LIBRARY OF CONGRESS. -^ %p_V-© W ; 5 pi n . Shelf. jB.S3 UNITED STATES OF AMERICA. Ssu mmm BY THE SAME AUTHOR. I. The Philosophy of English Literature. Lectures delivered before trie Lowell Institute, Boston. i2mo, cloth, $i . 75 II. The Principles of Psychology. i2mo, cloth, . . . 1.75 III. Comparative Psychology; or, the Growth and Grades of Intelligence. 121110, . . .1.75 IV. Science, Philosophy and Religion. i2mo, cloth, . 1 . 75 V. The Philosophy of Religion ; or, the Rational Grounds of Religious Belief. i2mo, cloth, 1.75 VI. The Principles of Ethics. i2mo, cloth, . . . . 1.75 VII. The Principles of Natural Theology. i2ino, cloth, 1.75 G. P. PUTNAM'S SONS, New York. THE SCIENCE OF MIND BY JOHN BASCOM AUTHOR OF ".ESTHETICS," "PHILOSOPHY OF ENGLISH LITERATURE, " PHILOSOPHY OF RELIGION ", " ETHICS " n* n ' ] NEW YORK G. P. PUTNAM'S SONS 27 and 29 West 23?j Street 1881 0- COPYRIGHT, 1881, BY G. P. PUTNAM'S SONS PREFACE It has been a reproach to philosophy, generally and persistently put forward, that it makes no progress, that it lacks established elements, that it is a field of extrava- gant and contradictory theories. "We do not accept these assertions in the unqualified way in which they are thrown out. So made, they are the result of ignorance and un- grounded contempt on the part of those who so easily utter them. So far, however, as these statements are true, they are a common reproach and misfortune, to be removed only by more patient, more protracted, more guarded inquiry. To scorn and reject philosophy as pre- sented under its own, its metaphysical form, subject to its own conditions, is simply to deepen the difficulty, and postpone indefinitely an answer to the most fundamental and central inquiries. If more than the usual number of mistakes have been made in this department, it is because more than the usual obstacles lie in the path of progress. These are not to be removed by discouragement, or by opening ways in other directions. All success to the stu- dents of physical science : but each of its fields may have its triumphs, and the secrets of mind remain as unap- proachable as hitherto. With philosophy and not with- out it, under its own laws and not under the laws of a iv PREFACE. lower realm, must be found those clues of success, those principles of investigation, which can alone place this highest form of knowledge in its true position. The fol- lowing treatise is at least a patient effort to make a con- tribution to this, amid all failures, chief department of thought. If asked why I hoped this volume might re- ward study, I should answer, Not because the system pre- sented is new, but because the statement it here receives is at once succinct and elaborate, is strengthened by new points, by a consistent maintenance of all that belongs to it, and by the rejection of that which, essentially alien to its principles, only embarrasses it. I trust the Intui- tive Philosophy will be found hereby to have gained somewhat of that proof which springs from completeness and proportion of parts. I have acknowledged my obligations to ' others in cases in which they have been direct. I here especially ex- press my indebtedness, in the general tone of the philos- ophy presented, to the eminent explorer and instructor in this field, Dr. Hickok. Holding my work amenable to thorough criticism, I shall yet expect but little profit from the facile applica- tion of previous opinions to detached points ; or from any discussion of the principles involved less penetrative and systematic than that here presented. I believe this treatise to have the integrity of a system, and to call, therefore, for a joint and complete judgment. To such handling I hopefully commend it. In the present edition secondary points are more fully presented than before, and the work is better fitted for PREFACE. V the purposes of instruction in higher education. I have been diffident in claiming for the philosophy here offered the independence, coherence, and strength which I believe belong to it. As, however, I am zealous for the system, and have found critics easily overlooking points not forced upon their notice, I now invite at- tention to the clear definition given to the doctrine of the intuitions ; to the care with which they are enu- merated, with which their relations to each other are pointed out, and their constructive office in thought is assigned them ; to the development of higher powers in connection with lower ones ; and to the support which liberty receives from the spontaneity of the intellect. Herein are secured a certainty of conviction, a strength of defense, and a clearness of explanation, not other- wise attainable. The system lies in direct continuation of the Intuitive Philosophy, but is put upon advanced ground, in a form more self-sufficient and defensible than hitherto. CONTENTS INTRODUCTION. PAGE § 1. Value of Philosophy 1 § 2. Determines the rank of Man 2 § 3. Correlative of physical knowledge 4 § 4. Its connection with moral and religious truth 6 § -5. Disparagement of metaphysics 8 § 6. Postulates of Philosophy 13 BOOK L—The Intellect. CHAPTER I. THE FIELD OF MENTAL SCIENCE AND ITS DIVISIONS. § 1. Field of Philosophy, Consciousness 19 § 2. Difficulties of Philosophy 22 § 3. Aids to inquiry — Language — History — Physical organs 27 § 4. Division of faculties 30 § 5. Volition and choice, not separable 31 § 6. The relation of consciousness to our faculties 33 § 7. Mental phenomena below consciousness 34 § 8. Intelligence and consciousness 46 • § 9. Dependence of the Mind on the Brain 56 • § 10. Abnormal states, Hypnotism 67 § 11. Physiological and mental facts 69 § 12. Is the mind always conscious ? 72 CHAPTER II. THE INTELLECT — ITS DIVISIONS — PERCEPTION. § 1. Divisions 79 § 2. The Senses — Perception .. ... 80 § 3. What do we see ? 82. vni CONTENTS. PAGE § 4. Perception and common conviction 91 § 5. The doctrine of direct perception 94 § 6. Direct perception and consciousness 99 § 7. The action of each sense 101 § 8. Importance of perception — History 107 § 9. Primary and secondary qualities — Criteria of 116 § 10. Consciousness as a source of knowledge 123 CHAPTER III. THE UNDERSTANDING. § 1. What the understanding includes — Memory 125 § 2. Theories of memory — Hamilton's theory 128 § 3. Association as connected with memory — Habit — Growth 133 § 4. Qualities of memory — Kinds of memory. 136 § 5. Imagination . . 142 § 6. Theories of imagination — Bain — Hamilton 143 § 7. Influence of imagination on passions — On judgment 147 § 8. How cultivated — The word conception 151 § 9. The judgment — Importance of 152 ^ 10. Nature of judgments 156 § 11. Erroneous views of judgment — Sir William Hamilton 160 § 12. Judgments, classes of 166 § 13. A second use of the word conceive 170 § 14. Judgment and association 172 CHAPTER IV. THE REASON. § 1. Disagreement as to existence of this faculty 176 § 2. Existence— Bain „ 179 § 3. Number 183 § 4. Resemblance 1 85 § 5. Space 188 § 6. Consciousness 201 § 7. Time 204 § 8. Cause and effect 209 § 9. Spontaneity. 221 § 10. Truth.. . .' 224 § 11. Right 224 § 12. Beauty 240 CONTENTS. IX PAGE § 13. Infinite . . 243 § 14. Criteria of regulative ideas. . . 254 § 15. Objective character 256 § 16. Not to be referred to inheritance 260 § 17. Grouped in their relations 263 § 18. Have all been given ? 265 CHAPTER V. THE DYNAMICS OF THE INTELLECT. § 1. Growth of mind 268 § 2. Acquisition of knowledge — Resemblance 278 § 3. Sensation — Judgment — Reasoning , 283 § 4. The control of the mind over its phenomena 290 § 5. Difference of endowment in man and in the animal 296 BOOK IL— The Feelings. § 1. Distinction between feelings and thoughts 306 § 2. Division of the feelings — Use of words , 309 CHAPTER I. PHYSICAL FEELINGS. § 1. Division of the physical feelings 312 § 2. General sensations 313 § 3. Special sensations 315 § 4. Appetites — Purposes subserved by sensations 317 CHAPTER II. THE INTELLECTUAL FEELINGS. § 1. Divisions — Desires — Not primitive — Division of 322 § 2. Feelings dependent on desires — On success 329 § 3. Feelings which accompany failure ' 332 X CONTENTS. CHAPTER III. THE SPIRITUAL FEELINGS. PAGE § 1. What these are — Their divisions — Truth 337 § 2. ^Esthetical feelings 342 § 3. Moral sentiments — Effect on other emotions. 346 § 4. Affections — Religious sentiments 350 § 5. Classification of the feelings 352 'CHAPTER IV. DYNAMICS OP THE FEELINGS. § 1. Offices of the several classes of feeling. . « 355 § 2. Order of development in the three classes successive 357 § 3. Communities take up the law of growth 359 § 4. Animal life in its feelings — Utilitarianism 361 § 5. Laws that control the feelings 363 BOOK III.— The Will § 1. Relations of the will — Subdivisions 369 CHAPTER I. THE NERVOUS SYSTEM. § 1. General relations of the nervous system 371 § 2. Life as plastic power 372 § 3. Parts of a nervous system 373 § 4. Various forms of a nervous system 375 CHAPTER II. THE NERVOUS SYSTEM OF MAN. § 1. Principal parts 378 § 2. Offices of each part 881 § 3. Forms of nervous action 385 § 4. Cerebrum in man 387 CONTENTS. XI CHAPTER III. EXECUTIVE VOLITION. PAGE § 1. Executive volitions 391 § 2. Interaction of voluntary and involuntary stimuli 392 § 3. Relation of voluntary and involuntary action 394 § 4. Relation of cerebral action and thought 398 CHAPTER IV. PRIMARY VOLITION, OR CHOICE. § 1. Divisions of volitions 400 § 2. Choice 401 § 3. Motives must be unlike in kind — A moral element 403 § 4. The will governed by the strongest motive 404 § 5. Effect of liberty and responsibility of this view 407 § 6. Whence the idea of weaker and stronger motives 408 § 7. Objected to liberty that it admits no control 409 § 8. Proof of liberty — Not found in consciousness 411 § 9. Proof of liberty — Spontaneity 417 § 10. Proofs summarized .. 421 CHAPTER V. DYNAMICS OP THE WILL AND OF THE MIND- § 1. The will strengthened by use 425 § 2. Order of activities — Reflex relation 426 § 3. Control exercised by the will 428 § 4. The feelings central — Their relations to pleasure 430 CHAPTER VI. THE RELATIONS OP THE SYSTEMS HERE OFFERED TO PREVALENT FORMS OF PHILOSOPHY. § 1. Value of general convictions 434 § 2. Materialism — Its a priori character 440 § 3. Hamilton — Direct perception 445 § 4. Idealism — Excellences — Defects 447 § 5. The philosophy now presented — Its relations 450 § 6. Divisions of Philosophy 454 INTRODUCTION. § 1. Though a knowledge of the value of a subject is not necessary to its successful pursuit, yet it inrparts to our inquiries peculiar zest. We shall never fully understand the advantages connected with any science, till we have mastered it; and it is thus natural that each should praise his own favorite pursuit, experiencing daily the enjoyment and power it confers. Nor is this commendation usually, in itself considered, excessive ; it is chiefly at fault, as it disparages other investigations, in themselves possessed of rival claims. As the fashion of thought in our time is to underrate philosophy, a brief space bestowed to urging its importance will not be misemployed. We shall not enlarge on the pre-eminent mental dis- cipline it gives, the acuteness of analysis, the steadiness of attention, the breadth of principles. All study im- parts more or less of this training, and some are willing to believe that metaphysics bestows an unprofitable sub- tility of intellect, a gymnastic dexterity of thought, more fit for show than service, more likely to mislead than guide their possessors. There are certain peculiar and pre-eminent considerations on which we would chiefly rest our estimate of philosophy. The facts which it furnishes are most intimate to our own actions, to the mastery and ordering of our own thoughts, and to the influence we are to exert over others. 2 INTRODUCTION. It is indeed possible, that there should be healthy and suc- cessful intellectual action, a wise play of the emotions and of the moral nature, without understanding them. So may there be physical health without hygiene; yet who will deny an influence of the knowledge of the laws of life in the government of life ? To pick up a few facts so per- sonal, so of our very selves, as those which pertain to mind, cannot but be of the highest moment in ordering our ac- tion. Indeed, every man who has any claims to general knowledge is a philosopher, however much he may deny it, and however false and limited his conclusions may be. It is not a question whether there shall be philosophy among men ; this there must be, if men are to think and act at all ; but whether this philosophy shall be a true or false one. Yet we do not wish to dwell on the value even of the facts which mental science gives, their direct practical worth in affording rules for intellectual training, and for influence over others ; but rather to point out cer- tain broader relations of philosophy, which make its acqui- sition yet more imperative. § 2. In the first place, no true notion of the dignity of man will be attained without it. If we consider man ex- clusively in his external relations, in his physical organiza- tion, and the ministration of nature to him, though we shall certainly assign him, if we reflect wisely, a pre-eminent po- sition, we shall by no means measure his true worth. The forces and lives of the world grade up to him, and grade down from him ; and while he is the highest and latest of living things, he is nevertheless of them, ruling by a supe- riority, not by a complete separation, of nature. The body of man is very perfect ; but those other organisms are also in kind marvellous. The brain of man is very large ; but those other brains are large also, and apparently thoughtful. Having travelled in classification all the way up from in- DIGNITY OF MAN 3 fusoria, the last strides of progress, great as they are, do not impress us as throwing man out of the general range and fortunes of the life of which he makes a part. As a matter of fact, those whose attention has been most external in its objects, who have studied nature, and man in nature, have held comparatively disparaging views of the rank of the human race. They have put it in the direct line of development with the life below it; they have thought it to share its intellectual and moral endow- ments with the higher animals ; and they have subjected it, in common with all life, to the fatalistic lock of physical forces. Approaching man from below, we interpret him from the types of power we find in nature, we limit his liberty or rob him of it, we expound his moral nature by the law of utility, so obtrusive in the acquisition of physical good ; while we seem to find the germ and outline of his intellectual constitution in brute instincts, perceptions, asso- ciations. We are thus as those who contemplate in a statue more the pedestal on which it rests, the marble of which it is made, the measurements to which it conforms, than the living, spiritual power it expresses. There is no adequate defence against this tendency, no reasoning man out of this grasp of scientific classification, from the position of bimana among quadrumana, from his rank as co-ordinate in structure with the gibbering mon- key, the grinning chimpanzee, the brute-headed gorilla, except through philosophy — without reversing the process, beginning at the top and moving downward — without con- sidering that which is internal, and overshadowing with it, transient,, external conditions. Suppose, for instance, as the result of such direct, independent inquiry, it is found that liberty belongs to man, a power altogether unique, with no prediction or type in nature ; that the moral intui- tion, the necessary accompaniment of freedom, transforming 4 INTRODUCTION. it into self-poised, responsible life, is equally independent and primary ; do we not in these two pillars of personality discover supports which lift the spiritual life into an en- tirely new region, which cannot be broken by all the blind giants of simple, physical induction that may bow them- selves against them ? If also it shall appear that the intel- lectual action of man is throughout different in kind from that of the animal ; that we have no proof that the truly rational elements of thought ever enter the lower field of life, ever transform associations into comprehension, £hen shall we again see, that we have reached a new plane ; not the completion of that which is below, but the commence- ment of that which is above ; not to be explained from the earth upwards, but from the heavens downwards. To estimate man outwardly, physically, is to judge a temple from the exterior, is to decide upon it by the order of its architecture, the bevel of its stones, the greatness of its workmanship, without entering its shrine, seeing its worship, or studying its ritual. So to judge man is as if we should pronounce on the supernatural claims of Christ by an inquiry into his human features and Jewish charac- teristics, in perfect oversight of the subject matter of the question. Man is to rank according to his spiritual consti- tution, and that it is the office of philosophy, and philosophy alone, to inquire into. We must go within the mind, see its structure and appliances, before we can know the dig- nity of the race. If this is denied us, if these portals are locked against us, we can only remain mute till the key shall be brought us. § 3. The second great office of philosophy is to fur- nish a counterpoise, a complement and corrective to the methods of natural science. It is not because we overlook the legitimacy and practical value of these methods, nor be- cause we disparage induction, a chief builder in the tern- COMPLEMENTARY TRUTHS. 5 pie of knowledge, one that has commenced and is carrying briskly onward some of its most showy and serviceable por- tions, that we urge the rank of philosophy ; but for this end, that the two may be seen to be truly supplemental each to each, that the arrogance of science and its supercilious de- nials may be seen to so cut down the scope of human facul- ties and hopes as to make knowledge itself comparatively trivial and nugatory. It is the nature of the mind that knows that gives signiiicancy to knowing, and if this term, the one most intimate to ourselves, in which alone we are deeply concerned, is to be excluded from knowledge ; if the disembodied spirit, the mind itself, is to be left wandering in the limbo of things forever uncertain and unknowable, then, indeed, is it a most minute and unsatisfactory gain, that our unexplored and unfathomed powers lay hold for a little of the things about them ; a small matter that the stream, rushing on, we know not whither, yields a troubled reflection of the shrubs on its banks. We claim that the knowledge that centres directly in mind, in its moral and intellectual powers, and in the social, civil, and religious actions that arise immediately from them, is a full half of all knowledge ; and that the methods of reasoning employed in these departments, while very different from the naked inductions of science, constitute the nobler moiety of intellectual life. We urge attention to philosophy, because the sphere of thought cannot be complete without it, cannot be rounded into a well-balanced and stable orb. If there has been one development more preposterous than all others in the growth of knowledge, that develop- ment is Positive Philosophy — a scheme that scouts meta- physics, and yet can do it on no other than metaphysical grounds ; that determines what may be known and what may not be known, and puts among the things to be dis- 6 INTRODUCTION. carded the knowing faculties ; that uses philosophy to ex- plode philosophy, and on the ground thus cleared builds up a cobble-house of facts, every one of whose connections must yet be as purely intellectual as those of mental science itself. This is as if the eye, failing to look backward as well as forward, inward as well as outward, should deny the existence of anything in that direction, and affirm the objects before itself to be ultimate, the only resolution of facts into ideas. To save us from such pitiful philoso- phizing, we need philosophy. * We are, then, in a peculiar want of this branch of knowledge, since it is a hemisphere of itself, holding in equipoise the world of truth ; since in it are found new regulative ideas, new laws, new lines of order, and also the tests of the validity of knowledge, and the rational grounds on which the limits of inquiry are established. Patches of truth may be given here and there by science, but land- marks, a synthetic rendering of the whole, can only be se- cured by the aid of philosophy. • § 4. A last reason we shall urge for these lines of inves- tigation is, that intelligent moral action and religious faith must rest upon them. Fortunately, considering the prem- ises from which they start, men are so illogical, that they find no difficulty in believing much which in consistency they ought not to believe, no difficulty in doing that for which their own philosophy can render them no adequate reasons. Eut in spite of the fact that there is often an interior coherence in action, in the unconscious workings of our constitution, which does not appear in our reason- ings, a false, deficient philosophy will, from time to time, come to the surface in unbelief, irreligion, immorality ; the ground will soften under long trodden paths of faith ; and many blind pilgrims, plunged into an unexpected quag- mire, will fail to reach the farther shore. All the ideas on FUNDAMENTAL TRUTHS. 7 which morality and religion rest are established and de- fined in the realm of metaphysics, and to deny ns this branch of knowledge, or to treat it slightly, is to put us, in the conflict with unbelief, at such disadvantage that we can never maintain our ground. "We may, indeed, shut our eyes, and stand fast ; we may stop our ears, and run from the questionings and claims of scepticism; but we cannot maintain our position in quiet and serene conviction, with- out searching for those foundations of truth found in the discarded field of philosophy. The nature of right and its obligations, of liberty and its responsibilities, of the infinite in its application to God, as well as the positive and negative knowledge we have of his existence and attributes, are to be established by an inquiry into the phenomena of mind, the truths present to it, their source and authority. To hope, therefore, for morality and religion, and yet to sink out of sight those abutments on which they are to rest, is infatuation. Those do not so hope who wittingly do this work of denial and overthrow — quite the contrary. Very many of them well understand that their mines run beneath the sacred edi- fices of religion, the spiritual labors and history of the race, and that, if they can be fully and successfully fired, these will sink, a mass of rains, into a black, sulphureous chasm. We see, therefore, that the intellectual battle be- tween belief and unbelief, religion and irreligion, must be fought, in large part, in the fields of philosophy. The truths of revelation must be vindicated or overthrown by their relation to man's constitution, his powers of knowl- edge and obedience, and the rational stretch of his hopes. Simple, then, are the reasons for philosophy, if philoso* phy be possible. We must abandon ourselves later than alii things else, consent to darkness everywhere, if we can only strike a cheerful light at this fireside of our home. Urn 8 INTRODUCTION. fortunate, indeed, would it be to lose the reins of power wherewith we guide the forces of nature, but far more un- fortunate to miss the right handling of ourselves, and that serene strength which wins the rewards of life. § 5. But is philosophy possible ? Is there not rather foundation for those many taunts and denials, asserting the endless, hopeless round of conflicting theories, the entire want of progress, the inevitable uncertainty attaching to every conclusion, and all conclusions, in metaphysics ? If philosophy be not possible, if there is ground for the scorn and incredulity with which labor in this department is often regarded, so much the worse for us all. Nothing can take the place of philosophy. If we are doomed to igno- rance here, our ignorance is hopeless and pitiable. We fail to understand the satisfaction with which some snuff out this light, when they have nothing wherewith to replace it — nothing better to propose than the desertion of this whole region, and a surrender of it to confusion and chaos. The injunction, "Know thyself," the revered precept of all time hitherto, thus becomes impossible, and to modern thinkers, ridiculous. Outside of ourselves, we move with patient inquiry ; we may feed our senses, and through them the mind ; but we harvest home this knowledge, we know not for what ends. We gather facts, ignorant of their ul- terior, spiritual uses, as the ox grazes, letting digestion and nutrition care for themselves. We see no grounds for con- gratulation in such a result. If it must be accepted, it yet remains a painful and sad alternative, turning the key in a door which above all others we would fain open, hiding from us things which most reveal the invisible world. It is as if some one, moiling long and patiently and profita- bly in the bowels of the earth, knowing how to pick and blast and shovel, and sure of the productiveness of those processes, should, hearing of the miscarriages, accidents, SCIENCE RESTRICTED. 9 and embarrassments of the upper world, begin to deny this region to himself and to others, and to make it the dogma of his life, that there was but one form of sure, safe, and re- munerative labor, but one unmistakable and positive good, and that was mining. We console ourselves, in view of such conclusions, with their entire falsity, and the utter im- possibility of their general acceptance. Other departments, moreover, besides philosophy, are to suffer from this rejection of the philosophical spirit. The positive sciences themselves require for their successful cultivation something beyond an observation of facts — a classification of resemblances. There is ever kept hover- ing before the mind some idea of the causes, the concealed grounds and reasons, of phenomena; and it is these supersensual notions which guide inquiry, direct the eye, and teach it what to observe. Without these, the classi- fications of science would come to little more than the child's art in grouping its bits of crockery by size or color or the conceits of fancy. It has been, for illustration, some notion of the nature of light, either as a material emana- tion, or a movement in a generally diffused ether, that has directed inquiry, instituted experiments, and interpreted facts. Yet there is nothing in philosophy itself more subtile, more impossible of conception, more evasive and evanescent than either of these supersensual conceptions, which have presided over this department, and resulted in most brilliant discoveries. Deny a search into intangible and inconceivable causes, causes that in their inception are purely theoretical, and we lose at once the clew of Our labyrinth, and henceforth wander at chance, with no fore- cast of thought, through its endless passages. Another illustration is furnished by the correlation of forces. Some notion of a hidden equivalence between very diverse phenomena haunts the mind, of a concealed agreement 10 INTRODUCTION. where no apparent agreement exists. This it is which sets the inquirer at work, quickens his thoughts, and leads him to new observations and experiments. An idea of a super- sensible thing termed force, is present to the mind. For this force in its very diverse forms, as mechanical action, heat, electric action, chemical action, it strives to find a measure, and so to establish an equation between these dif- ferent expressions by virtue of this their common term. Fruitful as this inquiry has been in science, it turns on an interpretation of things quite inscrutable to the senses. But how vain is it to demand positive, direct knowl- edge through the senses of this notion itself, so serviceable and indispensable? If we are to banish, as the ghosts of past superstitions, all the disembodied ideas the mind fur- nishes to positive science, we shall shortly be left without guidance, deserted of these good angels of thought, in whose absence eyes and ears are of no avail. We are in science, no less than in philosophy, constantly reaching and handling supersensual notions, purely mental phenomena ; we are ever making them most fruitful sources of further acquisitions, though certainly with no more full, definite, and positive knowledge of their very nature than that we possess of mental phenomena from consciousness. Indeed, the moment we penetrate a very little below the surface, Positive Philosophy is of the same nature with that which it discards, is dealing with causes, forces, and reasons which are wholly the offspring of the mind, and the limits of whose legitimate use must be determined on purely intel- lectual grounds. Nor is philosophy itself without its fixed, settled facts, as generally admitted and as incontrovertible as those of any science whatever. The laws of recollection, attention, judgment, imagination, of the emotions, of responsibility, constitute a large department of accepted conclusions. ACCEPTED TRUTHS. 11 The principles and precepts therein involved are running hourly through our processes of reasoning, our persuasion, our judicial action, our social opinions. Indeed, no single science, unless, perhaps, we except mathematics, is furnish- ing so many, so constant, so undoubted guides, both to those who maintain, and to those who deny, its theoretical value, as philosophy, with its adjuncts of logic, aesthetics and ethics. Totally untrue, then, is the representation that metaphysics is a hopeless medley of contradictory and un- verified theories. An appearance of truth is given to this assertion by directing attention from established facts to those skirting and partially explored fields of ontological inquiry, of the sources of our mental furniture, and of the authority of our faculties. We might thus discredit the established facts of electricity on the ground of conflicting opinions concerning the nature of the activities or physical states which constitute it. Now, it is evident, from the nature of the case, that more of these ultimate questions, more of these points at which direct, sensible knowledge ends, must belong to philosophy than to any other branch. The postulates and definitions of knowledge are conditioned on the faculties of mind, and to state these in their safe, ultimate form ; to settle where knowing, in all its phases, begins, and to give the reasons and grounds of these statements, is a late and difficult task, and one which should not, by its laborious and partial re- sults, prejudice a department which is highest in rank, as it is most recondite and ultimate in its conclusions. What act more indolent and unscientific than to jump to the conclusion, that these deepest questions are unsearchable and fruitless — than to turn our back on a region that does not at once yield its secrets ? ISTor are we without progress in these most obscure directions of philosophical inquiry. In some cases, the true conditions of the problem are bet- 1 2 INTROD UCT10N. ter seen — what is to be hoped for and what not ; in others, the grounds of attack and defence are shifted. Many arguments and presentations have been exploded, and, though others have taken their place, there has been pro- gress, progress toward an ultimate decision. The battle surges and rolls onward, and is not endless. The doctrine of human liberty is an example of the first sort. A more consistent statement of what it involves can to-day be made than ever before. It can be better distinguished from every form of necessity, and set apart with proper limits,. and more defensible boundaries than hitherto. To be sure we cannot explain freedom in the ordinary meaning of the word, but we can see why such explanations are not, and ought not, to be applicable. As an illustration of the second form of progress, we instance the discussions as to the sources of knowledge; whether among these are in- tuitive ideas. The doctrine, that experience is the ground of all knowledge, is a very different one in the hands of Spencer and Bain from what it was as expounded by Locke. The latter champions pronounce the earlier proofs and de-^ fences insufficient. Confessedly, then, this school has been driven in part from its line of argument. Herein is move- ment, looking to an ultimate solution of the problem. Though inner lines succeed one another, the city cannot be besieged forever. The grounds of conflict and the balance of strength are suffering daily changes, and though the con- clusion may be yet far off, we see that it is slowly prepared for by what occurs about us. This discussion is not simply the dogged reiteration of affirmation and denial ; the strik- ing of shadowy forms with immaterial weapons, the wounds of to-day closing against the battle of to-morrow. Quite the reverse; old points are yielded, new points are made ; light in turn is thrown upon these, and we move forward toward a conclusion — move slowly it may be, but as cer- POSTULATES. 13 tainly as when the discussion pertains to the nature of heat or light. Reid dogmatically asserted as a tenet of common sense what philosophy ever since has been defending, limit- ing, settling on rational grounds. Much work, indeed, remains to be done. The grounds of reasoning are to be more definitely fixed in this higher department; the logic of philosophy to be unfolded, re- straining erratic, fanciful movement, bending effort to fruitful results, and urging discussion to a speedy issue. If the inductive sciences owe so much to a new orga- num, a new form of logic, and that, too, to one lacking the strict proof of previous, deductive branches of inquiry, is it not rational to expect that further modifications of method, a new estimate of the nature and qualities of the proof ap- plicable to the unique and remote questions of metaphysics will be equally productive, will yield fresh fruits to wiser investigation. § 6. Before proceeding to the facts of philosophy, I wish to lay down a few of its postulates most frequently violated. First, all phenomena of mind, as facts of some order, demand sufficient causes for their existence. These phenomena are by no means* of equal significance in what they indicate, are not all normal, but they are all facts, and may not any of them be overlooked by any sound philoso- phy. "We may not select a portion, and reject a portion as the result of some vague and vaporing process which we have chosen to decry. We have disposed of no facts by calling them metaphysical or theological or any other name expressive of disapprobation. The entire facts of mind must be stated, accepted, and harmoniously covered by the theories of mind. The same test applies here as in physics, the ability of the explanation offered to expound all the phenomena that come under its consideration. A second postulate is, that there are different kinds of 14 INTRODUCTION. knowing, each independent of the others, each incapable of affording any light within the field of the others. The various forms of knowing show the various powers of the mind. The independence and diversity of the matter given reveal the independence and ultimate character of the faculty through which it is reached. If one knowing faculty could overlook another, the second would by that very fact be lost or merged in the first ; since for the two there would be but one line of perception. We have two eyes, but only one power of sense or sight, and this sense can do nothing, absolutely nothing, to cover the phe- nomena of taste or of smell. The additional and inde- pendent action of each intuitive faculty is involved in the very fact of its being a faculty, a distinct power of doing a distinct work. The reverse statement is evidently equally true, and gives us a third postulate, that we have as many faculties as we have distinct forms of impressions or primitive knowl- edge. The presence of a feeling, a perception, a conclusion, an idea in consciousness, must be explained ; and if it can not by analysis be resolved into simpler forms, or by deduc- tion be derived from a more primitive action, it must be accepted as itself primary, and the power to attain it be recognized. The question of elements is not different here from the kindred question in the physical world. Each form of matter ranks as an element, till chemical analysis has resolved it. The classified fruits of knowing imply as many powers of knowing, till the classification can be cor- rected by a reduction of the number of genera. A fourth postulate is, that we have intuitive as well as reflective powers. Reflection by its deductive processes exhausts primitive material of its instructional value, and by its inductive processes combines it in forms most suita- ble for general knowledge. But these methods of action POSTULATES. 15 imply material already present. We cannot derive all things from something more ultimate ; nor combine things, till we have the things to combine, and the idea of the order we pnt upon them. Reflection cannot furnish its first premises. The mind must have starting points, and these must be arrived at directly, intuitively. It is irra- tional not to recognize the beginning, or to strive to get back of it with an explanation. What these points of com- mencement are it is the office of philosophy to decide, and to arrest explanation and all effort toward it, when these have been reached. As powers are ultimate in their own field, the test of the correctness of their action in each case is its clearness, firmness, universality. This is our fifth postulate. That which is done uniformly by the mind, expresses the mind's normal power, and this normal power is to the mind suffi- cient proof. Though it may confirm its action at times in secondary ways, these ways will all of them involve the soundness of its permanent convictions. We may strengthen our reasonings by watching the concurrent progress of events, but we shall never allow those events to contradict a plain, logical process. We escape this result by the fresh inference that the premises in the mind and those embraced in the physical facts are not identical. Clear- ness in the mind's action is the first element of certainty. This clearness may be so comprehensive and complete as to enable a single mind to oppose its convictions to the gen- eral convictions of men. Pre-eminent mental power does not allow itself to be matched with majorities. Even in- sanity cannot escape this law, and the conclusions of the insane may even be more imperturbable than those of the sane. This fact does not destroy the value of the law. Health remains a power, though we cannot deny the exist- ence of disease. This first clearness confirms itself by re- 16 INTRODUCTION. iteration. What the mind firmly holds, under shifting cir- cumstances, is well held. And this again is further tested by universality. Universality is in some sense the proper test of a normal power of mind, but it is one that must be applied with great discrimination. Scarcely any truth in science or philosophy could endure its heedless use. There is no average mind which is a law to mind in all its mani- festations. It is only the decision of minds of like power, scope, and advantage that confirm each other. The widest experience is called for in the wise application of this test ; while the mind bears with it, all through the formation of this experience, an unwavering confidence in its own clear convictions. This test, moreover, will vary with the power under discussion. The senses more readily contradict or correct the senses, than the thoughts the thoughts, or the intuitions the intuitions in their higher range. A last postulate is, that what is conceded — avowedly, tacitly, or impliedly — at one point, must be freely conceded at all points. Processes which themselves assume the good- ness of our faculties must not conclude with a denial or im- peachment of their integrity. A doubt must have a prem- ise, and if this premise involves confidence in the very reasoning by which the foundations of reasoning are dis- turbed, that doubt is self -destructive. An idea, whose valid possession is denied, must not be allowed to enter furtively into those very processes of thought by which it is profess- edly eliminated. If it cannot be removed in the mind's ordinary action, it must not be removed in an exhaustive scientific statement of that action. If these postulates are truly adhered to, we shall cut ourselves off from a great deal of impossible and absurd effort to assimilate one form of knowing to another ; from a feeling of dissatisfaction because our analytic inquiries are brought at length to a halt ; from denying any knowledge POSTULATES. 17 because it does not assume a familiar and specified form of knowing; and from deceptively using ideas in the very attack which we make upon them, knitting together our reasonings with axioms stolen from an adverse system. By these postulates we secure several advantages. We safely start our knowledge ; we start it theoretically as we do practically in our intuitions. We prevent the trespass of one form of knowledge upon another, or the concession of an undue pre-eminence to any one process of mind. We fortify the foundations of knowledge against irrational at- tack. The intuitive powers which at any stage are yield- ed by analysis are freely accepted by us, and if there is a disposition to distrust any one of them, we are carried back immediately to the process by which its claims are to be tested. If we believe any knowledge not to be simple and primary, we have only to show it to be compound and derived. So long, however, as we accept it as a distinct, unanalyzed conviction, we must assign it a mental power, and concede its entire validity. These postulates keep our philosophy at work on the familiar mental facts offered us for explanation, and check it in any erratic speculation which is proceeding in oversight or subversion of the phe- nomena under consideration, the hourly thoughts of men, the knowledge current in the human mind. BOOK I THE INTELLECT. CHAPTER I. The Field of Mental Science and its Divisions. § 1. There is no branch of knowledge more distinctly defined in its limits than mental science. It lies in a unique realm, cut off from every other — that of consciousness. All the phenomena of this field in their separation, classification, mutual interaction and dependencies are the subjects of this science, and its only subjects. There is thus little oppor- tunity to confound the inquiries belonging to philosophy with those of any other department. Logic and Ethics most nearly approach it ; but the one considers abstractly the products and processes of thought, and not the think- ing powers ; and the other, the moral constitution of the mind, and is so far a branch of philosophy, adding thereto, however, an evolution of practical precepts from moral principles. Anatomy and physiology, on the side of the natural sciences, are most closely allied to philosophy, yet, after all, deal only with the physical conditions and instruments of mental action, and, without the key and interpretation of 20 THE FIELD OF MENTAL SCIENCE. mental science itself, can cast no light whatever upon it. The facts of philosophy lie in consciousness ; here they are to be sought, and every fact therein contained is to be made the subject of consideration. Consciousness is commensurate with all mental states and acts. It accompanies feeling as much as thinking, and volition -as much as either. The only possible way in which a mental state or act can be testified to, is by con- sciousness; some mind at some time has known or felt it. An event that happens nowhere in space is not a physical event ; an act or state that is not found in the field of consciousness is not a mental act or state. There are either facts that are neither physical nor mental, that exist neither in space nor consciousness, but in some unintel- ligible form in some third region, or all facts fall under these two divisions ; and it remains the criterion of one class that they occur in space, and of the other that they occur in consciousness. A third state is inadmissible as unknown and unnecessary. Consciousness is neither a knowing nor a feeling nor a willing, is neither this nor that mental act, but a condition common to them all, a field in which they appear, in which they arise and make proof of their existence. A consciousness of knowing is neces- sary to knowing, a consciousness of feeling is necessary to feeling, and of willing to volition ; and as these three cover all states and acts of mind, consciousness is involved in the very conception of a mental act or state, It is an insepara- ble something which defines the nature of the phenomena to which it pertains. Consciousness gives — we use familiar language, a more careful expression would be, in consciousness is found — the mere fact of a mental state, that it is, and what it is, whether one of thought, feeling, or volition ; or a complex one involving two or more of these. It renders phenomena TRUTHFULNESS OF CONSCIOUSNESS. 21 as they exist, not analytically but synthetically, as the eye colors, or the ear sounds. To reach the primary colors which constitute the tint, the separate notes which form the harmony, calls for attention and discrimination. The mere facts of mind as facts are rendered in consciousness, and to be found there and Only there by all who meet the conditions of search. Discussion is had as to the truthfulness of consciousness. There is no ground for such discussion, since the discussion itself involves the thing doubted. Nothing can be better known than a fact of consciousness, since nothing can be known save through such a fact. Consciousness pervades all knowing, all thinking, distrust equally with trust, denial with affirmation. No man ever does doubt, nor can he philosophically doubt, the existence of a present fact of mind. To do so would rob language of all meaning. The only way in which such a dispute becomes possible is by wrongly regarding consciousness as a faculty, giving direct testimony to certain things, instead of something involved in the very fact of feeling and knowing, making them what they are, and, therefore, never present except through veri- table, and, for the instant at least, undeniable, feeling and knowing. Whether the thing known has an indepen- dent existence, or the thing thought is correct, are quite other questions. A doubt of the truth of the testimony of one or more of our faculties to the various things de- clared by them is a scepticism by one step less central and I less absurd than the distrust of consciousness. In this \ there is no show of rationality. There are involved in the one act an affirmation and denial of the same thing. There can be no rational objection taken to an inquiry into the facts of consciousness, and no uncertainty can attach to them as facts which does not attach in a yet higher degree to all other facts. The phenomena of 22 THE FIELD OF MENTAL SCIENCE. psychology are of the most primary character and certain order, even as compared with those of science. There is no knowledge so direct as that which the mind has of its own states ; all other knowledge is indirect being condi- tioned on this knowledge. The phenomena of mind are given truthfully and synthetically. Its facts are evanes- cent in each mind, but they are, for the moment at least, certain. In their several forms and their diverse lines of succession, they are exceedingly complex and changeable. Their obscurely synthetical character is their most strik- ing characteristic. § 2. The facts of mind are confined, then, to the field of consciousness, and there they are to be sought. In this search there are peculiar difficulties. It is with most an unusual effort of mind to direct attention to interior phe- nomena. External objects have been the chief subjects of consideration, and to turn the sight of the mind on itself is an unfamiliar and delicate process. It is like an effort to reveal to the eye itself its own chambers, by casting in light and by adroit reflection. Neither are the several phases of mind observed as transpiring, but as remembered. In the very act of think- ing, the mind is so occupied with the subject matter of thought as not to make the process itself the object of at- tention. JSTow memory is at best but a dim and obscure vision, and especially so of internal states, which less draw the mind's eye than the objects and facts which are the occasion of them. If natural science were to proceed by the memory of things, seen at periods more or less remote, its progress would be comparatively uncertain. Nor can the phenomena of mind be restored perfectly at pleasure, and thus the recollection of them freshened. This is more pos- sible in thinking than in feeling and volition ; yet even in thought, for its natural and full flow in a given direction, DIFFICULTIES OF INQUIRY. *2S the mind must be disengaged from conflicting states and considerations, and be left to the unobserved and sponta- neous action of the connections and impulses peculiar to the mental movement. This inability to hold directly the state considered be- fore the mind, as the plant or mineral is watched and re- tained by the eye, is connected with another difficulty ; no one can join us in our investigation with the directness and certainty which pertain to other inquiries. The object be- fore the mind of each observer is hidden from the other, may not be of exactly the same character as that with which he is occupied, nor looked at in the same direction. This confusion of objects and observations is most perplexing. It is as if the eye were turned a little askance, and the movement and the blow, therefore, directed at an image before it, and not at the very thing itself. Much skill and time are thus consumed between different observers in drawing attention to exactly the same facts. They often, through the deceptive effect of agreeing words, seem to have reached this result, when they have not attained it, and thus fall into inextricable confusion and contradiction. The feebleness of direction and construction is akin to that experienced, when, by the sense of touch alone, groping in the darkness, we strive to understand the parts, proportions and relations of even a familiar room. It is also incident to this search of consciousness, that no one observes more than the phenomena of his own mind, and those, too, of a comparatively recent period. It is diffi- cult, therefore, to determine how far a peculiar balance of faculties, as individual habits and associations, may have modified the mind's action, giving prominence to certain forms and connections of thought, and obscuring others. This fact also embarrasses us in deciding how far the mind's later convictions are due to protracted association, 24. THE FIELD OF MENTAL SCIENCE. and how far to native, inherent tendencies or powers. Is the normal, adnlt mind in forms of action the fruit of growth, or are these forms native and indispensable to it ? The consciousness of the child or of the savage, so far as these questions may there seem to find an experimental answer, is beyond our direct exploration. Another embarrassment in philosophy, though not pe- culiar to it, is the blended way in which its facts are pre- sented. Not only do thought, feeling, volition unite in one state, diverse and conflicting feelings struggle for the mas- tery, and, in the simplest judgments, are interwoven per- ception, memory, reasoning, imagination, intuition, and the subtile effects of association, rendering analysis a difficult, yet an indispensable condition of success. Analysis of this obscure character, with phenomena in themselves evanes- cent and fluctuating, requires the utmost skill and tension of mind. Another obstacle to success to be mentioned are the peculiar deficiencies of language in this department. Lan- guage, always an essential, is here a chief, instrument of investigation. It is the precision of the word employed, that separates and holds fast the faculty, element, or rela- tion designated. In natural science, objects exist apart, though not named, and hence do not lose their identity, are not so merged in the ebb and flow of shifting phenomena as to escape all observation. The very sense of existence is largely due in mental facts to a clear, specific,, generally recognized name ; since we handle the states of philosophy exclusively through their names, and without these, readily lose all traces of them. Moreover these names are applied somewhat in the dark. It is by description and suggestion that we are taught what the internal states are to which given words are set apart. The word is the same, but the internal fact which explains it is, in every single case, dif- DIFFICULTIES OF INQUIRY. 25 ferent, that is, lies in a different mind, and mnst be hit on as the thing meant, by the sagacity of that mind. We are as one who puts together a complicated machine by a printed description and directions before him. Careful observation is required to determine the parts referred to, and failing of this, all is confusion. Yet in this illustration the parts are fixed, separate, with a permanent, independent existence ; while the parts of a complex, mental state ad- mit of various divisions, or may disappear altogether, like some rivet in the dust of the shop. , To attach words, there- fore, to their objects ; to unite the two so that there shall be no escape for either, is a delicate and uncertain process. The ambiguity of words embarrasses all forms of statement and reasoning, but is never elsewhere the source of so much idle discussion and fruitless inquiry as in philosophy. A further obstacle presented by language is, that it comes to mental phenomena saturated with the imagery o\ the external world. Words are born amid sensible facts, and thence transferred to the mind. They come, therefore, to this new service with the images and associations ac- quired in the old. They subserve a popular, familiar use only the more aptly for this reason. It is when they are made the subject of careful analysis, when they are treated as the exact expression of the thing named, that their physi- cal root and relations reveal themselves disastrously. The mind reaching this interior analogical thread of interpre- tation is pleased by it, and overlooks the fact, that investi- gation is thus sure to be led astray ; to be turned entirely from true mental phenomena, and to be sent wandering among their shadows and reflections in the external world. Thus, from the very beginning, every discussion concerning liberty has been embarrassed, and in most instances has miscarried, through the application to motives and desires, in a figurative sense, of words begotten amid the necessary 26 TEE FIELD OF MENTAL SCIENCE. connections of physical things. These half-reclaimed ser- vants, when closely questioned, have betrayed their low relations, and in so doing have lost to liberty its high, ether- eal form. Like its household, it has been thought to be mud-born. The facts before us for discussion must one and all re- appear in language as a condition of their consideration ; and this difficulty subdivides itself into four obscure and shift- ing forms. (1) There may be no word in the language to express a mental fact or relation. Thus we are now bor- rowing noumena to oppose to phenomena. (2) The exact attachment of a word may be uncertain, and different for different persons. Thus consciousness is still very vague in its application. (3) Words of fundamental importance may have several meanings, and play in a most perplexing man- ner between them. Thus right may designate that which is intellectually or socially or morally right. (4) The images of the physical world so pervert our psychological words as oftentimes to constitute a false connection between the things covered by them. Thus the doctrine of association has been built up on the idea of a certain coherence, affinity, or what not, between thoughts and feelings aside from the powers of mind expressed in them. The last difficulty is allied to this, and arises from the uniqueness of the department. It refuses to receive illus- trations from the analogies of matter ; or rather it refuses to accept as the exact types and counterparts of its own facts and dependencies those of a realm at the farthest possible remove from it, at the very nadir of the sphere of being. Yet the mind, familiar with certain processes, certain forms of explanation, certain couplings of thought, is uneasy and dissatisfied with all others, is only content when it has put the new matter under the old law, the new wine into the old bottles. Unable to hold it in these stiff, inflexible case- AIDS IN INQUIRY. 27 ments, such a notion as that of the infinite perplexes and vexes the mind, simply because it is not the finite, and thus stands opposed to other forms of knowledge, and is ex- cluded from them. Equally is it annoyed with liberty for not yielding to some analogy of necessity, some interpreta- tion drawn from the physical world ; for not taking upon itself, in a more subtile way, the iron-bound connections of matter ; and with right, because it will insist on being final, and refuses to be merged in any other form of good what- ever. To accept a new department — so new and so novel as this of mind when contrasted with that of matter — as new, to lay aside prepossessions, and to commence again with simple intuitive convictions, the axioms of this field, involve a sore conflict, and the more a conflict in propor- tion as the inquirer has gained great victories of know- ledge in the material world, and dwelt long amid its methods of action. This is probably the gravest of all the obstacles to philosophy, and the more so because it is gen- erally entirely overlooked or forgotten. § 3. While the phenomena of mind are to be obtained directly and only directly from the mind itself, there are very important indirect auxiliaries of inquiry. Language is one of the prominent of these aids. Language, as the product of the mind, as the external, visible trace of the mind's movements, reveals of course the forms of its action, and, in the designations of mental phenomena, a part at least of the facts of the interior world. On disputed questions of analysis, also, the spontaneous, general convictions of men are betrayed by the words they use ; and a distinct designation is so far proof of the general recognition of a; distinct idea. That certain words are always and every- where floating in popular speech indicates that the thoughts^ of men find rest in them, something valid, sufficient to> steady and sustain the mind as for the moment it lights 28 THE FIELD OF MENTAL SCIENCE. upon them. These traces of the mind, indicating its own spontaneous convictions, that which is actually woven into the web of its thinking and feeling, must be included in every sound theory of philosophy, and furnish the suggest- ions for its construction. Of the same nature exactly, though not as easily accessi- ble or explicit, are the facts of daily life, and of history. The shadow of the mind is cast upon them, and we may reason thence to the powers and capacities they indicate. A theory which utterly confounds, as do some metaphysical theories, all the convictions of daily life, and makes the facts of history and those of philosophy rest on utterly di- verse conceptions, so much so that no region seems so re- mote or even preposterous as this metaphysical dream-land to the very beings who are said to inhabit it, by that fact reflects on itself extreme improbability. History must be felt to be, and found to be, the very shadow, the intimate reflection, of that inner life which is revealed to us by men- tal science. Another aid to philosophical investigation is found in an inquiry into the instruments of the mind, the physical organs which it uses ; and into the incipient and rudiment- ary development of intellectual action shown by animals. We are thus able to give more correct weight to the purely physical element, and to separate more intelligently the lower and instinctive forms of animal life from true, men- tal powers. While not underestimating the secondary and inferential aid thus to be rendered to philosophy, we think that extravagant and absurd expectations of the results of investigations primarily physical have been entertained. One might look at a brain with utmost inquiry, and, with- out the interpretation of the facts of consciousness obtained by introspection, his observations, as initiating a science of mind, would not be of the least avail. To suppose that the PHYSICAL AIDS. 29 divisions of mental faculties can be fonnd either on the outside or inside of a skull is preposterous. Passing the experimental proof so fully given by Hamilton, that no such connection as that sometimes claimed between certain powers and certain localities of the brain can be shown to exist, we insist, that even if the fact of such a connec- tion were proved, we should still as much as ever need an independent philosophy derived from consciousness. These regions are not labelled in the human subject. They contain in themselves no suggestion of the purpose sub- served by the portion of the brain beneath them. The ob- server must have an antecedent idea of certain mental powers, and be ready to attribute one or other of these to the prominence under his fingers. Afterward he may con- firm the act by observation. This first condition, however, failing him, the bump under discussion might as well be a protuberance on a potato as a projection on a human skull. The one, in and of itself, as a mere prominence on a round body, makes no more declaration of ideality, benevolence, language, than the other. Suppose we have made from consciousness a wrong division of powers, what is there to hinder us from transferring these errors to our map of the cranium? Nothing; they will rather inevitably thus re- appear. The chart that is to guide us must be made out before we can begin to outline and number and name the divisions of our plaster bust, and equally also before we can attribute a faculty to a locality in the living subject. The absurd classification of phrenologists; such faculties as com- bativeness, philoprogenitiveness, secretiveness, are sufficient proof, if further proof were wanting, of this inability to find the invisible action of the mind in the visible form of its instrument. All the aid given to philosophy by an external fact is inferential, not direct ; and that invisible faculty or force which is thus to be reached, on which our 30 TEE FIELD OF MENTAL SCIENCE. conclusion is to land, must be given, in the only possible knowledge of its nature, by consciousness. The analysis of mental phenomena shows that firmness is the complex re- sult of various, and of different, mental states, and no loca- ting of a supposed faculty so called in one or another por- tion of the head can alter, or throw light on, these facts. The singleness of the name and locality imparts no new singleness to the mind's action, marks no division of its faculties. The invisible cannot be seen through the visible. Each must be determined independently, and the connec- tions between the two established by experience. It would be as rational to suppose that the letters contained in the word will should of themselves convey to every mind the notion of that power, as to suppose that a prominent eye should reveal the existence of a faculty called language. Regarding consciousness, then, as the only field of the science, whether reached inferentially, or directly under the interpretation of the light it itself furnishes, we pass to the general divisions of mental phenomena. § 4. The leading divisions of the faculties of the mind, so generally accepted since the time of Kant as scarcely to demand further explanation or defence, are those of knowing, feeling, willing ; the intellect, the sensibility, and the will. The desires are by Kant and Hamilton in- cluded with the will. They belong rather with the feel- ings. Desire is employed to designate a state of feeling toward a certain object or objects. We find things differ- ently related to our happiness ; we cease, therefore, to be in- different to them ; one object or line of action gets a hold upon us ; wa are drawn toward it, and this emotional state we call a desire. Language sustains this decision. Desires are constantly spoken of as feelings, never as thoughts or volitions ; the words in the first case are used interchange- ably, not so in the second. We apply the same adjectives DIVISION OF POWERS. 31 to them as to the feelings. We say of a desire as of an emotion, that it is strong or weak, firm or changeable, in- tense or feeble ; and sometimes, as in the case of avarice, speak of it as becoming a passion. Our desires, also, may be directly opposed to onr volitions. We greatly covet a certain possession, but our pride constrains us not to ask for it. We wish the pleasure of a given action, but through fear determine not to perform it. A state of desire, like every state of feeling, is antecedent to volition, and may or may not find play in subsequent choices. As a desire it may arise and pass away emotionally, like envy or jealousy or sympathy or love, and find no expression in action, awaken not the will at all. It may either meet with ac- ceptance by the mind, or suffer rejection by it. Desire, then, should be included in the field of the emotions, where it arises, and spends its power. It does not, in the fact that it gives occasion to the will for activity in providing for its gratification, differ from other feelings. These also, as long as they last, are springs of volition. § 5. An attempt has been made to divide the depart- ment of will, into choice and volition. A color of plausi- bility is given to this division by distinguishing between initiatory volition and executive volition. The first is termed choice, the second volition. When two diverse lines of action are contemplated, and the mind is as yet unde- cided between them, the desires have free play, the sense of moral obligation is present, and the conflict awaits a definite settlement by a choice, a fixed determination, in favor of one or the other. We sometimes, at this point, use the word choice out of the meaning which should attach to it as pertaining to volition. Thus we say, " My choice would be this line of effort," though we actually accept and pur- sue another. Choice is thus made to express a state of de- sire, not one of will. The word choice, however, in its use 32 THE F1FLD OF MENTAL 8GIENGE. in the third department of mental phenomena, expresses an explicit termination of all vacillation, a close deliberation by an act of will in favor of this and in rejection of that. The case thus being closed by a specific and peculiar act, there remains a longer or shorter series of efforts to be made in reaching the object proposed, in accomplishing the career marked out. There are no definite limits in analysis to these intermediate acts. Our division may extend to each muscular movement, or it may stop short with each specific undertaking. I propose to build a house; the number of distinct physical and intellectual efforts involved in the project are indefinitely great ; and while they are all under the control of the will, we have no occasion to place a distinct volition back of each one of them. The will has the power, by a few explicit volitions, to direct the current of the vital powers in a single channel of expenditure. A walk once entered on, the movement becomes in a large measure unconscious, and the mind is left at liberty to pur- sue any line of action it prefers. The voluntary and the involuntary play of physical members differ not so much in the manner in which they are sustained, as in the way in which they are initiated, and in the fact that the one is momentarily open to modification, and arrest. The distinction then made between a choice and a voli- tion seems to be found in their position in reference to an end, rather than in their intrinsic character. The one is initiatory of a line of action ; the other sustains and com- pletes it. Ths one is primary, the other subsidiary. The one is determinative and governing', the other executive. The first gives character to an action, the second sustains and develops that character. The one is immediately free, the other mediately so, through its dependence on the first. The division thus sinks into a classification of volitions, and removes neither choice nor volition from the phenomena of CONSCIOUSNESS. 33 the will. Choice, as an act of will, does not include the deliberation and the play of feeling from which it proceeds ; but only that final act by which they are brought to a close, and the powers of the mind made to unite in a line of effort. Executive volitions are the secondary impulses of will, by which its primary impulses are completed ; they are the prolongations of that power which is born of choice. The ball is driven in a given line, but receives accessions of force, and changes of direction, as the exigencies require. § 6. The relation of consciousness to the three forms of mental action is the same. Sir William Hamilton seems to have regarded its connection with knowing as somewhat peculiar. While he speaks of it as the condition of all mental phenomena, he says, "Those of the first class, the phenomena of knowledge, are indeed nothing but con- sciousness in various relations." The complete and expan- sive statement is rather that consciousness is the condition, and equally the condition, of all mental acts and states. It is merely through a deficiency, or peculiar use of language, that it seems to be more intimately connected with know- ing than with feeling. To know a thing, and to be con- scious of it, are used as interchangeable expressions ; and, hence, we have come to regard consciousness as a kind of knowing, or as an act of knowing, and not merely and purely the condition of such an act, that which permits knowing to be knowing. It is not strange, that a constant condition of an act should, in language, take the place of the act itself. Through this interplay of the words con- scious and know, we are able to say, " We know that we feel," " We know that we will ; " though we can with only doubtful propriety say, "We feel that we know," "We feel that we will ; " and cannot at all say, " We will that we know," "We will that we feel," This use arises through a peculiar connection in the language employed of conscious- 34 THE FIELD OF MENTAL SCIENCE. ness with knowing, and thus a transfer of the word know to both feeling and volition. Consciousness is no more an act of knowing than it is one of feeling, and is a condition in exactly the same sense and way for the one as for the other. We know in consciousness, we feel in conscious- ness, we will in consciousness ; and consciousness is neither an act of knowing, nor of feeling, nor of willing, but a con- dition of them all. Consciousness is not a something, a faculty or a light, which reveals acts, independently of knowing, feeling, willing, to the mind ; but that which makes an act of knowing to be one of knowing, of feeling to be one of feeling, and volition to stand forth as volition. Mind, by virtue of its own nature as mind, does and suffers what it does and suffers consciously, under this simple and inexplicable condition of being aware of its own acts ; a con- dition which is no more allied to one act than to another, to one state than to another, but is common to each in its in- divisible nature. A feeling is not a feeling and a knowing that we feel ; a volition a willing and a knowing that we will, but simply and singly an emotion and a choice, under the essential condition of such acts, to wit, consciousness. § 7. Two allied inquiries arise in this division of men- tal phenomena. Are there any mental phenomena below or outside of consciousness ? Are the states of mind, the acts of consciousness, consecutive or intermittent ? Sir Wil- liam Hamilton, and many other metaphysicians, recognize unconscious modifications of mind, we think without suffi- cient proof. The conclusion is too purely conjectural to command our consent. Mental and physical phenomena are cut broadly and deeply apart by the fact, that the one class appears exclusively in consciousness, and the other as exclusively out of consciousness. The last are actual or possible objects of some organ of perception, are somewhere located iu space, and thus open to the outside action of SUBCONSCIOUS FACTS. 35 mind through its senses ; the first are within the mind, evincing their existence exclusively by their effects in con- sciousness. Not to exhibit anywhere, to any actual or sup- posable organ of sense, any phenomena, is, in the physical world, not to exist. Existence is affirmed only on the ground of some effects, however subtile, in sensible objects, and directly or indirectly in organs of perception. We never hear of physical facts above or below space, beyond all possible tests of perception ; since such phenomena would be utterly unable to manifest this existence, to give any proof to it. The very notion of physical being arises from that of physical effects, under suitable circumstances open to observation. Thus also should mental phenomena be regarded. There is likewise only one known field for these — consciousness. All, aside from physical facts, that occurs outside of this, is necessarily unknowable. An al- leged fact, which is to be found anywhere as a fact, has but two avenues through which it can make itself known, the senses and consciousness. These are the sole means by which we take cognizance of any class of phenomena. To assert, therefore, the existence of other modifications or changes than those which respond to these two methods of knowing, is to affirm some third field, wherein events hap- pen whose nature is utterly unknown to us, and of whose being we can at most have only an hypothetical and infer- ential knowledge. Some imperative reason should be given for the accept- ance of phenomena utterly unknown, and from the nature of the case unknowable. By what principle are those un- known modifications, if thought to exist, classified as mental facts ? Something it would seem should be revealed more distinctly as to their character, before they are assigned to this class rather than to that of physical facts. If these unknown modifications are acts or states of mind, are in 36 THE FIELD OF MENTAL SCIENCE. any way phenomena of mind, we ought to have provision made for them in our classification of mental facts. The division would then run thus : the phenomena of knowing, of feeling, of willing, and a fourth class composed of certain unknowable states, acts, conditions, or whatever you please to call them, of which we have no knowledge either in per- ception or consciousness, and can say nothing by way of explanation. States, then, of mind may occur of which the mind itself knows nothing, and which furnish, neither in the field of thought or of forces, any direct proof of their existence. The argument for their being is thus of the most naked and inferential character. If it be said that these modifications are modifications of the mind itself, and not of the nature of actions, I think it must be granted, that they are thus conceived wholly under the analogy of material changes, and that if they are shown to be, and to belong anywhere, it is in the physical, and not the mental world — in the brain, the instrument of the mind, and not in the very mind itself. In this last, we know, and can know, of no organic changes. Its own acts and states constitute the sum of our knowledge concerning it. Nor are we hereby rid of these alleged modifications as phenomena ; nor of the consequent need of giving some clue to their mode of existence. We are thus brought to the fundamental difficulty of this view, that it tends to confound the broad distinction be- tween mental and physical facts — especially between men- tal facts and those which pertain to the brain and nervous system. 'No matter what relations exist in the brain itself, or what changes take place in it, an observation and knowl- edge of these are no part of mental science, and do not nec- essarily, do not alone, give a clue or explanation to any one of its facts. The organic functions and dependencies of the brain are matters of as distinct and purely physical SUBCONSCIOUS FACTS. 37 knowledge, as those of the liver, and no changes here can reveal to us the nature of a mental state, or of the powers peculiar to the mind. We can no more find the mind in the brain — because this is the organ of thought, than we can the life in the heart, because this is the chief organ of life ; or than the ancients could have searched it successfully for the affections, because they regarded it as the seat of the feelings. Listen for a moment to the words of one of these modern philosophers, who reject consciousness as the field of mental science. " Not only is the actual process of the association of our ideas independent of consciousness, but that assimilation or blending of similar ideas, or of the like in different ideas by which general ideas are formed, is no way under the control or cognizance of consciousness. When the like in two perceptions is appropriated, while that in which they differ it neglected, it would seem to be an assimilative ac- tion of the nerve-cell, or cells of the brain, which, particu- larly modified by the first impression, have an attraction or affinity for a like subsequent impression ; the cell so modi- fied and so ministering takes to itself that which is suita- ble, and which it can assimilate or make of the same hind with itself, while it rejects for appropriation by other cells, that which is unlike and will not blend." — Maudsletfs Physiology and Pathology of the Mind, p. 17. It is difficult to treat with respect explanations like these. Is the brain the only organ whose cells take to them- selves " that which is suitable," that which they can make of the same kind with themselves ? Why then do not the liver, the kidneys think, and unite like things in thought by resemblance ? No one thing is more separate from an- other than is cell-action from thought. To speak of the 38 THE FIELD OF MENTAL SCIENCE. two as the same is to use words for ideas. Who, by observ- ing the one, could come to a knowledge of the other ? One might watch at his leisure the operation of Morse's tele- graph, and, unless his previous knowledge furnished him the solution, make nothing evident but his own vacant mind. Yet the connection of this contrivance with lan- guage is far more mechanical and obvious than that of the brain with thought. The affirmation of subconscious phe- nomena is especially objectionable as playing into material- istic philosophy, as confounding the distinction between physical and mental changes, and referring real or imagi- nary modifications of the brain to the mind, as if the two were equivalent. But the views of Hamilton are not intentionally open to this objection ; let us briefly consider the reasons he gives for the acceptance of unconscious modifications of mind. The first of these is the extraordinary power the mind sometimes shows of recalling events, and even unintelligible sounds, as those of an unknown language, long after every trace of them seems to have passed from the memory. " Extensive systems of knowledge may, in our ordinary state, lie latent in the mind beyond the sphere of conscious- ness and the will ; but in certain extraordinary states of organism, may again come forward into the light, and even engross the mind to the exclusion of its every day posses- sions." In this argument we simply meet the old difficulty. How does the mind remember? How does it store up knowledge with no apparent store-house, accumulate men- tal vigor with no mental muscle wherein to lodge it, gain sharpness, precision, ease, with no underlying structure, in which these qualities may be thought of as inhering ? That memory shows unusual power under certain ab- normal conditions of mind does not essentially alter the SUBCONSCIOUS FACTS. 39 character of that power, nor introduce new conditions into the problem. Physical strength is not different in kind when exhibited in an astonishing degree by a maniac, from what it is in ordinary states of body. An ordinary act of recollection involves the whole question, involves neither more nor less than an extraordinary one. These queries— How does the mind remember ? How does it subjectively acquire and retain power? — we must submit are unanswera- ble ; questions which receive no light whatever from any supposed modifications of some supposed substance of the mind. If such modifications were granted, we should under- stand not in the least how they were equivalent to acts of memory, or productive of them— we should simply have two inexplicable things instead of one. The tendency to ask and answer such questions arises from the physical world, where we expect no change of powers without change of structure. The early solution given to this problem of memory, that certain films escape from objects, and are laid away in a secret store-house of the mind, is just as good philosophically as the latest ; and sprang from exactly the same false tendency to carry the analogies of matter into mind. The form of mental action is not revealed to us, and we have no clue to it except this false one of reasoning from things and processes totally unlike those of mind ; bringing the interpretation of physical phenomena to intel- lectual facts. We reject the explanation of mental power furnished by unconscious modifications of mind, because it is really no explanation, making the subject not the least clearer ; because these modifications themselves are wholly hypothetical ; and because they are inferred by analogy, from a field remote from the subject in hand, and alien to it. The second proof offered, is allied to the first. It is drawn in like manner from the analogies of the physical 40 THE FIELD OF MENTAL SCIENCE. world. The minimum object which the eye can perceive may be conceived as divided into halves ; neither *f these will be objects of perception, yet each of them must make a distinct, though unconscious impression on the organ of vision, in order that the conjoint effect may be perceptible. We have, then, the first conscious state in sensation secured by effects themselves unrecognized. Hence springs the inference, a conscious state of feeling or thought may be preceded by unconscious states as its conditions. We ob- ject to the analogy. The eye is a physical organ, lying be- tween the object and the perceptive power. There may be in it action too slight to reach the mind. In the case which this fact is brought to illustrate, there is no analogous mid- dle term between the mind and its own action. The ques- tion is, whether its own, its veritable, acts and states are always known to the mind? Now these actions are not occasioned in some intermediate substance by a foreign cause, and taken thence by consciousness, or overlooked by it, as the case may be. There is no such medium between the mind and its own acts. External, physical conditions, there doubtless are ; but these constitute no part of the mind itself. Keeping the inquiry itself clearly in view, Does the mind know all that the mind itself does, all that passes in it % it will be seen that the above analogy casts no light upon the subject. If the theory is, that external forces act on the substance of the mind, or, to put the same thing in appropriate specific terms, that nervous energy animates the brain, and that a certain amount of this influ- ence constitutes thought, while less amounts, though of the same nature, fall below consciousness, then indeed there is an analogy in the cases, and the argument so far holds ; but we have reached out and out materialism. The theory on this basis offers no more explanation of the problem, How does a pure act of judgment or of memory take place, than SUBCONSCIOUS FACTS. 41 would be found in the study of a piece of mechanism, a power-loom, or an electrometer. The brain is indeed more immediately the condition of the mind's action than any other part of the body ; but the brain, the body, every ma- chine and instrument it uses, are the conditions to one or more of its activities, and no one of them constitutes the very substance, the very nature of those activities. A third argument is found in acquired dexterities, as those of the equilibrist, or the musician. It is asked : How shall the separate acts involved in the rapid perform- ance of the musician, each of which was originally preceded by an act of volition, be explained, when established skill has banished from sight this directive power of the mind? One philosopher answers, " The movements of mind re- main, but take place too rapidly for distinct observation and memory." A second replies, " They remain, but remain as acts or changes below consciousness." Before we at- tempt to judge between these opinions, it may be well to inquire for the proof, that these impulses of mind remain at all. We believe that the supposed difficulty arises from overlooking the nature of the connection of the mind and of the body. Much of the nervous, executive play of the body never passes under the cognizance of the mind, does not penetrate the region of consciousness, is purely auto- matic. Some of this action, on the other hand, which is usually self-sufficient, is yet open to the arrest and modifi- cation of the mind. Of this character is the process of breathing. Few will claim that an act of mind is back of each inspiration and expiration, though we can at pleasure shorten or deepen, quicken or retard the movement. I may find myself breathing in a manner that is inadequate or injurious. I may for weeks laboriously strive to enlarge and deepen the play of the lungs. I may succeed, and the improved method become habitual with me. Will it be 42 THE FIELD OF MENTAL SCIENCE. claimed, that henceforward my inspirations are all volun- tary, each preceded by an act of mind ? I think not. The improved process is as automatic as the previous one, and no more requires subconscious mental acts for its explana- tion. There are still other physical movements more con- stantly voluntary, more rarely involuntary. We thus speak of them as voluntary acts, and seem to regard them as under the exclusive impulse of the will. There is no good reason for this. The fact that I walk whither I will, and modify my movement as I will, is not a sufficient reason for requir- ing a distinct, mental act, conscious or unconscious, back of each muscular movement made in passing over each rod of the road I am pursuing. The will, as it were, by one voli- tion, belts the automatic powers, and these run on till they are again arrested or redirected. If the play of the nervous energy to and from the nervous centres is sufficient to secure motion without consciousness of any mental action whatever, as in the case of the heart, is it not equally capa- ble of continuing a motion the will has established ? If we analyze each voluntary motion, so called, into the most single muscular movements of which it is composed, and place a mental act back of each, we have an absurdly com- plex result, and one not in the least testified to by con- sciousness, nor required by the known conditions of the problem. All the powers of life are not mental, and a great share of the labor of living is done by forces with a strength and movement more or less, as the case may be, independent of intellectual control. In acquired dexterities, volitions are required for a time to establish and confirm the automatic movement, but this, once settled, is able to sustain itself by a purely vital power, a play of nervous energies without the direct support of the will. The difficulty of the question seems to have arisen SUBCONSCIOUS FACTS. 43 from not marking the degree in which vital phenomena are independent of mental action. A fourth argument for unconscious modifications of mind, is found in the association of ideas. Links of associa- tion, it is said, are frequently omitted. The mind passes from number one to number ftve or eight in a train of con- nections without distinctly recalling the intervening steps. How does this happen ? Does the mind move through the entire series, though too • rapidly for memory ? or does the unbroken thread lie below consciousness, there traversed by the mind ? The last query is thought to indicate the true solution. But is there any sufficient reason for shutting us up to these alternatives? Is it so certain that the mind never makes a leap, that it cannot associate five with eight directly, omitting altogether six and seven? Is not this also an act and a method of association, as much so as that which originally united the ideas marked five, six, seven and eight, respectively? Six scholars stand before me in the recitation room. This fact of itself is a ground of as- sociation, but it also gives occasion for fresh associations of various kinds, and so may cause the memory on the pres- ence of one to recall any of the remaining five by some new nexus. Take the case of acquired meanings. A word may have stolen from application to application along an ob- scure path of subtile connections, till it has reached the twentieth meaning. How many of these successive uses any one mind shall recall in employing the word will de- pend in part on knowledge, and in part on the frequency with which the word recurs. The last meaning may be the only one suggested to the majority of minds in the majority of cases, though the previous ones and their connections may be known to them, in whole or in part. The word becomes at length a 44 TEE FIELD OF MENTAL SCIENCE. literal term in its twentieth meaning, attached in this sig- nification directly to its object ; though there lie between the first use and the present application nineteen images, each of which has been carried in the imagination, impart- ing to the word a figurative force for a greater or less length of time. What is to hinder the mind's going directly ? Nothing : association itself prepares the way for it. The explications offered by unconscious mental acts in- volve facts far more obscure than those explained. This movement under the surface of consciousness, is in itself a most perplexing riddle, a strange something we know not what. Nor, if it is granted, do we at all understand how it can or does change its nature, and suddenly issue in a movement within consciousness. The supplied links in this theory are of an unintelligible nature, and do their work in an unintelligible way. The whole result is more perplexing and obscure than if we accept the naked phe- nomena, and suppose the mind to pass from idea to idea, now by a more direct, now by a more circuitous route, able to do the first, because it has done the second. The facts presented in consciousness are more manageable by them- selves than when surrounded by suppositions, which involve phenomena unknown and unknowable. The dip of the thread of connections below consciousness is a loss of it for all practical and explanatory purposes in chaos and night. If it re-appears in the realm of knowledge, it comes like a ghost from Hades, in a mysterious method and an inex- plicable guise. The answer, then, to this proof is double. (1) New as- sociations are constantly obliterating and overriding old as- sociations. (2) And the intermediate steps of an association often repeated drop from the mind. The mind is able to unite directly the first and last terms, when these are the only significant ones. Thus the Koman characters for four, SUBCONSCIOUS FACTS. 45 six, seven, eight, nine, come to designate these numbers as readily as the corresponding Arabic numerals, though at first they involve addition and subtraction. The abbre- viation of mental processes by the omission of familiar steps, and by the more direct union of remote ones, is a cardinal fact in mind. The connection of this idea of a subconscious region with materialism plainly appears in Lewes' Physiology of Common Life. He affirms : " that all nervous centres in action, give rise to Sensation, and thus furnish elements to the general Consciousness" Thus we are made to be con- scious of all the muscular and involuntary movements that take place in the body. This strange affirmation is thrown into the very teeth of consciousness itself, momentarily affirming the reverse truth to us all, on the purely a priori grounds, first, that a similarity of ganglionic structure in these nerve-centres implies similarity of office ; and second, that constant, physical impressions must be made upon them, and hence, must enter consciousness. " Every such excitement of the sensitive organism must be a sensation. These sensations will necessarily be very various, as the organs excited, and the exciting causes, are various ; but they must all be sensations, they are all active states of the general property of sensibility. Ergo, they must all be elements of consciousness." Thus, this author so thoroughly identifies physical with mental states, that having established the first, he out-faces the mind itself, and declares that they must be consciously found in its record. This is only a bolder movement in the one general direction, since it pretty much annihilates the dis- tinction between conscious and subconscious phenomena, and brushes lightly aside any testimony the mind itself may offer, as to anything that is, or is not, passing within it. If there is any mockery of consciousness more extreme 46 THE FIELD OF MENTAL SCIENCE. than every other, it is this affirmation, that every peristal- tic motion of the intestines is a phenomenon of mind. So one mind at least classifies its activities. § 8. There has been developed in recent philosophy, especially in Germany, an increasing disposition to extend intelligence, as a state or energy, to all forms of activity that manifest intelligence, and thus to confuse the physical acts with the mental energies from which they proceed. The strength with which the conviction has taken posses- sion of philosophy, that certain physical states are the equivalents of certain mental ones, is very great. We must, therefore, meet it in its various forms of assertion, since the phenomena of mind cannot be profitably discussed till their independent primary character is established, and their limits laid down. A bold, firm line of division be- tween the physical and the intellectual realms is a cardinal necessity. J. J. Murphy, in a work on Habit and Intelli- gence, presents this phase of opinion under which intelli- gence blends with and is lost in the physical. Sensation and thought are regarded by him as in their own nature un- conscious ; consciousness is quite a secondary phenomenon. " So far from consciousness being necessary to intelligence, unconscious intelligence is the rule, and conscious intelli- gence the exception. Intelligence presides over, as an in- dwelling power, all vital action — formative, motor, mental — and is as significant a term in one portion as another of the vital process."* The author does not mean by this that a conscious Divine Intelligence orders the organic process, but an unconscious constructive intelligence. There is, by figure of speech, intelligence in the steam-engine, but it is the intelligence of the machinist. This view implies some- thing more, it identifies intelligence with physical organic processes ; or with some inherent unconscious force that is * Consult the earlier chapters of the second volume. INTELLIGENCE. 47 supposed to control them. We have, therefore, in it simply a degradation of the word intelligence. The conscious act of thought, and a physical fact in nerve-tissue, — or in plant- tissue — must forever remain incomparable phenomena, things most distinct, most diverse from each other. To call both intelligence, is only to lose a division in one form, which we must speedily restore in another. The things themselves we can not long confound. A physical fact, however subtile, occurring in no matter how impalpable a medium, can not be, nor in any way represent, a thought, a feeling, a volition in consciousness. Even if we were to grant that a definite physical fact is the antecedent of a mental one, it would not follow that the two are equivalent. We can not talk of shadows without light ; shadows are not latent in darkness, no more than statues are latent in mar- ble. Shadows are to be thought of and discussed only with the light. When this comes, there are shadows ; when consciousness is present, there are thoughts. If we call any lower states thoughts, feelings, intelligence, when the true thoughts approach, we must look about us for a new and more princely appellation. To speak of unconscious intelligence is to discuss shadows, images, and reflections, with no mention of light. To speak of intelligence which is not conscious intelligence, is either to use figurative or unmeaning language. If we say of the plays of Shake- speare, that they are intelligent, we mean either that they sprang from intelligence in him, or awaken intelli- gence in us ; or, under human experience, we mean noth- ing. Things may be intelligible, they are not intelligent. Neither thought nor language justifies the idea of intelli- gence in things. A feeling of which we are not aware, a thought of which we are not conscious, are simply a feeling we do not feel, a thought we do not think. We can only fall back, in 48 THE FIELD OF MENTAL SCIENCE. connection with such a use of language into utter vague- ness, or onto purely physical states. He also says : " Sensation and consciousness are both feelings. To use logical language, feeling is the genus of which sensation and consciousness are species. 7 '* Hamilton regarded consciousness as a general term for knowledge, and here it is ranked as a specific feeling with the feelings. Hamilton would say that we know that we feel, and Mur- phy that we feel that we know, while the fact is that we both know and feel as simple, sufficient states of mind. Mr. Murphy proceeds to refer, in a curious way of his own, distinct states of mind to distinct nervous acts ; but the theory is, in reference to every one of its significant asser- tions, absolutely, purely hypothetical, is semi-mechanical and completely physical ; if granted throughout it explains nothing in mind proper, but serves rather to obscure and destroy the fundamental connections of thought and thought, thought and feeling, thought, feeling, and voli- tion. These physical theories of mind are wonderful ex- amples of explanations, fictitious in their data, futile in their expositions, and destructive of the facts expounded. There hold against them, one and all, these objections : (1) They rest on no one known fact broad enough to sus- tain them. The correspondence of a definite state of brain with an exact and pure mental state has not in a single instance been made out ; much less has it been shown, that the first is the invariable antecedent of the second. (2) These solutions proceed on qualities and relations which belong to matter rather than to mind. (3) They thus subvert instead of expound the phenomena to which they are applied. (4) The explanations fail at the very moment at which they should take effect. As long as the terms are physical, they are coherent, but at the instant of * Habit and Intelligence, vol. ii. p. 13. UNCONSCIOUS CEREBRATION. 49 transition all light disappears. Nor can this failure to show how or why the physical fact is productive of the mental one be helped by the analogy of the senses. That a series of physical effects in the senses should produce effects in the mind every way unlike themselves, and so be the ground of a knowledge of the external world, is a fact of a very different order from the assertion, that certain physical states and relations are the invariable equivalents of corresponding mental ones. Two points are overlooked by the analogy, that perceptions and thoughts are sharply distinguished from each by this very dependence of the one on physical facts, and the independence of the other ; and that the whole nervous structure has plain reference to, the one form of intercommunication, and no known ref- erence, obvious or obscure, to the other form of relation. The notion of subconscious phenomena has been one so vague, so vacillating between physical and mental facts, that it could not long hold its ground. There were three fatal objections to it : (1) It had no assignable place or form for its alleged facts. A subconscious region was as unintelligi- ble as a post-conscious or ante-conscious or super-conscious region would have been. Indeed, it implied them all. (2) It had no means of arriving at its facts. The senses might reach physical things, consciousness might lay open mental phenomena, but what perceptive power should explore this subconscious territory? (3) These phenomena, as wholly beyond the mind's construction, could serve for it no possible explanatory purpose. They were merely verbal lumber. As the advocates of this new tendency began to fathom more deeply the conclusions contained in their premises, the inadequacy of the old statement became apparent, and has been mainly displaced by that of " unconscious cerebra^ tion." Cerebrations, actions of brain, are supposed to have the force of mental facts, and are of two orders, conscious 50 THE FIELD OF MENTAL SCIENCE. and unconscious. Thought is made so dependent on cere- bration that if cerebration proceeds, it is regarded as im- material whether it is accompanied by consciousness or not. The mind may be borne forward in its intellectual processes by acts of unconscious cerebration. We may travel by day or by night, waking or sleeping, to our intellectual destination. The wheels roll on without our observation. This theory of course involves the pre-eminence of the cerebral state, and the progress of thought is made incident to its progress. The sun casts a shadow on a dial ; it is hidden by clouds for a time, and then again shines forth ; the hands have ad- vanced on the disk, and the index line, as if it had stolen on its way unobserved, falls at the appropriate figure. This view should recognize distinctly, and state clearly, that the intellectual movement is incident to, and controlled by, the physical one. Certainly an intellectual process as an in- tellectual one cannot progress in unconsciousness, any more than a shadow as a shadow can travel in darkness. This is the fatal objection to unconscious cerebration ; it subverts or obscures the true line of dependence. We do not deny, that, as the organ limits the activity of the mind its special states enter as a factor of moment into each re- sult. But so also does the condition of our muscles settle the effective force of the will ; yet the physical energy does not predetermine the voluntary power. No more does a process of cerebration precede and causally determine the mental activity it expresses. Dr. Carpenter, in his Mental Physiology, presents fully, and in its best form, the doctrine of unconscious cerebra- tion. Many of the reasons which sustain it are those already sufficiently considered in connection with unconscious men- tal action. The additional points made by Dr. Carpenter we will consider. This doctrine is so generally accepted, and is so destructive of true mental powers, that we feel UNCONSCIOUS CEREBRATION 51 desirous to return to it as often as any additional light can be shed upon it. It is a common experience, if a difficult problem, or a theme to be discussed, is called before the mind and then passed by for a time, that the thoughts revert to it later with unexpected advantage ; that a certain mastery of the topic seems to have been achieved in the interval. This new power, often very considerable, is referred to un- conscious cerebration, a process of thought that has gone forward, as it were, in the substance of the brain. The moment the favorite and favorable words are dropped, the argument, it will be observed, loses probability. " Uncon- scious cerebration," guides the mind to the conclusion more smoothly than the equivalent expression a physical change in cerebral states. We have here the trick of a phrase. These gains of thought, we think, may be much more wisely ascribed to the frequent reversion of the mind to the subject, and its leisurely consideration of it in a variety of lights, though the times of such secondary occupation, extending over considerable periods and thrown into the shadow of other events, are not conspicuous in memory; indeed, like any transient under-current of thought, may have quite escaped it. Few of these interstitial states can we recall at the close of a week. If the topic is not a familiar one, does not lie in the line of our pursuits ; if it is not a habit with us to return more or less frequently to a discussion once present to the mind, we shall find the gains of delay very slight. If, however, we are accustomed to restore a theme, to recast the thoughts at odd moments, and to gather new material as the process proceeds, then the yield of the under-drift will be correspondingly large. This fact, which is the significant feature of the general fact of acquisition by delay, shows that the mind does keep its intellectual garden in growth by indirect and un- 52 THE F1FLD OF MENTAL SCIENCE. obtrusive, though real, attentions. Cease to sustain uncon- scious cerebration by voluntary effort, and it will cease to be a noticeable fact. Alien subjects will gain little by de- lay. Insight, following time and rest, is the fruit of a vig- orous habit of mind. This result is due to the normal ac- tivity of the thoughts. It is no matter of surprise that the slight, forgotten exertions of the passing days may, at the close of a year, yield a respectable aggregate of results. Unconscious cerebration as a fact is certainly not easier of comprehension than that which it is here brought forward to explain. If a physical activity, self -directed and self- sustaining, can be the exact equivalent in intellectual results of the wisest thought, then thought as thought is no longer coherent, and a chain of reasoning can be made up of alter- nate links of physical and mental, conscious and unconscious facts. When we do understand a subject, reach a conclu- sion, we understand it from beginning to end; and in that final act of comprehension we leave no room for any merely physical facts, facts which in their transpiring were not acts of knowledge, nor are now acts of knowledge. The intellectual act of comprehension is complete, and receives no known aid from a previous unconscious act of cerebra- tion. The conscious act is the act, and this is sufficient to itself. It is not in the least plain how an intellectual diffi- culty can be overcome in unconsciousness, ignorance or error be flanked in the night-time. We must first say that physical states determine mental states, and are themselves determined by previous physical states, before unconscious cerebration can afford us any aid ; that is to say, we accept this philosophy first, and then get what light we can from it ; we are are not led to it by its own light. But the phi- losophy takes away all coherence from our intellectual life, subverts all its relations. A second fact urged by Dr. Carpenter is the sudden en- UNCONSCIOUS CEREBRATION. 53 trance of a new idea or ideas into the mind ; the return of something which the memory had struggled for in vain ; the instant presence of a fortunate conception completing an invention. These facts are plainer to us as ultimate facts, as successful efforts of mind, allied to that by which we correctly articulate a sound we have long striven to no purpose to utter, than they are when burdened with facts of cerebration whose very being is conjectural, and whose mode of operation is unintelligible. A subject is not ex- plained by two difficulties. Still less proof is there in the inventive moods of gen- ius, the unusual power that falls to the mind on one occa- sion and deserts it on another. The periods between these hours of advantage are often very irregular, and may have little to do with the direction in which the thoughts have been tending. They are preceded by no indications of an unconscious cerebration, but more frequently follow upon lassitude, indolence, restf ulness, passing again into activity. Careful preparation sometimes fails of its object, and an inventive flow will at another time be present to the speaker or writer independently of previous effort. These shifting moods of mind are partially explicable by physical fatigue or vigor; but involve many conditions not traceable to uniform causes. The mind, if not capricious, takes up and lays down its strength in a way often too subtile for our analysis. Invention remains invention, a quick putting- forth of power, sometimes from restful energy, sometimes from irritable force. Observation lends no support to the doctrine that there is a slow unconscious accumulation of thought-products in the brain, which, like waters in an in- termittent fountain, are suddenly poured forth. The mind, like the will, has reserved force by which it can achieve great and sudden results, and these, too, in unex- pected directions. 54 THE FIELD OF MENTAL SCIENCE. There are some facts in abnormal states especially diffi- cult of explanation under the idea that conscious states are predetermined by acts of cerebration. There are those double experiences, double states of consciousness, which proceed each under its own impressions, and totally suspend each other with an abrupt transition. Can two series of physical states utterly diverse, each coherent within itself, and incoherent in reference to the other, go forward in the brain, arresting one another in an irregular yet decisive way ? Possibly, but few would have the boldness to affirm that such a fact is plain enough to be offered as the solution of any other fact, that capricious physical states are more explicable than capricious intellectual ones, so much more so as be able to account for the latter. It may be said that the two hemispheres of the brain take up a disconnected instead of a concurrent action, and so give grounds for a divided consciousness. If this theory is to have any weight, if the implied facts are not far too obscure and uncertain to explain anything, it must still in- volve, we think, some transfer of attention on the part of the mind, akin to that by which we see through one or the other Qye. The physical states of brain in its two halves are doubtless continuous, and, if the controlling source of impressions, must give continuous impressions. We ought, therefore, if states of brain determine men- tal states, to have two coetaneous mental experiences, instead of a consecutive experience made up of alternate parts. The latter fact seems to imply the unity of mind, and that superiority by which it shifts its organs, calling them into service or letting them drop from it. Thus, if the axis of the eyes are thrown out of relation, the mind soon learns to use one eye to the neglect of the other. This idea of cerebration proceeds on that of an ante- cedent physical causation of mental states, and a strict UNCONSCIOUS CEREBRATION. 55 equivalence of these effects with their causes. Each thought, feeling, volition, every combination of these must be the exact counterpart of a correspondingly definite state of the brain. Every thought, and every possible thought on every possible topic, stand in correspondence to some form of cerebration. We think of Paris by one state of brain, of London by a second, of Pekin by a third, and of each person and thing in these cities by still other forms of physical activity. This is a complexity of conception quite startling and wholly unnecessary. It follows from placing efficient causation solely in the physical world. Admit independent power in the mind and this reasoning at once loses its force. No definite state of the muscles decides whether the strength put forth shall be expended in lifting a pound of lead, or one of iron, or of stone; whether the person shall walk toward the north, or the south, or the south-east. A certain energy, for what purpose soever employed, taxes the muscles to a certain de- gree ; and this fact is the simple condition of its exertion. It is not a given molecular state of the muscles which set- tles directions and offices, but the living agent who employs the muscles. The antecedent cerebral state may neither determine the form of the force, nor, save under general conditions, its degree. The efficiency deciding on the special kind of * activity is found in the mind, not in the brain ; for the mind is something other and more than the brain and its functions, as the engineer is something more than the engine. Assigning a definite intellectual state to an equally definite state of brain, we shall be unable to ac- count for the slight effect on the mind of accidents attended with a loss of the substance of the brain, or for the restora- tion of memories which have once been lost by disease. It is certainly not easy to- refer this renewal to a subtile super- intendence that in the passage of months slowly builds up 56 THE FIELD OF MENTAL SCIENCE. the brain- tissue under the previous pattern, with exactly the previous action. It may be thought that the senses, as sight, show a like complexity in the molecular changes by which facts are indicated to the mind. In this analogical example several things are overlooked. There has been a very protracted and complicated development of each special organ of sense for this very purpose, the registration of external causes. Just here lies the widest difference between perception and pure thought. Even in the senses each impression, though complex in itself, gives a clear field for the next impression. It exists only so long as an external cause sustains it, and wholly disappears in making way for another. If we press the organ beyond these limits, as the eye by the rapid motion of the spokes of a wheel, its power of discrimination is lost. If we search this analogy, thoroughly, therefore, as we must to make it an argument, it tends to the opposite conclusion, that there are no definite lines of causation other than those manifestly laid down in the senses. The senses, also, work by discontinuance not by continuity. To make the analogy applicable, the eye should not merely see the new thing, but retain all previous things. § 9. We need to understand more definitely the relation of the mind to its immediate instrument, the cerebrum, as a means of determining the proper character of mental phe- nomena, and their degree of independence. The mind and the brain are reciprocally dependent on each other, the states and activities of the one affecting the states and activities of the other. We may represent this relation by that of two wheels which press each other at their circum- ferences, and so, in their revolutions, mutually impel each other. The moving force may be transferred from one to the other, but expends itself in either case in the revolution RELATION OF THE BRAIN AND THE MIND. 57 of both. Indeed, a concurrent impulse may be applied to each wheel ; and in any given revolutions the eye may not be able to decide in which the balance of force or the entire force is .to be found. The relation, however, be- tween the mind and the cerebrum is not doubtless of that exactly reciprocal, equivalent character implied by the com- parison ; the two do not stand to each other on the same plane of causation, nor is it in their normal action a matter of indifference by which of the two the power is applied. Their real relation to each other is complex, can only be disclosed by a consideration of the facts, and then but partially. Activity of brain always accompanies activity of mind. (1) This is shown by the destruction of brain-tissue inci- dent to energetic thought. A sensible increase of the waste due to nervous tissue attends on such action. ¥e instantly conclude, and probably correctly, that the corre- spondence indicated is complete, and that all mental effort has an exactly equivalent expression in brain-action, and so in decomposition. (2) The fatigue and nervous exhaus- tion that accompany thought contain the same conclusion. We attribute these to the expenditure of nervous ener- gy, with its destruction of tissue. (3) The renovation inci- dent to rest sustains the argument. Sleep is the best restorative of mental functions, and sleep seems to bring peculiar repose to the brain through repose of the volun- tary powers, and so to make way for nutrition and refresh- ment. The reverse of this proposition is not so evident, that 'all activity of the cerebrum is accompanied by activity of mind. (1) We should not know what mental state, for in- stance, to refer to the reconstructive processes that pro- ceed in sleep or in restful hours ; (2) nor what modifications of thought to ascribe to incipient disease of the brain that 58 TEE FIELD OF MENTAL SCIENCE. has not proceeded to the point of overthrowing mental equilibrium. (3) The loss of brain also that has frequently attended on accidents, when a comparison is made between previous and subsequent mental states, does not show any exact equivalence between cerebral and mental activity, nor that the first, is in reference to the second, a measureable force, calling, in every variation, for a measurable corre- spondence. The brain may be materially reduced even in bulk with immaterial or vacillating results. (4) Its en- tire automatic action, as well as organic action, proceeds also, with little or no trace on mental states. States of brain at all times affect and at times control states of mind. (1) The general dependence of mental power on the size, form, quality of the cerebrum shows this functional connection. None can doubt that mental power, in its manifestation, is proportioned to the vigor of nervous action. The condition of the physical instrument is, in a large measure, that of the mental agent. This is a fact of daily observation. (2) Yivisection puts this truth beyond question. Though mental powers do not disappear in definite order and degree with specific portions of the cere- brum, they are disturbed by its injury, and lost by its re- moval. (3) Diseases of the brain complete the proof. In- sanity, partial hallucinations in a great variety of forms, are the constant accompaniments of cerebral disturbance. Not only does such disease diminish mental power, it strangely modifies that which remains. The thoughts seem to be the sport of abnormal, physical conditions. If the mind struggles occasionally for self-possession, it is soon overwhelmed again, and floated on by the current. We must, therefore, grant that, at least in some instances, physi- cal states seems to be the efficient, determining causes of mental ones ; x though even in these cases the 'final result combines the two series of forces, physical and mental. THE BRAIN AND THE MIND. 59 The proof does not carry this conviction, that cerebral con- ditions establish and define mental ones, but only that they are often an immediate, irresistible provocation to them ; as bad digestion to bad dreams. Physical states may over- power the mind, and the form of the hallucination still be due to the mind. The fever may induce a very diverse delirium in different persons. The two sets of causes are concurrent in the result. The reverse proposition we confidently offer as present- ing more important and more obvious truth. Pure mental states affect and usually control cerebral ones. (1) This is the manifest and unavoidable conclusion from the fact of thought. Thought involves the evolution of one mental state from a previous mental state, the attaching of a second conception or judgment to a former one. If it is not this, it ceases to be thought, and becomes illusion. Even imag- ination evokes its images one from another, unites them by a mental connection. If one thought, as in the proof of a proposition, follows another, not by an inherent connection, a mental dependence, but by the relation of successive phys- ical states in the brain, then we are utterly at fault in the entire thought-process, and it is something quite other than we have supposed it to be. The images of the imagination even can not be shadows that chase each other on the screen by an outside, alien law, much less can the successive judg- ments of coherent thought be united independently of the thinking agent. Reasons cohere, thoughts coalesce, con- clusions are evolved from premises, and these facts imply that a previous mental state is, in connection with the un- derlying powers of the mind, the efficient source of a suc- ceeding one ; that the nexus is a mental efficiency, and not a physical one. Reasoning, as the result of a series of cer- ebral states united by unknown physical forces, is utterly incomprehensible, is subversive of our most direct primitive 60 THE FIELD OF MENTAL SCIENCE. and constant convictions. Here is a proof that can not be escaped without quite displacing the foundations of truth. Argument itself is destroyed by such a conclusion, and so this conclusion is lost in the general wreck of all conclu- sions. We quite deceive ourselves in argument, if convic- tion is only a series of states induced by causes entirely blind, wholly alien to the process. (2) The relation of intellectual feelings to the convic- tions that call them forth presents a kindred proof. Words are spoken in our hearing which stand connected with our own sentiments, actions are performed which affect our interests; immediately there spring up decided feelings, we accept or we disapprove the opinions or the conduct. These states of feeling are plainly due to the mental ac- tion by which the bearings of the words or the transactions are disclosed to us. The intrinsic order and dependence are quite subverted, if it turns out that thoughts and feel- ings alike have been thrown in upon us in a secondary way by the motion of physical forces interlocked among them- selves, covering all real efficiency, and expounding the se- quence of the shadowy states of conciousness by their own independent and firm connections. Our higher emotional life is unintelligble on this supposition, and its apparent dependencies quite illusory. (3) The same is true with a like startling and fatal con- tradiction of our daily convictions in the case of our voli- tions. We refer these to the motives, to the intellectual and emotional states, that precede them, while the actions that follow them we attribute to the volitions themselves. This interpretation, under the hypothesis that cerebral states wholly determine mental ones, is completely erro- neous, and thoughts, feelings, volitions, actions are impotent in reference to each other; they are rolled off, as the pano- rama proceeds, in the order in which they are there intro- THE BRAIN AND THE MIND. 61 duced, by synchronous, physical agencies. Herein is the subversion of all mental processes. "What matters it, what we think, if there is no logical coherence in thought ? how we feel, if there are no just or unjust grounds of feeling ? how we act, if there is for action, within itself, no cohe- rent law % By this view, truly understood, intellectual and moral life are alike suspended, and are left the unsubstan- tial shades of their former selves. Under this hypothesis words must act as physical forces, inducing a cerebral:" state incident to which is a certain thought or feeling. It is not by virtue of the conception which is called forth that they are efficient agents, but the conception itself is one among the secondary states that attend on the primary sequence of forces in the nervous system. Thoughts, feelings, voli- tions, are as the shadows of the cars ; the cars are coupled, and the links of the shadows are the images of these coup- lings. (4) It would be difficult, also, under this reference of mental states to physical causes to explain the extended and exact agreements between men in their mental action. Why are the principles and processes of mathematics the same for all, except through their relation to the mind ? And why are other truths so diversely viewed save through divergent intellectual conditions ? These remark- able agreements and disagreements are referred, and must be referred, to the relations of the mind to truth. Truths that are simply deposited by similar physical processes, should show no such complete agreement, nor no such wide diversity. Constitutional resemblances, the physical condi- tions of habit and inheritance should control them, as they do the texture of the skin, the quality and color of the hair. (5) "We urge as a farther consideration— if, indeed, any farther consideration is called for — the feeling that we have on recovering from deep sleep, or a troubled dream, or a 62 THE FIELD OF MENTAL SCIENCE. partial delirium, of an effort on our part toward self-control, self-possession ; the taking up anew by the mind of its vol- untary activity. The mind returns to self -guidance by an exertion, and false impressions are dispersed. It recognizes two states, a normal and an abnormal one ; that in which the mind controls its impressions, and that in which the mind is controlled by them ; and it asserts itself in behalf of the former. A kindred experience is often a salient feature of incipient insanity, and may, when a firm will accompanies the effort, oppose a strong barrier to its pro- gress. (6) For the very reason that we ascribe delirium to the overpowering effects of physical causes are we disposed to refer rational action to the control of the mind. If insanity is due to disease, if a disordered brain brings disordered imagery, and it is an incident of this state that physical conditions control mental ones, then we readily believe that a healthy brain may prepare the way for the reverse action, and yield itself as an obedient instrument to the spirit. Thus cramps and convulsions in the muscular system are due to the escape of stimuli from the control of the will and its automatic relations. There is something certainly in the coherence of the two kinds of facts which seems to show these dependences. Hallucinations are fixed, obsti- nate ; sane impressions are flexible, amenable to influence. We seem to be dealing in the one case with a stubborn, physical tendency ; and in the other with a changeable, moral state. In delirium, the senses cease in part to be the media of facts ; the ways of ingress to the mind, like those of egress, are choked. In health, the movement inward and outward is alike free ; the brain is the medium, as it should be, of activity starting from either extremity. "We affirm, therefore, a reciprocal interdependence of the brain and the mind, with a normal government of the THE BRAIN AND THE MIND. 63 brain by the mind. Each may initiate action, but in all our high, characteristic activities, the agent is the mind it- self. We object then to the statement given below from Spencer, and to many kindred statements, as at once un- proved, improbable, and wholly destructive of the integrity of mental facts. "What is the meaning of the human brain ? It is that the many established relations among its parts, stand for so many established relations among the psychical changes. Each of the constant connections among the fibres of the cerebral masses, answers to some constant connection of phenomena in the experience of the race." — Principles of Psychology, Vol. I. p. 468. The relation of the brain to the mind is a subject un- dergoing close inquiry and careful experiment. The con- clusions already reached by Ferrier and others make it com- paratively certain, that the structure of the brain and ner- vous system in man is one throughout of definite relations, fitted to the double work of conveying distinct impressions inward and correspondingly distinct energies outward. In- deed, if there are any well adjusted connections in these channels of communication, we may well expect them to be complete. Their value and fitness must evidently depend on this continuity. The exact cannot pass at pleasure into the vague, the well ordered into the unordered, and retain their value. While these nervous connections are relatively precise, they are by no means single, similar results may be reached in more than one way. We may figure the cerebrum as offering definite termini for definite impres- sions from the exterior world, and also definite starting points for reciprocal energies. Here is a result like, and yet very unlike the statements of phrenology. The ner- vous system, has, as an instrument, fixed relations as cer- tainly as a type-setting machine, or an organ, though, un- like these, it conveys impressions both ways, as the recep- 61 THE FIELD OF MEMTAL SCIENCE. tive and active powers are in close correspondence. The definitness of the nervous dependencies of the organs of sense carries with it a like fixedness in the means of expres- sion. Thus by experiment and observation the conclusions have been reached, that ' the third left frontal convolution ' is the initiative surface for intelligent speech ; closely con- nected therewith are the starting points of that action which issues in writing. The parietal region or crown of the head constitutes the motor region of the brain, and may be farther subdivided into specific forms of action, as grasp- ing, clenching the fist, swimming. The temporal lobes are centres of sensory perception, both passively and actively, and are divisable with some certainty between the several senses. The occipital lobes or back portion of the head are conjecturally centres to organic activity, as digestion. The frontal lobes are the surfaces for the initiation of intel- lectual and moral action ; of attention, inhibition and direc- tion. It must be borne in mind, that experiments of excita- tion in any portion of the cerebrum can go no farther than to indicate the nervous sources of those movements which express certain mental states ; the conditions of the state themselves are not disclosed. It does not follow, therefore, from these conclusions, that pure thought and the communication of thought are equally dependent, or dependent in the same way, on the nervous system. All that the experiments of Ferrier, Fritsche, Hitzig, serve to establish is the general complete- ness of the mechanism of expression. Indeed, the very connections of thought in things belongs to the mind ex- clusively ; even when a verbal sign is put for them, the word of relation is no more than a sign. We may well be- lieve that the same thing is true of that more subtile lan- guage of molecular action in the cerebrum. Guiding or at least steadying, our thought by the im- THE BRAIN AND THE MIND. 65 age of an organ under the hand of a musician, it is suffi- ciently plain, for reasons already given, that we cannot wisely affirm this organ of the nervous system to be a hand-organ, whose tunes are pricked into its revolving barrels, and whose revolutions take place mechanically under external forces. Nor is it probable that the tune— the pure thought and feel- ing — lies open, like the music above the key-board; that the mind reads its own impressions in the molecular lan- guage of the cerebrum. In other words, it is not probable that thought as thought has any other than an instrumental, symbolic expression in the brain, such as it finds on a written page, or in spoken words. The proof of our first proposition, that activity of brain always accompanies activity of mind, implies a close rela- tion between the two. What is that relation ? The brain on the one hand, is the recipient of definite impressions from the exterior world. These, offered in their last physi- cal form as molecular charges, are not by the mind received in that form, but are translated into states of consciousness. These states give occasion to other more active states, which, with a similar subtile transfer, are accompanied by molecular changes, the agents of outgoing action. The word, articulation, offered to the eye or sounded in the ear, carries a distinct change to the brain, which becomes to the mind the occasion of an idea. On the other hand, the mind having occasion through the presence of an idea to write or utter that word, initiates in the brain in a fixed way the needed action. This translation in either direction is an ultimate mystery, both of whose terms even the mind does not directly recognize. "Words are the counters of thoughts, and thought can not proceed far without them. But words are sustained in the mind passively or actively by one or other of these molecular occasions. When the mind, there- fore, is thoughtful, those incipient impressions, which are' 66 THE FIELD OF MENTAL SCIENCE. the first terms of expression, and which are the physical symbols of words, may well be present and take the place in their supporting power of written or spoken language. Thus the musician may lightly touch the keys of his key- board, as a means of helping onward the inventive process. In each case the perception of the mind is not in its familiar symbols, but is simply sustained by them. Intense thought is fatiguing, because it involves con- stant movement among the physical counters of the mind, these incipient symbols of utterance in the molecular changes of the brain. We know what a relief it is to substitute the passive for the active symbol, a bit of paper with its few figures, for the vanishing figures of a problem worked out, as we say, in the head. The fatigue of extemporary speech is very great, for all these active signs of thought are fully present, and completely uttered, with even-paced rapidity. Intense thought without utterance is less fatiguing, because the process of expression is abridged, — the words hastening by us with large representative power — and the movement is left to shape itself to the inner impulse. Mere revery is scarcely fatiguing, because an indolent impulse is obeyed, both in direction and rapidity. Our inference then is that the brain, as the medium of all expression, becomes, in its definite molecular changes, a sustaining force to the lan- guage of the mind ; as written words, through another set of impressions, more passive, are an occasion of an appre- hension by the mind of another's thought. The very essence of thought therefore, is no more held in the molecu- lar type of the brain, than in the type of the printed page ; indeed hardly so much so, as the one is an unrecognized, and the other a recognized, condition of the appropriate thought. In each case it is the presence of pure spirit that evokes thought, and carries it forward ; the symbols, whether those of receptivity or activity, whether those ABNORMAL MENTAL STATES. 67 starting with the outer termini of the nerves and passing in, or starting with the inner termini and passing out, act simply as supports to its rapid footsteps. § 10. There are certain abnormal mental states that de- serve a passing notice. The chief physical change in sleep is a large reduction of blood in the brain. Its external fea- tures are the suppression of voluntary action and of the action of the senses. There may always remain, and there certainly often remains, the play of the imagination known as dreaming. The mental action seems to be sympathetic with the bodily state, and to be attended with very little control. While complete sleep involves the large arrest of voluntary life incident to muscular repose, there are many partial forms of it. The senses may remain cognizant of very many events ; a slight uneasiness or a gentle push may call forth a change of position. Words may be spoken ; or, more rarely, words may be listened to and answered, if in- troduced in the line of existing impressions. In somnambulism these states of partial wakefulness assume an extreme and troublesome form. They are characterized by an unusual acuteness of impression in some directions, with the ordinary want of it in other di- rections. The dividing line between waking and sleeping, active and dormant, powers is drawn with unusual decision, and in a new direction. Incident to this is also a new rela- tion of voluntary to involuntary action, the latter taking up what usually falls to the former. Hypnotism, mesmeric states, table-tipping, second-sight, and kindred facts, are phenomena of somewhat the same order. They involve an unusual suspension of some pow- ers, and an unusual activity of others. Normal associations in the action of faculties are broken up, and abnormal ones take their place. They are induced and established by unbalanced tendencies, by inheritance, by habit. Rev- 08 THE FIELD OF MENTAL SCIENCE. ery presents a like condition in a very moderate degree. A succession of images is vividly present to the mind, while the action of the senses and of the will is suspended. The degree of excitement to which an abnormal state may bring a faculty or a sense is sometimes illustrated in sick- ness. The slightest light or the least sound may be in- tensely painful, and passing events may impress themselves in quite a new way on the feelings. The nervous system under excitements or tension takes on an action quite novel to it. In hypnotism and mesmerism an abnormal state of wakefulness and of repose is induced by artificial means, the activity of certain faculties being as remarkable as the suspension of others. In the mesmeric state the patient — for we may fitly call the person subject to such disordered action a patient — becomes inattentive to the ordinary con- ditions of action, and highly sensitive to those which pro- ceed from the person inducing the state. In hypnotism there is a like suspension of habitual sensations, and a kin- dred attention to other relations determined by previous association. We may ally the action to that by which we listen intently without seeing, or look through one eye to the exclusion of objects in the other. The states implied in hypnotism, while akin to these, are much more extreme, much more abnormal. In these and kindred conditions unconscious and auto- matic connections gain ground on conscious and voluntary ones. The eye, our most voluntary sense, is least attentive, while touch, or rather the organic stimuli allied to it, may be very active. Persons who have united hands thus be- come the unconscious mediums of impressions passing from an active agent at one extremity to a passive agent at the other ; and the latter, abnormally sensitive, marks the slightest change in the former. The least movement ac- companying the recognition of the right word or the right ABNORMAL STATES 69 letter on the part of the active agent, is transferred to the passive agent, and he, when allowed a choice of actions, words, or letters, reads correctly the mind of the former by virtue of impulses which quite escape ordinary observa- tion. In table-tipping, by mechanical tests, pressure is shown to be present when the parties to it are wholly unaware of it, and are exercising a measure of volition against it. In- voluntary states triumph over voluntary ones ; confused, sec- ondary and unconscious ones over clear and conscious ones. In the planchette we have a visible record of automatic impressions escaping from the control of the voluntary life. Those who are the most coherent, rational, and self -guided in action are the least subject to these abnormal conditions, while those most impressible, excitable, weakest in their voluntary life, are especially liable to them. By repetition these states gain power with a corresponding loss of self- control. Notwithstanding the exalted susceptibility im- plied in them, they are to be regarded as intellectually and spiritually unwholesome. In these states, the automatic life, the life of obscure, physical impressions, gains ground on the reflective life, in a confused and confusing way. (1) There is a new and abnormal division of activities be- tween the two ; (2) in the unconscious life there is intense activity in unusual directions ; (3) in the conscious life, un- usual inertness in usual directions. §11. We dwell at length on consciousness as including the entire range of mental phenomena, because thus only can we adequately define the field of mental science, and keep it forever distinct from all physical inquiries. Physi- ological facts are of incalculable interest and value, but are perfectly distinct from philosophy. Each branch is capable of independent development, nay must receive it, and neither is as obscure as the connections between the two. 70 THE FIELD OF MENTAL SCIENCE. Only by a double light on either hand, the mind being made known to itself, and the brain and nervous system being carefully inquired into, can we hope to trace obscurely and slowly the dependencies of the physical and spiritual worlds ; even then reaching everywhere ultimate facts be- yond our solution. Metaphysics, with all its erratic and fanciful reasonings, never gave explanations more absurd and inadequate than those sometimes rendered of intel- lectual phenomena from a study of physical organisms. The assertion that the brain secretes thought, is the crude idea out of which, with more subtile and obscure phrase- ology, those impotent reasonings from matter to mind arise. This premature and preposterous union of the two realms, or rather absorption of the one by the other, is greatly aided by the admission of a region below conscious- ness, a region in some way attached to the mental field, though not fairly located in it. The mind thus allies in conception its phenomena to those of the physical world, taking place under a blind play of forces, and then readily unites them to nervous and cerebral action. Hypothetical, unlocated, unknowable facts are thus made to furnish a passage between the two departments ; to give inlet to lower physical causes, whose service it ostensibly is to ex- plain, but which really obscure and destroy, intellectual and spiritual powers. We reject this region of subconsciousness as unexplored and inexplorable, either by the inner or the outer eye ; as furnishing no ground for induction or safe deduction ; as necessarily a region of myth and fancies, offering no solid explanations which can be subjected to any form of experi- ence. Let positive science give us its positive facts, estab- lished with sufficient inquiry, located in the brain and as- sociated organisms — facts as material and sensible as those of brass or iron, oxygen and hydrogen, heat or electricity, EXPERIENCE. 71 and as physical facts we will recognize them ; let philosophy declare what the common consciousness can verify, and its statements shall be accepted as at least of equal value and validity with those which creep into the mind through the eye and the ear ; but let neither form of investigation bring alleged facts from a region which it itself puts beyond the entire range of our critical faculties. Consciousness pre- sents a distinct and independent field. On it no purely physical inquiry can enter, and in it philosophy can lie in- trenched beyond the power of any form of ignorant or jealous scepticism. The students of Positive Philosophy, ready to desecrate this sanctuary of our spiritual nature, will, like the blind men of Sodom, weary themselves in vain to find the door. Mental science will also be aided, by this divorce of the unknown from the known, the conjectural from the estab- lished, in bringing its own doctrines to a more decided test ; and in expelling some of those dogmas, which, unin- telligible., yet possible to a bold and blind faith, have hovered about it, and given it a superstitious, visionary, and unphilosophical appearance. Of this nature is the asser- tion, that one may sin below consciousness, or the belief that sin is transmitted from parent to child. If all the acts and states of mind are conscious ones, then, of course, all moral phenomena must transpire in the light. We are ready to accept and consider any well-established dependencies between mind and matter, but many of the theories on this subject transcend by almost their entire breadth all known facts, and bring no light to what they discuss. We are now prepared to receive physical phe- nomena as expressed in physical terms ; mental phenomena as stated in mental terms ; and any relations that can be es- tablished between the two. Our philosophy is thus on the ground of experience. Data of which our experience takes 72 THE FIELD OF MENTAL SCIENCE. no cognizance, states of mind beyond consciousness, states of matter that are in effect intellectual, intelligence out- side the range of mind, are all swept away. We get back to our primitive phenomena, and satisfy ourselves with striv- ing to analyze and classify them, and to point out their character and dependencies. This is the true, and the only true attitude of a truly empirical philosophy. Words can never be more clear than the very phenomena which they designate. Words of mind, therefore, should be constantly and exclusively ex- pounded by phenomena of mind, and, words of matter by the phenomena of matter. Having duly grasped and grouped each set of facts under their own forms, we are prepared to discuss inductively and deductively any relations between them. To systematically set ourselves the task of expressing the facts of mind in terms of matter and motion is to institute an effort, patient and skilful it may be, to subvert the facts of mind. Physical philosophy erred in the beginning by using words of mind to expound physical facts, mental philosophy now errs by substituting physical words and imagery for mental facts, presented in conscious- ness. This is the logomachy of our day. § 12. The second preliminary inquiry referred to — Is the mind always consciously active ? — is closely allied to the one now answered — Is the mind ever unconsciously modified ? A negative answer to the second inquiry would seem to prepare the way for a positive answer to the first. If no phenomena of mind transpire below the surface, then we should anticipate that the continuous existence of the mind would be productive of continuous activity above the surface, and that some phase of thought, feeling, or volition would be ever transpiring. The second question of course contemplates a modification of mind in the nature of an action, or an induced change of state, and not at all the CONSTANT ACTIVITY OF MIND. 73 admitted fact that the mind increases in power. The sub- jective method of this increase is beyond present explica- tion ; we are simply not to figure it under a material form, as if it were a substantial change. If, on the other hand, we say with Sir William Hamilton, that there are uncon- scious modifications of mind, we have prepared the way for denying its constant, conscious activity; since some mo- ments of being, at least, would seem to be sufficiently ac- counted for by the occurrence of these subconscious facts, and the existence of such facts would prepare the way for their hypothetical occupation of the mind in periods of ex- ternal repose. Yet, Sir William Hamilton answers this question, justly we believe, in the affirmative. The mind is always consciously active. The reason which most avails in bringing us to this con- clusion is one which will probably have little weight with most minds. It is of an a priori character. The only proof of existence is some form of phenomena. Existence without phenomena is unevinced and unintelligible. Mat- ter that should manifest neither active nor passive effects anywhere, under any conditions, would cease to meet our idea of matter, would be non-existent. ISTow the sole known phenomena of mind are those of consciousness ; and to sup- pose a total arrest of these leaves the mind, for the interval, without the proof or the form of existence. We may figure, in some vague way, under the analogy of matter, some pas- sive state or power as belonging to the mind and maintain- ing for it a phenomenal existence during the hours of sleep ; but here again we are in the region of pure hypothesis. We know nothing of mind save as the source of certain activi- ties, and if these are gone, the only grounds on which we ever predicated its existence are gone. To suppose it capa- ble of existence in a passive state, is a supposition alto- gether beyond knowledge, and made tenable only by analo- <4 THE FIELD OF MENTAL SCIENCE. gies carelessly caught up from the physical world. We believe, therefore, in the constant activity of the mind, as the only state under which we know it at all, or in consistency with what we do know of its nature, can at all conceive it. The notion of total rest leaves the mind as mind without any possible manifestation or proof of existence to any being under any circumstances. The only known phenom- ena of mind are removed, and with them pass away the evidence of its present being. Urging, however, no farther this consideration, we be- lieve the strictly inductive proof sufficient to render the conclusion, that the mind is always active, at least probable. As it is dwelt on at length by Hamilton, we shall treat it briefly. The chief difficulty to be overcome in the affirma- tion, is the admitted fact, that the memory does not retain and report the movements of the mind in hours of sleep or of syncope. How strong is this objection ? Much the larger share of the thoughts and the feelings of yesterday have en- tirely passed from the mind, and yet we readily believe in their existence. We have no doubt of the continuity of thought in our waking moments ; yet we arrive at the con- clusion more from our present experience than because we can recall one in* ten thousand of the feelings which have passed through the mind in the last dozen years. Now the impressions of dreams, when these are known to have oc- curred, are of a much more evanescent character. At the very instant of waking we may be able to recall them, and yet we lose all hold on them in a few moments. We also know that in proportion as sleep is sweet and sound these impressions of the night are fleeting, and must be caught almost in the very act of passage, or they are wholly lost. It has happened to many, perhaps to most, to awake in a dream, and to take delight in the images left by it, and yet after another hour's sleep to be unable to restore them. CONSTANT ACTIVITY OF MIND. 75 The memory also seems to be especially affected by phys- ical conditions. Fatigue and nervous exhaustion for the time being greatly diminish its power ; some forms of dis- ease erase its impressions in whole or in part, while the weakness of age first betrays itself in this faculty. Since, then, physical conditions so obviously and directly modify this power, it is but natural to expect that so great a change as that from wakeful activity to sleep might decidedly affect its action. The thoughts which pass through the mind in revery or abstraction often leave very slight traces. Sud- denly startled from such a waking dream by a practical claim, wc can scarcely, the moment after, recall what it was which so occupied us. These facts are sufficient to over- come the antecedent improbability of continuous mental action, arising from the want of memory, and to leave the way open for proof. The most obvious facts which go to establish the con- stant activity of mind are dreams. The memory does tes- tify to a large amount of movement in hours of sleep not to be distinguished by external signs from other periods of repose. Some habitually dream : that is, the play of imagery, the dumb show in the hours of darkness, the spectral troop of the sportive thoughts pass and repass within the scope of mental vision, and the person, on waking, remains mindful of this fleet, flitting assemblage — of this masquerade of his thoughts escaping the control of the senses aud the vol- untary life. Now, though others rarely dream, that is, rarely recall these shadows of the mind, leaving no more visible traces on the external life than do the clouds that fly through the heavens on the earth which they darken for the moment, this fact goes but a little way to weaken the presumption, that they are not very different from their fellows ; that the rehearsal of dreams is only a little more interior and close-locked in the one case than in the other. 76 THE FIELD OF MENTAL SCIENCE. This supposition is strengthened by the fact, that the habit of recalling and relating dreams is said to confirm the ten- dency to them, and to deepen their impressions. It is not probable that one dreams more as the result of reciting dreams, but rather that the deepened attention strengthens the memory of dreams. The nature also of dreams is a proof of their continuous presence. There is shown in them a certain freedom, yet also a certain weakness, of the mind not found in the wak- ing moments. The intellectual powers are plainly divorced from the usual restraint and guidance of the senses and the voluntary activities. Nothing seems monstrous, that is unnatural. The most incongruous events are accepted with perfect composure. The laws of nature are largely set aside, and the mind binds together, with its own fanciful connections in its own fanciful creations, the events that arise before it. The inner wheels are ungeared from the outer world, and revolve in their own rapid and irregular way. This fact goes to show that the senses are in full re- pose, while the mind retains this wild, free, sportive, un- tiring activity. In dreams, also, the will, through the repose of its physi- cal instruments, seems utterly powerless. Flight, however urgent the apparent necessity, is impossible. ~No personal exigency is met with physical prowess and strength. This seems to arise from the fact that the will finds itself thwarted by the inert, sleeping body, and not inducing its wonted effects in this torpid mass, throws back on the mind fear, faintness, and a sense of hopeless failure. Some- times, indeed, the effort it puts forth is so great as to run, like an electric shock, through the muscles, and the awak- ened body is landed at a leap, startled and astonished, on the floor of the chamber. These facts all indicate that physical repose is accompanied with mental activity, and MIND ALWAYS ACTIVE. 77 not simply that sleep is partial and disturbed. Such a state, indeed, affects the character of dreams, and deepens their impression, and thus aids us in recalling them ; but does not seem to be their cause. A third fact looking to the same conclusion is the familiar one of talking in sleep, though the person on wak- ing retains none of the impressions which occupied the mind. In such cases, mental activity is fairly shown to exist without corresponding recollection. The dog even will bark in his sleep, tickling the motor nerves with some tantalizing image of cat or rabbit. Allied to this is the fourth general proof furnished by somnambulism in all its forms. In these cases, the mind acquires a partial control of the body, and, while leaving the senses chiefly at rest, guides and stimulates its mus- cular powers. The wonderful precision and daring with which this is sometimes done evince great calmness and activity of the faculties, enabling them to reach results im- possible to the frightened, swimming senses. Of this char- acter are those familiar instances in which the somnambu- list passes through positions of great peril without failure or disturbance. A student in my own college class had been greatly interested and perplexed by a difficult prob- lem. He could not hit upon its solution. He retired to rest, and, in the night, rose in his sleep, and wrought it out on the board in the room. There, to his astonishment, he found it in the morning, the whole labor having left not the slightest trace in the memory. A fifth fact looking in the same direction, is that tes- tified to by Sir William Hamilton, and open to anyone's verification : " I have always observed that when suddenly awakened during sleep (and, to ascertain the fact, I have caused myself to be roused at different seasons of the night), I have always been able to observe that I was in 78 THE FIELD OF MENTAL SCIENCE. the middle of a dream. The recollection of this dream was not always equally vivid. On some occasions, I was able to trace it back until the train was lost at a remote distance ; on others I was hardly aware of more than one or two of the latter links of the chain; and sometimes was scarcely certain of more than the fact, that I was not awakened from an unconscious state." One more fact remains of very general prevalence con- firmatory of those now given. The mind is found to ex- ercise a certain measure of watchfulness over the body in hours of sleep. We sleep, as popular speech has it, with one eye open. Anything unusual, though slight in char- acter, arouses us, while familiar sounds pass unheeded. There is evidently a sentinel posted, who reports at once anything alarming, while he suffers ordinary events to pass unchallenged. We see something of this even in the tor- por of intoxication. The mind makes an unsuccessful ef- fort to arouse the body on the approach of danger, and, if the danger is extreme, sometimes sobers the man at once. We assign the mind a specific duty. We lay upon it as a task, that it shall awaken the body at a given moment. The mind is frequently disturbed and made nervous by the imposition, and arouses the vexed body in a tentative way half a dozen times before the hour arrives ; or, better trained and more familiar with its service, it leaves the re- pose unbroken till the moment has fully come. These and kindred facts of observation seem sufficiently to establish the constant activity of the mind, and to render it certain, that this invisible agent of invisible phenomena has a continuous and manifested existence, whatever the condition of its factor, the body, may be. CHAPTER II. The Intellect — Its Divisions — Perception. § .1. The first great class of mental faculties are those of the intellect. When we speak of faculties, we mean the different ways in which the one individual mind acts, rather than a combination of distinct powers under the analogy of our physical organs. The forms of knowing are treated first, not because they necessarily arise first — feel- ing doubtless precedes them, and chiefly occupies conscious- ness* in the first months of life — but because, in the activity of mind, they prepare the way for emotion and choice, and chiefly determine their form. The knowing are the recep- tive processes, and give material to the feelings and alterna- tives to choice. The intellectual powers have been divided into three principal classes ; the sense, the understanding, and the reason. The first furnishes the direct facts, the forms of existence which the mind contemplates, whether of the outer or inner world. The second carries on and sustains the processes of reflection concerning these, elaborating them into knowledge. The third furnishes those necessary ideas under which only the movements of a rational mind can go on. We shall not pause to speak of these divisions, as all that we have to say under each of them is requisite for their perfect comprehension. We proceed to treat of the first of these classes, that of sense. This term is somewhat awkward, but as it has already been used in this connection, we avoid, by its retention, one 80 INTELLECT. great evil of metaphysics, a perpetually shifting nomencla- ture. The sense includes two, and quite diverse sources of knowledge ; the power of perception and the immediate cognizance which the mind has of its own states. Under an image, but very partially applicable, they may be spoken of as the outer and inner eye of the intellect. § 2. In perception we shall not, as is usually done, in- clude all the senses. A portion of these seem primarily avenues of feelings rather than of percepts. When the sensation is manifest, lying in the organ, and contemplated there as an occasion of pleasure or displeasure, the sense is evidently one of feeling rather than of knowing. Though we may make the peculiar character of the odor or of the taste a ground of inference as to its source, and thus of knowledge, this fact does not destroy its primary connection with the feelings. JSTor is the fact that an odor, a flavor are, as it were, a form of knowing, a knowing that cannot be otherwise arrived at, a ground of classifying these sen- sations with the intellectual faculties; since the same is true of love, sympathy, anger. The perplexity arises, as has been already intimated, from the fact that every feel- ing involves consciousness, and to know a thing and to be conscious of a thing are constantly used as interchange- able expressions. As consciousness belongs necessarily to thought, feeling and volition, it is not in this common con- dition of their existence that their differences are to be looked for ; but in the nature of that existence, conscious- ness being conceded. All, then, that abides in the organ as a distinct, local sensation, an incipient, or a positive pain or pleasure, is a matter of feeling rather than of perception, and should be classified as a portion of our emotional na- ture. With this distinction in view, we have but two un- mistakable organs of perception, the eye and the ear. Even these, under certain conditions, may give rise to sensations. PERCEPTION. 81 The light may become so bright as to be painful ; the sound so loud or so sharp as to be disagreeable, that is or- ganically disagreeable, and thus these senses serve for the time as avenues to feelings rather than to perceptions. The pleasures that enter the eye and ear in painting, sculpture, music, not being organic but mental, do not interfere with the purely perceptive action of the senses. In perception, material of knowledge, or of subjective emotions simply, is, through the medium of the organ of sense, brought to the mind. It is only by observation that we know that the eye is the means of sight, or the ear of hearing. Neither of these organs, in their healthy state, give any direct indication of their office, or excite us by any passing sensation in the performance of it. To this fact our language conforms, and we speak of perception, an acting of the mind, through, rather than in, the organ em- ployed. The sense of touch seems more mixed than any of the others. It declares its locality, and lodges its results as dis- tinct feelings in the finger-ends. Its sensation should, therefore, be primarily ranked with the feelings, and it be regarded as an organ of feeling. Indeed, this conclusion language seems unmistakably to indicate, and in designa- tion we have passed over with the same Word feeling ', from the external sense to the internal emotion. Touch, how- ever, approaches the two higher senses, in the fact that its sensations are made almost exclusively the ground of infer- ences rather than of enjoyments, and when highly devel- oped are clear and ultimate in the information imparted,, and almost wholly overlooked as forms of feeling. The- blind doubtless cease almost entirely to contemplate the agreeable and disagreeable in touch — indeed the tactuaL character of these sensations — and find in them a direct, uiit conscious medium of knowing. Under such circumstances,. 82 INTELLECT. the sense is one of perception rather than sensation. A difference between sensation and perception is fonnd in the direction of the lines of activity ; in sensation it is inward, in perception it is outward. Sensations are converted into perceptions by making their data a subject of analysis and of inference. As the reflective element gains ground, the feeling is obscured, and the particular sense becomes the inlet of knowledge. Perceptions then are sensations trans- formed into terms of knowledge by the mind that lies back of them. Hamilton's statement, that perception is inversely as the sensation, if not mathematically true, is proximately correct. § 3. Taking the eye as the type of the intellectual senses, we ask, What do we see ? Most multiform and per- plexed have been the answers to this question, and most fatal, and, to the common understanding, preposterous have been the conclusions drawn from them. It is no part of our purpose to dwell on these either by exposition or refu- tation ; but rather to state what we regard as the just view, and with passing indications of its bearings to leave this to displace them. The nature of this view, and therefore its grounds, are so much involved in our idea of the intuitive action of the mind as to turn upon this fundamental fea- ture of philosophy. The full reasons of our conclusions cannot therefore at once be spread out, but will be slowly made up as we present the entire furniture and action of the mind. The separate parts of our structure can show neither their strength nor fitness till the survey of the whole is finished. In the first place, the eye as an organ of perception deals with color, the ear with sound. The sources of these colors and sounds are known only inferentially. It is a necessary belief, arising under the notion of causation, that these organs can become means of cognition only through PERCEPTION. ©o effects which have been wrought in themselves, and that unaffected they can be the medium of no knowledge. Effects not only demand causes, but causes efficiently pres- ent in them, interpenetrating them. The last, the imme- diate cause is inseparable from the effect. Now the vibra- tions of light and sound are the agents and the only agents that reach these organs, and it is a matter of experience that perception is immediately dependent on these agents as they penetrate into and work their changes on the organs of sense. Each organ is obviously fitted for the action of its own agent, and every interference with these internal adjustments destroys perception wholly or in part. While, therefore, our necessary beliefs demand an immediate effect on the organ of perception, experience clearly points out the agents of this effect, and the contrivance by which it is wrought. The purely intellectual character of sight, the extent to which the eye is an unconscious, translucent medium of the mind, is shown by the number, delicacy, variety, and fur- tive character of the judgments inextricably involved in vision. The earlier years of life are evidently busily em- ployed in learning to see, not in the scientific but in the familiar use of the word. These facts harmonize with the further recorded fact, that the eyes of one couched in mature life, seemed to report all objects under the analogy of touch ; that is, as directly in contact with the organ of vision.* These spaces, greater and less, which the educated eye now reveals ; this opening up and spreading out of the universe before it, this unsearchable depth, this height, this * The case referred to is that describe^ by Voltaire. The operation was performed by Cheselden on a lad who had been born blind. " It was long before the patient could distinguish objects by size, distance or shape. Several other like cases have been reported." — See Diderot, by John Morley, p. 54. 84 INTELLECT. breadth, are not the products of direct vision, but of vision modified by innumerable judgments, and mingled with them. The most of them we form unconsciously, and learned to make early in life, their accuracy and ease being increased by every day's experience. How many things come in to determine our estimates of the distances of sur- rounding objects, the clearness or faintness of colors, the depth of blue cast upon them by the atmosphere, their apparent size, intervening objects and the muscular ad- justment of the eyes in their perception. Most have prob- ably experienced in some moment of relative abstraction, an exaggerated or false impression made by some object or objects seen, but not observed, and marked the instantane- ousness with which these flashed into their true form upon the first distinct direction of the eye toward them. The relative position and size of objects are also almost wholly a matter of judgment; the eye itself only records their angular separation. It reduces them to a map-surface, and leaves their relations and distances unrecorded. Angles, not lines, are contemplated by it. The distances outward from the eye, and hence laterally also, are wholly a matter of experience. To these judgments are to be added those which turn on light and shade, and from these data arrive at the most complex surfaces. We thus see that the pure visual data of sight are very meagre, and bear no more resemblance and intimate connection to the world in which we live than do the canvas and the paints thereon, as canvas and paints merely, to the landscape represented. This saturation of a sense by the understanding, this inflation of a single drop by the breath of rational thought into a brilliant sphere, and the acquired ability to do this as child's play, are the noticeable features of this, our highest organ of perception, quite distinguishing it from such an organ as that of taste, PERCEPTION. 85 from which with smack, pause and reiteration, we reach one or two uncertain conclusions. The ear is akin to the eye, though considerably below it, in the number of judgments its habitual use involves. The direction, distance and source of sounds are plainly learned by experience ; though in most cases we hardly separate the mere phenomenal fact from the judgments on which our knowledge depends. To these are to be added all the variety of feelings expressed by intonation, and also that representative power of articulate sounds instituted in language yet through familiarity employed and interpreted without thought. Here again the under-play of the un- derstanding is very great, exploding a single ictus of sound, like a thimble of powder, into a death-warrant, or open- ing the gates of blessedness by the key of a monosyllabic assent. Thus does the mind work up the crude material, the physical nutrition of an organic susceptibility into the daily food and the special feasts of the soul. The point of most philosophical interest in these senses is the approach we make to a more exact answer to the in- quiry : What do we perceive ? Is it something external to the organ ? or is it something subjective to it ? or is it sub- jective to the mind itself ? If, in the word perception, we include all the mind's action therein, its direct and its infer- ential knowing, then plainly we perceive something exter- nal to the eye and to the mind. If, however, by perception, we mean only the arriving at those simple intuitive data around which these judgments cluster, and which they con- struct into the well-ordered and complete vision of mature life, then the mind perceives that only which is subjective to itself, and knows directly no more about the intermediate organ it uses than it does of the external object which is the joint, final product of its perceptive and reflective powers. The first spontaneous answer of philosophy has been, the 86 INTELLECT. direct perceptive action of the mind is confined to the cir- cle of its own activity, to consciousness; and probably no other answer would have been sought for, had not the con- clusions drawn from this earlier statement led to a recon- sideration of it. These conclusions have been idealism, and have compelled those who have wished to establish the in- dependent existence of the external world, and have had no other means at hand to do it, to re-analyze perception, and find therein a valid objective element. Overlooking the inferences of the mind, they have given it a direct knowledge of matter. The proof of idealism runs thus : 1. " We cannot know things in themselves ; all knowledge is subjective ; it is confined to unseen states and changes. 2. " If this is so, then still more is what we name the objective, only a state or change of us as subjective, it is a mere fiction of the mind so far as it is regarded as a beyond, or a thing in itself. 3. " Hence we do know the objective ; for the skepti- cism can only legitimately conclude that the objective that we do know, is of a nature kindred to reason, and that by a priori necessity we can affirm that not only all knowable existence must have this nature, but also all possible exist- ence must. Self-conscious intelligence must be, according to its very definition, subject and object in one, and thus universal." Hamilton has striven to break this charmed circle of the mind at the point of perception, affirming that a real objective element is directly recognized therein. He says, " We have no reason whatever to doubt the report of con- sciousness, that we actually perceive at the external point of sensation and that we perceive the material reality." " The total and real object of perception is the external ob- ject under relation to our sense and faculty of cognition." PERCEPTION. 87 " Suppose the total object to be twelve, that the external reality constitutes six, the material sense three, and the mind three ; this may enable you to form some conjecture of the nature of the object of perception." Is there any good ground for the very general and very stubborn conviction that the mind cannot, by way of direct apprehension, act on anything external to itself; or are Reid and Hamilton right in regarding this as a pure as- sumption ? It is very difficult and very important, in a discussion of this character, to be aware of the physical images which cling to our words and mislead the thought by material analogies. In and out, where it is and where it is not, are expressions applicable to matter rather than to mind, and we must not confound the intellect even with its instru- ments, the brain and the nervous system. The effects which take place in these are one thing, and what enters consciousness as a purely spiritual product, an inner experi- ence, is quite another. The connection between the two, an affection of the organ of sense and an affection of the mind, is unknown, and for the present at least insoluble. They are as wide apart in kind as any two known things can be, since the one is physical and the other spiritual, classes of phenomena for which we have found no common term. There seems some plausibility in the notion of external perception, when we contemplate the organism of any one sense, as that of the eye. The light enters. A sensible, visible effect — visible to another eye — is evoked on the retina. To this compound effect to which two agencies are contributing, the eye and the light, it may seem rea- sonable to regard the nerve as sensitive, and therefore to suppose it .to take cognizance of the immediate presence of a foreign agent. If, then, we could identify the perception of the mind with this condition of its organ, there would 88 INTELLECT. seem to be in it a direct knowledge of one force at least, that of light, alien and external to itself. But even on this supposition, farther reflection would modify our conclusion. In purely physical causation, the cause, though entering into the effect, is not as a cause recognizable there. Indeed it seems probable that there is not invariably the same transferred force in one series of effects as in another, and that in some results the prime agency quickly- disappears. A ball is struck by a bat and set in motion : after the ball has parted from the bat how much of the antecedent fact could be found in the sub- sequent one of independent motion ? How far would the second phenomenon directly disclose the first, or what com- mon term or force could be detected in the two '? The force is not discernible aside from the results it occasions, and antecedent effects are not given in subsequent ones. Suppose the same ball to be observed falling under the influence of gravitation. How far would this new cause be discoverable directly in this new phase of movement? Again, chemical action is initiated by a rise of temperature ; water is instantly frozen under certain conditions by a slight jar ; the brain is quickened by a full stomach ; in these and a thousand other cases of causation, what portion of the cause is in the effect, to be found there as a part in a whole, as the numbers, 6, 3, 3, in the sum twelve. Evidently in a purely physical effect it is impossible for us to detect the cause as a cause — as a second, alien agency, entering into and constituting a distinguishable part of the new, simple state before us. We perceive phenomena only, not the underlying forces, not the very causes ; these, and the antecedent facts they may have occasioned, are matters of inference and of ex- perience exclusively. If, then, the phenomena transpiring in the eye were, as they are not, identical with those of the DIRECT PERCEPTION. 89 mind, it would be impossible that these should include a knowledge of the very cause, and still less possible that they should include a direct knowledge of antecedent, ex- ternal phenomena reached only by inference through this hidden, unsearchable force or cause. We may direct atten- tion in this discussion to two things : the very cause, or efficiency which necessarily coexists with the effect and sustains it, and the immediately antecedent phenomenal effect, more often spoken of as the cause. The first of these is not discoverable in the eye, since no causes, as causes, can be directly known. To know phenomenally the very cause, would be to make that cause a phenomenon, that is an effect, that is not a cause. Pure causal being, the being or force that lies back of phenomena, cannot be known perceptively as a result. To affirm this is to deny causation, and make a phenomenon its own cause. The second of these, to wit, the immediately antecedent, outside effect, cannot be perceptively found and known in the eye for the obvious reason that it is not there. If, therefore, we were to direct the attention to the eye alone, and identify its states with those of the mind, we should still be unable perceptively to discover anything in it but its own phenomena, which are neither the outside object nor any cognizable portion of it. We are not to regard the eye with the facts that transpire in it as at once inseparable from the mind and external to it. If its changes are the changes of the mind, then all that is outside to it is equally so to the perception. So truly subjective, then, is even the organic state of the eye in sight, that were this the thing revealed in consciousness, we should still not be able to separate or distinguish the external element, " six," in the sum twelve, and know it directly as a foreign agency. The phenomenal six alone should we perceive, and still be compelled to infer hence the causal six supporting it. 90 INTELLECT. But when we pass beyond the condition of the organ as itself unknown to the mind and outside of it, and contem- plate the true immaterial content of consciousness, the case is, if possible, still plainer. Perception as an act of mind does not reveal to us the instrument of sense employed, or the state of that instrument. The connection between a mental state and the physical state which accompanies it, is mysterious and unknown ; it is not so much as hinted at in the very act of perception in consciousness. For aught that we can see, the last might be very different from what it is, and the first remain the same. Indeed, that there are to sight and hearing accompanying physical states, what these states are, and even where they are, constitute facts which require to be learned from experience. It has helped to obscure the doctrine of perception, that a distinct image on the retina has been found to in- tervene between the object and vision. This image has drawn attention as something especially necessary and ex- planatory. It is, however, in reference to vision a mere accident, as much so as the possible images which may be formed within the tube of a telescope. These images one and all play no part as images, but simply as causal links. The image on the retina is only one term in the physical agencies, which finally express themselves in the molecular changes of the brain. If we were to examine the eye in delirium, we should find no images in it of the objects evoked by the mind. Even in advanced life we do not always recognize at which ear a given sound chiefly enters, and tentatively test the question by turning the attention first in one direction, then in the other. The content of consciousness, then, is not of such a nature as to reveal in perception the states of the retina, or of the auditory nerve ; or whether there is in them more or less of foreign action. These changes are DIRECT PERCEPTION. 91 sunk foundations on which the visible structure rests, but are not in the least disclosed in their nature by it. They are the submarine cable, neither declared in its length nor its depth, nor in the mechanical nor electric conditions of its structure, by the messages sent and received at either terminus. To introduce causes into consciousness, that they may be there directly known, is either to assert their supersensual and immaterial character, is to grant the as- sertion of idealism, " We do know the object, and there- fore it is of a nature akin to thought," or it is to break down the fundamental distinction between mental and physical phenomena, affirming that both transpire in con- sciousness, and that the physical facts of the brain are the spiritual facts of mind. Yet having made this inadmissible concession, we are confronted with the fact, that conscious- ness does not of itself indicate whether the brain, or the heart, or the bowels, are the seat of thought ; whether we see with our fingers or our eyes ; and the further fact, that causes, as causes, are never discoverable even in purely physical effects. The assertion, then, that we cannot directly know things in themselves, follows inevitably from the two assertions : consciousness is the sole field of perceptive knowledge ; no material phenomena, as material, can appear in conscious- ness, interpenetrated, so to speak, by it. Consciousness covers all intellectual knowledge, and excludes all else ; lays down a line of demarcation impassable either from within or from without, cutting apart matter and mind. This conclusion we believe all experience confirms, and that no one would have thought of denying it, save under the pressure of certain difficulties to be evaded, and certain conclusions to be reached. § 4. How far pure idealism, that professedly knows only mind, is entitled to these assertions which we are 92 INTELLECT. ready to make in common with it, is a question of more donbt. We, in our position, arrive at them by an inferen- tial knowledge both of matter and mind, by a discovery of their mutually impenetrable character. If we were, as idealism asserts, in every way debarred access to matter — to matter as believed in by the masses of men — it would certainly not be so plain how we could come so universally to form a distinct, uniform and controlling idea of its char- acter, and be able also to affirm, that this most omnipresent and fixed of our notions is, in its essential features, a mere figment of the brain. Why a series of physical conceptions which is removed by the very nature of mind from even the bare possibility both of knowledge and being should nevertheless be the most uniform and universal of men- tal states, is not explained by idealism. How a form of thought, necessarily false, comes to be a fixed product and characteristic of mind ; how it happens that we continually talk, think and act in reference to matter — matter which by the constitution of the mind, is beyond all forms of knowl- edge ; how science and philosophy come to so utterly differ from each other in their beliefs, are mysteries which must ever, to the straightforward, practical thinker, reflect the highest improbability on idealism, and leave it among those strange, remote conclusions, which when not directly dis- proved are too far off to disturb the orbit of our daily life. When philosophy subverts knowledge, instead of expound- ing it, and denies the validity of the most settled, familiar and unavoidable judgments of the mind, it assumes an anarchical character, removing the foundations, if not of thought, yet of conviction. We believe the true doctrine of perception to be, that the state of consciousness therein is purely subjective both in action and object, indeed that the action and object are in- separable. To perceive a color, is to put forth a complete, PERCEPTION. V6 primary, simple act of knowing, complete in that something is known ; primary in that no further explanation can be forced upon it, the act standing in its own light, appre- hensible for what it is in itself ; and simple in that it is incapable of successful analysis. On the occasion of such a perception, the mind, of its own interpreting action, under the notion of causation, infers an external source of the impression, which, as a necessary, certain and uniform con- clusion, becomes to it as valid as any that it ever makes. Its validity, like the validity of all mental acts, is referable to the clearness and constancy with which it is made and repeated. We reach, then, the external world not directly by perception, but indirectly, inferentially, along a bridge of thought, whose farther abutment our rational nature supplies, and whose connections are established by varied, repeated and protracted experience. Shifting the figure we strike the shore with the grapple of causation, and by this guy we swing. If asked why the mind supplies the idea in connection with one mental state, that of perception, more than with another, as that of thought ; how it knows where and when to fling into the air its coil of rope, that it may thereby be lashed to the physical world, the answer comes : It is the fruit of varied experience. A sensation is found to be a new, distinct, sudden, independent state. As such it de- mands explication in an outside cause. A thought is a con- secutive, evolved, dependent product, that can be renewed in the mind at pleasure, and by this fact finds explication through the mind itself. The various senses also, in their diverse yet independent reports, mutually aid and guide the mind in this reference of sensations to external causes. Impressions in distinct organs are found always to accom- pany each other in certain forms, under a fixed order. Thus experience is constantly disclosing the character of 94 INTELLECT. phenomena, and the mind rapidly learns to distinguish those inwardly dependent on its own action, from those dependent outwardly on foreign agents. This class it can not, from its own constitution, leave without this causal reference and exposition. The confusion which sometimes overtakes the mind in perception, illustrates its method of education, and the manner in which it is commenced. A pressure is felt across the forehead, as if the band placed upon it had been drawn too tightly. "We cannot tell with certainty whether the impression is due to this cause, or to the astringency of a fluid with which the fillet was saturated. We test the point by raising the hand, and determining whether or not mechanical force is present. In the absence of this, we refer the feeling to the condition of the nerves. Again, we seem to hear a sound, as the anxious parent the crying of her child. She cannot at once decide whether the im- pression was the suggestion of her own thought, or the actual effect of the supposed cause. The attention is more carefully directed, the phenomena that enter the mind from without being discriminated from the mere play of fancy; and by this more complete separation of its own action from the action of other agents the point is settled. § 5. It has been thought, and much has been made of this point, that a denial of direct perception is an impeach- ment of the veracity of our faculties, or, as it is expressed by Hamilton and others, of consciousness; and that the way is thus logically opened to universal scepticism. Idealism is certainly not a denial of the facts of conscious- ness. Perception as a fact of mind, is accepted, and the first exception taken is as to what perception is, what it gives us. Now the veracity of consciousness is only in- volved in the mere fact of perception, the mere rehearsal and acceptance of its mental phenomena, not at all in the PERCEPTION. 95 nature and validity of its supposed revelations. Idealism does, however, set aside a most universal belief of mankind, and so far tends to scepticism. But this accusation does not hold against the view of perception now presented. The general belief of men in an external world is main- tained, though a careful analysis shows the grounds of the conclusion to be somewhat different from those at first ac- cepted. The accusation against idealism is not that it shows a general opinion to be groundless, but that it affirms simply and nakedly a general and necessary belief to be deceptive ; that is, the reiterated and constant action of the mind to be delusive. We may, on like grounds, pronounce the axiomatic conclusions of the reason unreliable. These are nothing more than its inevitable convictions. The affirmation in which the unaided powers of all men agree, which they spontaneously and inevitably make, is the existence of an external world, the opposition of matter to mind, a reference of a portion of our inner expe- rience to outer sources or causes. Whether this conclusion is intuitive, or involves one or more of the simplest acts of judgment, most men have never so much as inquired, and have therefore no convictions concerning it. It is doubt- less a matter of surprise to most persons to find, on in- quiry, so many judgments mingled with the simplest act of sight. These had been overlooked, and the act of see- ing regarded as more full, explicit and immediate than it is. Language favors this concealment of obscure, rapid judgments, and we are said to see the form of a sphere, when we merely infer it. Yet there is no ground for a distrust of man's faculties, because they are formed to act in ways and proportions not perfectly understood by those who accept results with no investigation of methods. To tell a man that the unlikeness of the images of the same object in each of his two eyes, is one of the grounds from $6 INTELLECT. which the impression of nearness is received, may interest and surprise him, but does not so shake his confidence in his own conclusions as to be told that the external world, in which he has so fully believed that he has never so much as thought of its existence as a matter of belief, is a mere creation of the mind, one portion of its own acts being thrown into opposition to another portion. The one assertion arrests and throws back in confused, eddying cur- rents, the whole stream of intellectual action; the other merely shows that analysis reveals more elements in mental phenomena than those at first caught sight of. There is no reason why the statement that there is a simple judgment involved in a belief of the existence of matter and of mind, should be regarded as any more destructive to the faith to be reposed in our faculties, than the generally accepted doctrine, that sight includes many judgments dependent on protracted experience. The assertion of Hamilton, " that consciousness gives a knowledge of the ego, in rela- tion and contrast to the non-ego," even if it were readily intelligible to all, would hardly, I think, be regarded as a satisfactory statement of a general and unwavering belief, when contrasted with the statement, there is found that in consciousness from which we directly and inevitably infer the existence of matter and mind. Most would doubtless regard the two statements as open to consideration, as lying alike in the line of the common belief in the external world. Indeed, to say that the mind is conscious of itself, or is conscious of matter, gives a shock at once to thought, and is far from being that explicit, indefeasible statement of the common faith which all at once recognize. The exact ground of the general belief is certainly open to inquiry, and one statement which accepts its va- lidity is no more exposed to the charge of a denial of the integrity of the human faculties than another. Indeed the PERCEPTION. 97 spontaneous conviction of the existence and nature of ex- ternal objects involves many judgments besides this one of causation. I see the apple before me. My present impres sion is — the steps of my past experience being unanalyzed — that I see it to be round, to be red, to be three inches in diameter and at a distance of three feet. How does this impression agree with what Sir William Hamilton says is the real object of perception 1 " Through the eye we per- ceive nothing but the rays of light in relation to, and in contact with the retina." Who ever perceived them, or came to so much as a knowledge of them, without diligent scientific inquiry? Light, as the fruit of much research, is found to be a form of motion, and this motion to affect the retina ; but no man ever knows the existence of the retina, or of the undulations of light thereon, save through an inquiry into eyes other than his own, and a careful in- vestigation of the physical world. What is here asserted to be the sole object of perception, the mind never per- ceives, but only employs it as a submerged, unknown cause through which it arrives at its own knowledge, to wit : a red apple of a given size and position. In this final pro- duct of perception, there are contained innumerable judg- ments, and it should certainly be no surprise to find among them this one of outside existence. That the spaces of the world are inf erentially given, is entirely in keeping with the fact that those of a painting are, by the previous habit and impulse of the mind, supplied under suitable sugges- tions of light and shade. The crude material granted to the mind seems to be a subjective impression of redness, of angular extension and various shades. From these, by the aid of muscular and tactual experience, and the help afforded by the color and relations of surrounding objects, it constructs an apple and assigns it independent existence in a definite locality. This 98 INTELLECT. it now does instantly, like a flash of light, though it has acquired the power of doing it slowly, by much and forgot- ten experience. The primitive, intellectual elements are wholly unlike this final result, these data of sense intershot with a few firm threads from the shuttle of reason. In- deed, no instance in our later knowledge, in which an entire system of principles is evolved from a few facts, more evinces the astonishing power which belongs to the mind, than does this simplest, earliest, most common case of rea- soning, that of perception. That color is known as the motion of an ethereal me- dium on the retina, or that there is any connection of the two, or knowledge of the one in and through the other, are statements not intelligible even, till science by second- ary inquiries has made them so. The transfer of motion at one sense into vision, at another into hearing, and in the brain itself into thought, are inexplicable transformations, whose terms we only know by independent investigation, and even then fail of their connection. To suppose that any portion of this knowledge comes directly in percep- tion, is the most obvious and violent perversion of experi- ence. If we were directly cognizant of the content in the organ of sense, cognizant of it for what it is and where it is physically, there would be no opportunity for deception or oversight in matters of perception. A force acting on a machine tells, and must tell, for exactly what it is. The effect is direct and inevitable. So would it be in percep- tion. We should never make a ghost of a stump, or over- look altogether the objects whose images are actually on the retina ; that have actually caused the light to impinge with customary power on this sensitive medium. It is because the mind gives a frightened attention or no attention, in- adequate interpretation or no interpretation, to these ob- PERCEPTION. 99 jects, that perception is distorted, or fails altogether. The mere physical effect in itself alone is nugatory. It is said that those whose eyes are distorted, use either one or the other as they choose, directing the attention to the right or the left as convenience requires, the impression in the neglected organ going for nothing ; and we all of us evidently take up and lay down at pleasure the physical effects on the retina, using them as means of vision only when the mind is at leisure to do so. These facts show without doubt, that perception is deeper than the organ of sense, is by no means identical with the appropriate action therein, nor is sure to follow it. It is, then, no impeach- ment of the veracity of our faculties to inquire into the ex- act mode of their action, nor any the more so because the inquiry discloses unexpected results. § 6. The only perfectly direct and absolutely simple form of knowing is the knowledge which the mind has of its own states and actions, because they are its own. When the state is itself simple and ultimate, we term it a sensa- tion, a feeling ; when the act is pure and simple, embracing all its terms within itself, we call it intuitive ; and the mind therein is an ultimate authority to itself. The intui- tive action and the consciousness of the intuitive action are one and the same thing. Thus that two and two make four is a statement that may receive illustration but is capable of no proof. The mind within itself sees and comprehends the relation expressed. It has a direct knowledge of it. When the act under consideration is one of inference, it re- mains known as an act directly, but is itself the ground of further indirect knowledge. Thus we see the kaleidoscope directly, but we also see in it something which depends for its form on the structure of the instrument. All phenomena immediately known to the mind are its own phenomena, since consciousness is the distinctive char- 100 INTELLECT. acteristic of this direct knowledge. Sensations, intuitions are its own states and acts declared in consciousness. From these phenomena it may infer noume'na, its own spiritual powers and the physical forces about it. The being of both stand on the same basis of valid inference. Phe- nomena of sensation and action being present, and the idea of causation present, the two coalesce reflectively into the conclusion of spiritual and physical agencies expressed in them. The three terms of the problem of perception are the external world, — external both to mind and body — the in- ternal or mental world, and the nervous organism which me- diates between them. The mind first recognizes its own phenomena of various orders ; from these it infers their sources in itself and out of itself; later it learns the part which the organs of the body take in this interaction. The analysis of Hamilton, and of a large share of the modern school of realists, (1) confounds agent and action, substance and qualities, noumena and phenomena, causes and effects. It for the moment overlooks the facts that causes can not be directly known, and that all direct knowledge is a knowledge of effects. (2.) It confuses the relation of consciousness and mental processes. Consciousness can embrace only phenomena of mind, and all direct knowl- edge belongs therefore exclusively to these phenomena. When an act of mind gives something not existing in the mind, that act is one of inference, whose premises simply are mental phenomena. The difference between intuitive and reflective knowledge for our present purpose lies just here, consciousness holds all the terms in the one case, and only the premises in the other. Intuition is a vision of the mind that completely searches the relation before it, and is wholly held in consciousness. If equals are added to equals the sums are equals. This axiom is not the statement of one PERCEPTION. 101 or of many known facts beyond consciousness, it predicates a relation of ideas within conscionsness. This odor comes from that rose. The odor as a state of mind is directly known to the mind, but the source of the odor in the rose is an inference only. (3) The analysis of Hamilton, while designed as a defence against idealism, opens the way to it, or to materialism, as the objector may choose. The facts of matter and mind are thought to meet in one field ; their identity therefore may easily be affirmed on either hand, the identity of material facts with mental ones, or of men- tal ones with material ones. Both it is said are known by a simple, indivisible process. (4) This analysis gets color by directing attention to one or two senses and neglecting the other senses. Take color and sound, and how manifest it is that our knowledge is made up of inner impressions, and the reference of these impressions to external causes. (5) The stimuli run along the neural route from with- out inward. The interior portion of this circuit remain- ing complete, the loss of the exterior portion is of no mo- ment, provided that the fitting stimuli can be imparted to the cerebrum. Apparent sensation remains in the hand after amputation. Hence the inevitable inference, that the molecular changes of the brain are the last and only neces- sary terms of cognition; that physical are translated into mental facts, not at the outer but at the inner termini. § 7. Each sense deals with a distinct medium, or with distinct conditions of the same medium. The eye as an organ of perception is affected exclusively by light; the ether in its vibrations is the only agent of vision. The light involves three points — gradations of intensity, color, and angular measurement. The first two are obvious, the third a little less so. It may seem to us that light gives directly through the eye a surface on which its colors are spread. These surfaces, however, in definite place and 102 INTELLECT. form, are the products of experience. Colors simply have their own character and subtend an angle. This is all that the mind, through the eye as an organ, directly declares. Surfaces are determined by distances both as to nature and extent, and the only fixed perceptive term out of which these judgments grow, so far as the eye is concerned, is an- gular measurements, which are the same for many different surfaces, the same for solids as for surfaces, and hence di- rectly involve no one surface. The floating muscse of the eye assume any size, according to the distance of the objects on which they are cast, though their angular dimensions remain the same. When we look out into space, the spherical impression of the concave is due to partial inter- pretation, with conditions of vision too narrow to make it complete. The stars are all thrown outward, but with, dif- ferences of distances undiscernible. The corners of a room would disappear, or the edges of a cube, were not our im- pressions of these made up of shades as well as angular measurements. The two we easily translate into position and surface. It is these elements that the painter deals with, renewing the impressions of vision by colors and shades, and by surfaces which act not by their absolute di- mensions, but by their suggestions of distance and position. It is true he reaches his angles through surfaces, but we do not contemplate them as surfaces of such and such dimen- sions on the canvas, but in their angular force through the eye as suggestive of the character and positions of known objects. The horse in the foreground covers more space in the picture than the mountain in the background ; and the size of the latter is impressed upon us, first by indica- tions of distance, and then by the angle subtended. The earliest and most fundamental of absorbed judg- ments is that of distance. Touch and motion come to initiate this judgment, and, the size of familiar objects be- SIGHT. 103 ing fixed, the eye carries forward the process by observing their angular or apparent size in various positions. Size and distance mutually contain each other ; if we know the one we can infer the other. We also judge of distance by intervening objects, them- selves interpreted in their relations by experience. Again we infer it from depth of color. This test, however, is ap- plicable only to remote objects, and, in comparative judg- ments, to objects quite unequally remote. The ridges of mountains, rising in succession beyond each other, are separated to the eye by the different shades of blue that rest upon them. The degree in which, in these estimates, we are dependent on our own experience is indicated by our wild conclusions under novel conditions. The pure at- mosphere and the unaccustomed dimensions of high moun- tains make the impressions of one who visits a country like Switzerland for the first time exceedingly deceptive. Weeks and months of laborious walking must be passed before these objects assume their true dimensions, and take on their real grandeur. In like manner the inexperienced landsman loses all accuracy when called on to estimate dis- tances on the water. A fourth aid in determining distances is the muscular adjustment of the eyes in bringing the image to the focal point on the retina. This test affords but slight assistance, however, and is applicable to objects comparatively close at hand. We are not conscious of a readjustment of the eye, except under a sudden change of vision at ranges whose inequality is very obvious. If an object near by is sud- denly thrust upon the attention, the effort to see it becomes even painful. The judgments of ordinary vision are vague, giving the general relation of objects in position more than their direct distance from the observer. A second perceptive judgment in sight is that of form. 104 INTELLECT. The conditions of this judgment are light and shade. In large and complex objects, like mountain ranges, as form involves position, our judgments in this particular are mingled with and modified by those of distance. A third judgment in vision is that of size. Though we infer distance from size, and equally well size from distance, the former is the more common case in experience. Well- established sizes, settled by close contact, are our ordinary data. Yet it not infrequently happens that distances are known, and we thence infer the dimensions of strange ob- jects. The variety and vagueness of our impressions as to the size of the moon are due to the fact that distance does not enter in as a measurable factor. The primitive data, then, of the eye are color, light and shade, angles ; its ac- quired data distance, involving position, form and size. Two eyes in vision give us an advantage besides that of protection against accident, or even that of stronger sight. The circles of vision in the two eyes of man do not quite correspond ; the one includes portions not found in the other. Also, by virtue of distinct positions, the two alter slightly, each as compared with the other, the relations of objects. These facts make the perspective more definite, especially as regards objects just at hand, in reference to which exactness is important. We secure by two eyes a triangulation available in defining distance and position. So important is this aid that the loss of one eye is attended with considerable confusion of perspective. An intervening object, also, brings less obstruction to two eyes than to one. The ear, the second leading organ of perception, deals with sound, with the undulations of the air within a certain range of rapidity. This range, however, is not identical in all persons, a rate of vibration being in some instances audi- ble to one .and inaudible to another. The physical charac- teristics of sound are quality or timbre, pitch and quantity HEARING. 105 or intensity. In addition to these primary qualities, there are secondary ones, which combine physical discrimination with rational association. Sounds are thus intellectual, emotional, and musical. The first two may each exist with- out the others, and the third quite modifies the two former in their combinations and force. Original discrimination and the modifications of experience enter, in a most com- plex and inseparable way, into the appreciation of these secondary qualities. The primary qualities of sound, like those of vision, develop a series of perceptive judgments, though these judgments are less numerous and important than those of the eye. We infer from a sound its source, as the presence of an acquaintance from his voice. These judgments rest wholly on the associations of experience. We judge of the dis- tance of any audible object by the intensity of the sound. This class of inferences arises under more uniform natural connections interpreted to us by experience. We also decide by sound, though with some hesitation, on the direc- tion of audible objects. In these conclusions we derive assistance from the possession of two ears. Direction is settled by the greatest intensity of sound, it being found in the line of the wave motion. In touch, taste, and odor we deal with matter in three form's — as offering resistance in masses, as floating in a gas- eous or most minute form in the air, as dissolved in water or saliva. In the first, the condition is mechanical, in the other two chemical. The things of which these senses take cognizance are, between the three senses, incomparable with each other, and, in the same sense, very numerous, with every gradation of difference. The sensations take on the variety and changeable forms of the feelings, as opposed to the narrow and definite action of the powers of the mind. Our perceptive judgments through these three senses are 106 INTELLECT. made up of the variable combinations of experience. They retain more firmly their distinct character than in the other senses, and are enlarged under a more directly guided effort. Yision contains the largest amount of completely absorbed judgments, judgments that turn on general principles, and are early taken up in an inseparable way with the percep- tion. The presence of these inferences are not only dis- closed by the mistakes we make under them, when our data are insufficient, — as in determining distances and relations under novel circumstances — but also by various contrivances by which we alter the conditions of sight, and so the appar- ent position and magnitude of objects. When we roll up a piece of paper like a funnel, and view a painting through it with one eye, the perspective is brought out more strongly. The increased effect of this monocular vision is due to the fact, that surrounding objects are cut off, the actual distance of the canvas, so distinctly given by both eyes, is obscured, and the data of the paint- ing alone are present. The eye, relieved from the contra- diction of near objects, proceeds at once to construct the landscape under the suggestions of the painter, with its true dimensions. A mask looked at in this way in the rear may, by an instant, unobserved transfer of light and shade from one side of the face to the other, be constructed with its features in relief, as if seen in front. In the skeleton form of a stereoscope, in which two pic- tures of the same object are separated by a card, so that one image is seen exclusively by one eye, and the other by the other eye, the two eyes unite to construct the view at a distance, as if they were looking at remote, real objects, in reference to which the card between the eyes would present no embarrassment. The eyes being restricted in vision, being set free from the contradiction of surrounding facts, each takes up its own representation in a sort of double PERCEPTION. 107 monocular action, and transfers it, with corresponding in- crease of dimensions, to the distance implied in the picture itself. It can then unite with the impression furnished by the other eye, and the two blend into one view. The mo- nocular character of the effort is plain from the fact that, when we fail to harmonize the two images, vision through one eye still produces the desired illusion. Still farther, the axes of vision in the eyes are made less convergent, as when directed to distant objects. To aid the construction, the two images are taken from slightly removed positions, thereby, in reference to the foreground, giving the same readjustment of objects in position as that which belongs to double vision. This advantage is, however, immaterial, except in connection with near objects. The lenses or mirrors introduced into the stereoscope do not alter the principle ; they still leave the eyes to do independently their constructive work, and to identify the images they have removed into the distance. § 8. Perception is a peculiarly interesting portion of psychology. Lying at the commencement of the study, it imposes upon us at once a most difficult case of analysis ; the results we reach go far to settle the relations of mind and matter to each other ; we have need to determine the full circle of mental powers and put them all in operation in laying these foundations of certain knowledge ; the auto- matic action of the mind, and its successive stages of growth require immediate recognition ; while historically many questions of philosophy have been settled at this point. Perception, in its broad meaning, includes the physical conditions furnished by the senses, with their nervous con- nections; the states or actions of mind directly incident thereto; the nature of the dependence of these states on those conditions ; the extent to which the first activity of mind is enlarged by experience ; and the character and the 108 INTELLECT. certainly of the knowledge both of the internal and the external world that accompanies perception. We will rap- idly review each of these points in their theoretical and historical bearings. The physical conditions of sight, hearing, touch, taste, and smell are matters purely of physical inquiry. A satis- factory knowledge of the intervening means of perception is comparatively recent, as is all exact anatomical science. An inquiry into the physical incidents of perception has served to displace some crude theories of its method, and to make way for a more careful separation of the material and intellectual elements in the process, or, in the minds of some, for their more complete identification. The notion of " images," " species," " representative ideas," which me- diate between the object and the mind, arose out of the ignorance of the organs of sense, and lingered in philosophy as late as the time of Locke. These images or ideas had one or another degree of materiality according to the age, the philosophy, the person, who dealt with them. Earlier, they were a physical film passing off from the object ; later, as the immateriality of the mind gained ground, they shared more of its nature ; but at no time could they per- form their office. To whichever extreme they moved, they thereby lost power to touch the other. If spiritual, they were out of relation with matter, if material, they were disassociated with mind. A knowledge of the eye, the ear, the brain sweeps away this intermediate mechanism, estab- lishes the complete physical character of the organic por- tion of perception, the complete spiritual nature of the mental portion, and leaves the interaction of the two an ultimate fact. Descartes sharply distinguished between mind and mat- ter, and so relegated physical inquiry to its own physical field, and mental inquiry to consciousness. All media PERCEPTION. 109 sharing the two natures disappeared ; the separation of the two fields, thongh often carried to a mistaken extent, be- came decisive. Thus the way was prepared for sound in- vestigation in both directions. The second point is the states of mind incident to the organic action of the senses. The content of the sense in each instance, with its definite, discriminated qualities, in some way controls the mind, and the mind in active, con- structive fashion responds to this external efficiency. If any physical force acts upon a body in reference to it inert, it is yet true that the result will turn quite as much on the qualities of the so-called passive agent as on those of the aggressive one. Every agent, physical and mental, ceases to be passive under action, and blends in the results its own activities with those of the efficient cause. We may forget this in mechanical facts, we can scarcely forget it in chemi- cal, vital, mental facts. The mind is barely less active in a sensation than in a perception, in a feeling, than in a thought. Each is determined as a conscious state by a sus- ceptibility called out, or a power occupied. As regards the fact, of activity, the two classes are not separable, though the activity in perception and thought is more voluntary, more modifiable than that in sensation and feeling. The mind, as possessed of a definite constitution, determines the nature and efficiency of each of its states. If the mind is restricted to organic conditions for particular sensations, so are these conditions, in turn, restricted to answering suscep- tibilities before they pass into mental facts. Nothing but a living agent can feel. When oxygen and hydrogen com- bine, there are conjoint activities, conjoint properties, con- joint products. The activity of the mind in sensation is shown not merely by the fact that sensibility is activity, but by the fact that the mind is ever interpreting and attributing its 110 INTELLECT. sensations. In each sense the whole complex nervous mechanism, including the external organ, the connecting nerves and the brain, lies between the mind and the ob- ject, as the telescope between the eye and the stars, the whispering-gallery or the telephone between the ear and the speaker. Yet it is not of one or all of these conditions that the mind is cognizant. It sees through them, hears through them, feels, tastes, and smells through them. It carries the facts of vision outward to their remote objects, and the qualities of touch or flavor to their sources. The mind is always taking up and using instrumentally its or- gans, and is no more subject to mere effects in them, than is the eye in using a microscope cognizant of its images and lenses. Intermediate conditions lie as submerged links in the mind's activity. ]STo image could well be more inapt than that of Locke in which he compared the mind to white paper. As all knowledge in chemistry up to the present time has been accumulating the conditions by which the plate of the photographer is made sensitive to light, so all evolution has joined in shaping that latest power by which mind feels. Mind is the one only agency which can so respond to the physical world. When we add perception to sensation, the marvel grows. All senses end in molecular changes. From these common terms, so like one to another, mind constructs, in the hidden joy of vision, and of every acces- sory sense, this universe about us, with its immeasurable spaces, its brilliant colors, its throbbing sounds, its sweet odors, its stimulating flavors, its velvety surfaces, and its thousand suggestions of life. If there is any fit image of omnipotence, it is this mind in the midst of the world which it momentarily evokes. The third fact is the nature of the interaction between the last physical facts and the first mental ones. Men have PERCEPTION. Ill struggled long and vainly with this ultimate truth, that matter affects mind and mind affects matter. They have striven to insert midway terms ; they have brought down mental to physical facts, and there identified the two ; they have reversed the process, and regarded perception and sensation as purely mental processes. It is well to stand with Descartes, and assert the radical division between the two sets of phenomena, so radical that facts of the one class cannot be intelligibly expressed in words of the other class. It is well to reject all explanations that explain nothing ; to make no assertions of the possible and impossi- ble which transcend experience, and imply on our part a knowledge of ultimate relations that violates their nature. It is well to stand quietly by ultimate facts, putting upon them no constructions which we cannot verify. Physical facts can be expounded under their own forms, mental facts under their forms ; while their interaction, to us at least unphenomenal, is without form. The method in which a specific, organic state is transformed into a sensation, or in which a volition in turn is converted into an action, is be- yond the terms of experience. So, indeed, is all transfer of forces in physics. The fourth point is the extent to which the primitive activity of the mind is enlarged by experience. The first full discussion of this topic we owe to Bishop Berkeley. The importance attached to this secondary element in per- ception, these inclosed judgments, has been on the increase since his time. The organic content, or rather the activity of mind incident to this content, is the dry sponge which absorbs the inferences of experience, is expanded and made flexible and serviceable by them. There is a tendency to refer these judgments in the man as in the animal largely to instinctive action and to inheritance. As the infant and the child are manifestly for a long period employed in form- 112 INTELLECT. ing these judgments, and as the process accompanies us all the way through life, we see but little occasion to ascribe their presence to a blind, organic tendency. Moreover, such a movement would not prepare the way for conscious, rational action, would not put the mind in possession of itself, but would tend in action to anticipate order, and so prevent such a result. The fifth consideration is the nature and certainty of our knowledge of the internal and external world. The mind under any view has the most immediate knowledge of its own impressions. The interior phenomenal world is necessarily known to it. Here knowledge gathers its pri- mary force. The chief difficulty is found in the relation of this knowledge to real being, internal or external. The dependence seems to us simple. We believe, with Des- cartes, that under the action of causation we infer decisively and correctly external agents from fixed, involuntary im- pressions. The reality and relations of these external causes are more and more disclosed to us by perception with its enlarging judgments, and are more and more com- pletely clothed with the phenomenal impressions in the mind which are attributed to them. This knowledge of causes in their effects is all the knowledge that is proper to them, and the effort to resolve the cause itself into a second phenomenon, serves only to push the cause one step farther back. In repeated instances this process, by which the mind habilitates the underlying reality with its appropriate phenomena, has, in the case of vision gained in later years, passed on under observation. There is no element of real doubt in this knowledge ; its links are close and sufficient ; its chief mysteries lie, as is wont to be the case, in first terms, in the perceptive and sensational elements, and in the completeness with which these are made objective. This last point is abundantly illustrated by facts like those of PERCEPTION. 113 sketching and painting. A few lines, as of a human face, on a plain surface, give us at once, under our construc- tive powers, form, substance, distance, character ; objective throughout and thoroughly realized. Yet around this relation of the mental impressions to the underlying facts, most of the divisions and denials of philosophy have sprung up, chiefly because the intuitive presence and the validity of the notion of causation have been overlooked, the firm yet inscrutable line between physical and mental facts been lost, and kinds of knowledge impossible from the nature of the case been sought after. If we accept these data, the separate, unmistakable charac- ter of mental phenomena, the soundness of our intuitions, and the distinct, incomparable forms of knowledge, there is very little ground for discussion. What the mind directly knows must be purely mental, for a fact becomes mental, is mental, by virtue of being found in consciousness. What it indirectly knows are the phenomena of space, and those of other minds, both interpreted by its own experience; and those permanent, efficient powers which underlie phe- nomena, known only as forces or causes, and not at all as appearances or effects. Facts that are placed in any other region than space and consciousness, or are to be known in any other way than directly as phenomena or indirectly as causes, are hopelessly unknowable, are so far chaotic as to lack any formative idea to define them, any condition of thought under which to appear. The impressions in the mind cannot be mistaken be- cause they are pervaded by consciousness ; the underlying and the outside facts which they disclose cannot be vision- ary, for all the intuitions and judgments of the mind vouch for them. The external world is known as the certain, cause of the fixed impressions which shadow it forth in the mind, and this knowledge is the exact equivalent of that U4: INTELLECT. which we have of ourselves as real persons. We allow no pre-eminence of one branch of knowledge over another; the perception and the inferences are equally decisive in reference to that which they disclose, and are inextricably blended. From confusion at this point has arisen the affirmation that all our knowledge is relative. By this is meant, that there is ever present an unmeasured, subjective element which separates the convictions of each person from those of his neighbor, and so from absolute truth. This assertion has meaning and force in reference to sensations, less in ref- erence to perceptions, and none in reference to intuitions, and the conclusions drawn from them. The sensation is more completely, the perception less completely, involved in a physical, organic state ; and all we can say concern- ing this organic element is, that it, under given conditions, yields in the mind certain impressions. The identity of the action of organs in different persons we cannot affirm, nor what forces are aside from the organs they affect. This relative knowledge, however, is real, and sufficient for all its purposes. We have no occasion to know matter save as the fixed causes of certain effects, and that its effects in us are kindred to those in others. To inquire whether matter is like its effects, or what it is aside from its ef- fects, are questions out of order under our organic intel- lectual law. The physical element makes the knowledge of the senses relative without affecting its value. But the intuitions with their inferences have no conditional, physi- cal element. The knowledge of relations is pure knowl- edge, identical knowledge, in all. This is obvious in mathematics, in logic, in all pure science. The mind is as capable of absolute as of relative knowledge, of a move- ment that goes out from itself in general principles, as of one that comes into it as specific facts. Its relative knowl- RELATIVE KNOWLEDGE. 115 edge is of substances in their properties, those interactions by which they define each other ; its absolute knowledge is one of forms, of regulative principles, which neither mat- ter nor mind can escape. This fundamental difference be- tween intuitive and sensational knowledge, no analysis has been able to break down or obscure. Truths of the higher mathematics rise quite above experience, and come at once with irresistible authority to it. It is strange that Hamil- ton should give us a direct knowledge of matter, and yet regard all knowledge as relative. This assertion, that all our knowledge is relative, has gained great currency ; yet we can look on it only as one of those false deductions which have followed the empiri- cal philosophy, and are fitted to sustain it. If all our knowledge comes from experience, it is all tainted by the quality of the senses, and hence is all relative to their forms and powers of discrimination. Pure knowing, pure in its object, and pure in its subjective process, becomes impossi- ble. Yet direct insight into abstract relations is evidently of this character, and evinces its nature by the uniformity and force of the conclusions incident to it. Empirical philosophy has no way of explaining axiomatic truth and demonstrative reasoning. The reference of the necessity of these convictions to inheritance is most lame. They pertain to abstract truths often very unfamiliar, and take effect at once in connection with quite novel statements. Matters of experience under the daily observation of many generations — as the rising of the sun, the blackness of crows, the whiteness of swans, carry with them by inheri- tance or otherwise no such necessity. That the truth of mathematics can be no other than they are, is a conviction that the mind takes with it to the study of nature, not one that it derives from nature. Of all branches of knowledge mathematics may advance most rapidly, and divorce itself 116 INTELLECT. most completely from previous experience. This power of the mind to deal with abstract relations, and push them entirely in advance of physical inquiry, belongs to it as possessed of pure insight. The abstractions of sensation, as greenness, sweetness, hardness, are accompanied by no cor- responding sweep of thought, and give no particular mas- tery. The abstractions of experience can carry no more force than the experience itself. The complete identity of mathematical truths in differ- ent minds, and this from their very first announcement, is a sufficient proof of their absolute character. They are plainly not modified by any special qualities of special forms of intelligence; and they stand, in this particular, in contrast with all empirical knowledge. § 9. A farther error connected with the doctrine of direct perception is the division of the qualities of matter into primary and secondary. The list of primary qualities is differently made out by different philosophers. Exten- sion and solidity are generally recognized as chief among them. The criteria of these qualities as compared with secondary qualities are, that matter can not exist without them, and, especially urged by Hamilton, that in the pri- mary qualities perception is peculiarly direct and clear ; " the objective element predominates," " matter is known as principal in its relation to mind." The distinction be- tween primary and secondary qualities seems on these tests to be untenable, and to arise from an oversight of those necessary intuitive ideas involved in the very existence of matter. Extension should not be regarded as a property of mat- ter, and if so regarded is not, in the form in which it exists, a necessary property. ]STo portion of matter is necessarily of one size rather than another. The actual quality of ex- tension, if it is to be so termed, is as variable as any other PRIMARY QUALITIES. 117 quality. The only universality in this attribute more than in other attributes of matter is found in the fact, that all matter must exist in space, and hence under the one form of extension. Space, extension, is a necessary condition of matter. Without it, those qualities, properly so called, which constitute matter, cannot have a being. It is in- volved in their manifestations, that they occupy some por- tion of space, and this primary quality, so called, is only this essential condition for the existence of matter. We might as well say that duration is a quality of matter, as to say that extension is such a quality ; since no form of matter can exist without occupying some period of time, more or less. ]STor is the second criterion any more satisfactory in its application. If there is any one direction in which the mind acts with a sense of establishing and defining its own data, it is this of extension. Odor, taste, color, are what they are, directly, through the nature of the outside cause ; but the form of a body is arrived at through meagre grounds of judgment unfolded by the enlargement and corrections of protracted experience ; while the notion under which alone this evolution can proceed, that of space, is furnished entirely by the mind. Perception, instead of being un- usually direct and immediate in extension, is more than elsewhere indirect and enlarged by inference. Solidity, as a primary quality, is open to a like form of criticism. That which must in this discussion be under- stood by solidity, is very different from the notion which the word ordinarily conveys ; it is the impossibility of com- plete compression, complete displacement. A gas is in this sense as much a solid as a piece of steel, since, when prop- erly confined in a cylinder, it is found to exclude the piston as certainly as the most solid substance. Compression can- not proceed to all lengths. Resistance accompanies pres- 118 INTELLECT. sure, and an increased and insuperable resistance remains as the final result. Without this capability of occupying space to some sense we withhold the appellation of matter. Here, again, any given degree of incompressibility is not necessary to matter, but only that there should be some degree of it, and some degree of incompressibility is neces- sary as involved in this method of occupying space. The general necessity, then, is evolved from this general condi- tion of the existence and recognition of matter, that it shall be a space-filling force, that it shall have, in the external world, a permanent substratum to its phenomena. As the necessary connection of extension with matter arises from the idea of space, so that of solidity arises from the occupation of space, the ideal of a local, fixed cause, the source of fixed phenomena. That the forces which lie at the basis of matter may in some cases penetrate each other, as in the union of two gases, and may in others entirely ex- clude each other, as in the contact of solids, are facts to be learned by experience. The very notion of matter, how- ever, is that it involves a local cause or force, and if a cause or force, that it has some means of showing itself as a force. If it be said that the distinction between primary and secondary qualities is valid, since solidity necessarily in- volves force, the substratum of matter ; answer is made that no specific form of force is necessary to matter, but only some form of force, and that this is as necessary to color, to flavor, to odor, when these are present, as to solidity when this is shown. Solidity has no permanent existence any more than odor or color, demands like them for its manifes- tation appropriate conditions, and does no more than they do, in demanding as a condition an external force. If, then, we speak of the effects matter is capable of producing as the qualities of that matter, odor, resistance are such quali- ties, but neither of them are constant ; both are occasional, PRIMARY QUALITIES. 119 and conditioned to fitting circumstances ; both of them im- ply that which is permanent and necessary. From taste as from touch, we may infer an external, local cause, a cause that must hence occupy space, be a space-filling force, at least to one sense, which is the entire conclusion derivable from resistance. If it be said that the circumstances are in all cases pos- sible under which the quality of solidity may be drawn out, while those which disclose odor are peculiar to a few bodies, we answer, this is a question of experience, is far from being proved, as in the case of ether, and, if estab- lished, could bring with it no sense of necessity, differenc- ing the two cases. Bodies which yield no odor under one form, may under another. Odor seems to involve chemi- cal change, and it might be found that every substance would yield it under fitting chemical conditions. This is a question to be decided by protracted and varied experience, and however decided, could only be the ground of an em- pirical, and not of a necessary division of qualities. Take such a secondary quality as that of color. It seems antecedently probable, that all bodies have color. Some gases are apparently colorless, but so is the atmos- phere in small volumes. Experience and theory would lead us to expect that the most diffused force in sufficient vol- ume would affect the transmission of light. These criteria, then, cannot be the sufficient and pre- vailing ground of this distinction, but we must look farther for something thought to inhere in the very nature of mat- ter, necessary to it and betrayed by every primary quality. This necessary something we accept, and believe the notion of it to arise under the intuitive idea of cause and effect ; but also believe, that this notion is called forth, as certainly by odor, by color, by taste, as by pressure ; and that we in- evitably put back of each of these subjective effects, a per 120 INTELLECT. manent force, occupying space, called matter. Tins per- manent force is necessary to the notion of matter, and is as appropriately reached by one sense as by another, by one effect as by another ; indeed, is indicated by any sensation which betrays an external world. The qualities which find entrance through one organ, have no more right to be called primary, that is fundamental, than those which enter at another. If the sense of muscular effort were wanting, we might still be able to arrive at the idea of matter ; though its alleged primary quality should not be directly recogniz- able by us. We should then understand color and flavor as indications of a local force, apprehensible by sight and taste. The second criterion more signally fails than the first in its application to solidity. Far from matter's coming most directly and fully in contact with mind through solidity, in many instances it is only in a secondary, inferential way, that this quality is at all arrived at. A gas makes no im- pression on the muscular system, offers no obstacle to move- ment, calls forth no sense of resistance, till closely confined ; and then by that very confinement is put beyond direct contact with any organ of sense. We are left to infer the solidity of gases, from the fact that the piston cannot, in the cylinder containing them, be forced perfectly down to its bed, and recoils as the hand is lifted. Perception is not more immediate and full here than elsewhere ; on the con- trary, there is no perception of the point at issue, the solid- ity of the gas, but only a judgment to that effect. Even the solidity of a solid directly handled is inferred from the muscular effort expended in the attempt to crush it, and only admits of an estimate by an indirect method. This doctrine of primary and secondary qualities, main- tained by so large a variety of philosophers, is of interest, chiefly from the way in which it has grown out of the PRIMARY QUALITIES. 121 errors, or betrays the errors, held by them ; and yet more from the indication it gives of an unconscious influence of truths not formally recognized. Thus, Locke speaks of the " inseparable " nature of extension as a quality of matter, while declining to accept the antecedent necessity of space as a condition of matter, and a knowledge of matter. Here- in he grants to matter the necessity which he has denied to mind ; whereas by necessity can only be meant something which the mind inevitably affirms, a union of things which it sees to be indissoluble. No matter how often things are practically connected, unless the mind can so far penetrate that connection as to see the one to be involved in the other, their dependence would not seem to be a necessary one. Yet this father of sensationalism speaks of insepara- ble qualities, when experience in many cases had neither by sight nor in any way tested their existence. Why this judgment of universal, of necessary extension and solidity ? Because of a conviction latent in the mind through its in- tuitive ideas, a conviction independent of the complete ex- pansion of experience. Hamilton, again, looking at this division of qualities through the doctrine of direct perception, jumps at the con- clusion that primary qualities are those more immediately revealed, whereas inquiry shows that solidity, the most un- deniable of them, is often wholly unapproachable to any form of direct perception, and is arrived at by reasonings from sensations which arise indirectly from the object of experiment. The staff so quickly clutched at has become a broken reed. Thus philosophers furnish undesigned and most valuable proof to an adverse theory, by recognizing and striving to use in a disguised form the truths which it proclaims, and assigns their true position. The acknowl- edged necessity of primary qualities is not in them, but in that intuitive action of the mind which they call forth. 122 IKTELLEQT. Necessity, which every philosopher seems ready to intro- duce at some point, is born not of experience, but of men's thoughts ; not of matter, but of mind. We briefly sum up the conclusions arrived at. Exten- sion is not a quality of matter, but its antecedent condition, and owes the sense of necessity that accompanies it to the necessary idea of space. The same reason that makes it, would make duration also, a quality of matter. The actual form or extension of bodies is contingent and inferential. Solidity, or the power of exclusion, is a quality of matter, and owes its necessity to our idea of the nature of matter, an idea arising under the notions of space and of cause. It differs not from odor, color in implying a permanent sub- stratum. Every quality of matter, every sensation and perception involve this, though they mutually deepen and confirm it. Solidity, a sense of resistance, is felt to be more necessarily involved in matter than odor and taste, because, in experience, we almost exclusively use this con- stant and convenient test of its presence. This, however, is an empirical distinction, arising from the nature of one senses, of an uncertain character, and of no particular im- portance. The distinction then of primary qualities, while involving important points in philosophy, in its common form breaks down. These qualities are not more directly perceived than other qualities; they are not in contrast with them nor known to be more necessary. If we reason from the quality to the substratum, each implies this, and the necessity is common and complete. If we reason from the substratum to its qualities, no individual quality is seen to be necessary, neither any kind nor class of qualities as hardness, color. We can not so penetrate the nature of the cause as to antecedently declare what its action will be. The greater constancy of one quality over another is learned by experience ; the intrinsic necessity of that constancy, if CONSCIOUSNESS. 123 there be any, is unperceivcd. A local substance perfectly penetrable, yet having odor, color, and flavor, would doubt- less be regarded by us as matter. § 10. Consciousness, or the inner sense, the remaining means of a direct knowledge of phenomena, requires but brief notice. Our chief difficulty in conceiving this source of knowledge, and in speaking of it, is found in the lan- guage we are compelled to employ, and the confusion already occasioned by it. Self-consciousness, or conscious- ness, or the inner sense, is not a method of the mind's ac- tion, is not a faculty of perception. These words are used by us simply to express the fact that the mind knows what it does; that its states, acts, experiences, are necessarily open to itself, not by any direct effort of attention on its part, but by virtue of the very fact that they are its own states. We cannot readily speak of this knowledge which the mind has of its own phases of activity without seeming to imply more than we intend ; to imply an explicit form, or faculty, or means of knowing. What we wish to draw attention to, as a second source of phenomenal matter, is the familiarity of the mind with its own thoughts, feelings, volitions ; and hence its power through memory to make them objects of attention, analysis, inquiry. By these pro- cesses primarily is philosophy established, the phenomena of mind separated into their elements, and the laws of their combination discovered. Consciousness furnishes only the bare data of mental facts, the perceptions or thoughts present, and is not in the least responsible for their accu- racy. Its verity is only involved in rendering them as they are, that is, as they lie in the mind. Whether we perceive what we think we perceive, whether we know what we think we know, that is, the objective justness of our mental action, these are quite different inquiries. The subjective state is all that is revealed in consciousness, and this is 124 INTELLECT. revealed by the very nature of mind. Concerning it, there is no opportunity for scepticism in the very moment of its transpiring ; later the question is one of memory. Consciousness gives internal phenomena, perception gives external phenomena. Did not perception constantly involve inference, did it not reach the external world through the reflective force of the mind, perception and consciousness would give but one and the same set of data, and the distinction would disappear. CHAPTEE III. The Understanding. § 1. The understanding includes all those mental activ- ities by which the data of sense and reason are wrought into knowledge. They are memory, imagination, and judg- ment. The first condition of rational activity is percep- tion, some object given to the mind towards which it may be moved, with which it may occupy itself. The second essential condition is memory, by which perceptions gain continuity, are united into one experience, are made ready to be woven into the fabric of belief. "Without memory our conscious states would be separate, incommunicable, save by direct sequence, with no more reciprocal play, unity, and growth than belong to particles of sand. Memory is involved in the coherence of intellectual life, as much as the constant interaction of its organs is included in physi- cal life. Memory is the power of recalling the phenomena of consciousness. The experiences of the past are restored to the mind by this faculty, with a recognition of their previous existence. Like all primitive powers, it has its own simple, unique action, explained only by experience. The words retaining, recalling, may, through the force they have acquired in physical connections, suggest the idea that some impression of the objects remembered is held in the mind, and again restored to its observation ; or that some trace or result of the first act remains with the mind, waiting renewal in memory. Indeed, looking more at the material suggestions of mental phenomena, than at 126 UNDERSTANDING. the simple, primitive character of the act of recollection itself, some have inquired, Whether the very thing first known is the object of memory, or whether the mind is occupied with some image of it ? We might as well in- quire Whether the artist's conception of a painting is the very painting itself, or an image of it ? It is certainly not the first, nor even the second in any other than a figurative sense. When I say that I recollect an event, my language is about as intelligible as it can be made. There is in it a direct appeal to the interpretation of every one's expe- rience, furnishing like simple, original acts. In memory a new impression of the event is present, accompanied with a knowledge of its previous presence. It is merely a futile struggle with physical images, the misleading effect of physical analogies, which prompt us to inquire, with an analysis more cunning than cognizant of the true condi- tions of mental experience, Whether, as our language seems to imply, we actually remember the very object that has passed away, or whether some impression of it is restored to us ? Each act of memory is a primitive act, efficient in itself for its own independent and peculiar end ; is, more- over, purely subjective, though often involving a knowl- edge of the objective. Memory is not a repeated experi- ence ; it is the cognizance of a previous experience without repetition. The renewal of awakened action in the brain, if it could be shown to accompany recollection, would be no explanation of it. Of a like character are all the expla- nations of memory, which spring from purely physiologi- cal facts. Whatever may be the effect of thinking on the brain, the connection of these physical changes in a physi- cal agent with the act of memory is wholly unintelligible. I might as well explain the recollection of a sword- wound by the presence of a scar on the body, as by any changes effected at the time in the brain by the suffering then ex- L MEMORY. 127 perienced. That a scar constitutes memory, is as apprehen- sible as that a modification of a nervous tissue is memory. It is a fact, that memory, like other intellectual powers, is dependent for its exercise on the conditions of the brain, but why, or how dependent, are queries beyond the circle of knowledge. The vital play of nervous changes along nervous lines is one thing, the action of the mind a totally different thing. The one is learned as an outside fact by outside observations, the other as an inside fact by con- sciousness. The synchronism of the two is an interesting point, but one, for the present, barren in strict philosophy. That memory is more dependent than other mental powers on physical states is generally believed, though we may be easily deceived in the grounds of this judgment. Memory is readily and quickly tested in its strength. A straightforward, categorical question betrays at once its weakness. "We observe, therefore, failure at this point, more certainly than at others. In moments of weariness the memory fails us, but so, evidently, does the judgment. Obstacles seem disproportionately great, the occasions of fear unusual and pressing. In old age, memory is said to be the first faculty that shows decay ; yet the old man, with- drawn from active life, naturally first discovers his failure here. It requires occasions of judgment to disclose the deficiency of judgment to others, while to ourselves, these failures are not betrayed from the very fact that the judg- ment, as weak, does not fully detect its own weakness. On the other hand, a dozen events every day expose inevitably and unmistakably the defects of memory. Moreover, the things chiefly forgotten are those of recent occurrence, a fact accounted for by the want of strong feeling, clear perception, and energetic attention. Diseases that weaken the memory by the destruction of brain-tissue, are especially unfavorable to the recollection of events that occurred in the periods im- 128 UNDERSTANDING. mediately previous to the sickness. Remote events may be retained with distinctness, while those of intervening years are wiped away. When memory is restored, its power re- turns in the reverse order. The more remote events are first recalled. A simple waste of tissue would seem to be liable to interfere with one set of relations as quickly as with another. The law of mind, however, that the power to retain impressions is proportioned to their first strength and to their reiteration seems to hold in this method of res- toration. These facts go to show that physiology is not prepared, I will not say to offer an explanation of the phe- nomena of memory, but even to point out with certainty and fulness the changes in the brain coincident with the changes of this power. A general dependence of all our powers on the vigor of this, their common instrument, is the brief summation of its knowledge. Language like the fol- lowing, conveys no intelligible idea : " All that has so far been said respecting the different nervous centres of the body cannot fail to demonstrate the existence of memory in the nervous cells which lie scattered in the heart, in the intestinal walls, in those that are collected together in the spinal cord, in the cells of the sensory and motor ganglia, and in the ideational cells of the cortical layers of the cere- bral hemispheres." — Maudsletfs Physiology and Pathology of the Mind, p. 182. What a famous stroke of explication — "ideational cells ! " What a liberal distribution of recollection from the sole of one's feet to the crown of his head ! Surely forgetfulness is inexcusable under such endowments. § 2. There are other theories of memory not so crude as these physiological ones, yet as deficient in proof, and rest- ing back almost equally though somewhat more subtly on physical analogies. Of this character is that one elaborately and repeatedly enforced by Hamilton. He affirms " that MEMORY. 129 an energy of mind being once determined, it is natural that it should persist, until again annihilated by other causes. This in fact would be the case were the mind merely pas- sive in the impression it receives ; for it is a universal law of nature, that every effect endures as long as it is not modified or opposed by any other effect. But the mental activity, the act of knowledge of which I now speak, is more than this ; it is an energy of the self -active power of a subject, one and indivisible ; consequently a part of the ego must be detached or annihilated, if a cognition once existive be again extinguished. Hence it is, that the prob- lem most difficult of solution is not how a mental activity endures, but how it ever vanishes." Is not this notion of the necessary persistence of force referable exclusively to physical forces ? What is the proof of its applicability to mental action ? The facts of mind inquired into on their own basis, seem to indicate quite the opposite conclusion. He proceeds : " If it be impossible that an energy of mind that has once been, should be abolished without a lacera- tion of the vital unity of the mind, one and indivisible, — on this supposition, the question arises, How can the facts of our self -consciousness be brought to harmonize with this statement, seeing that consciousness proves to us that cog- nitions once clear and vivid are forgotten ? The solution of this problem is to be sought for in the theory of ob- scure or latent modifications. The disappearance of inter- nal energies from the view of internal perception does not warrant the conclusion that they no longer exist." " All the cognitions which we possess, or have possessed, still remain to us — the whole complement of all our knowledge still lies in our memory; but as new cognitions are con- tinually pressing in upon the old, and continually taking place along with them among the modifications of the ego ; the old cognitions, unless from time to time refreshed and, 130 UNDERSTANDING. brought forward, are driven back, and become gradually fainter and more obscure. The mind is only capable at any one moment of exerting a certain quantity or degree of force. This quantity must, therefore, be divided among the different activities, so that each has only a part, and the sum of force belonging to all the several activities taken together, is equal to the quantity or degree of force belong- ing to the vital activity of mind in general. This obscu- ration can be conceived in every infinite degree, between incipient latescence and irrecoverable latency. The ob- scure cognition may exist simply out of consciousness, so that it can be recalled by a common act of reminiscence. Again, it may be impossible to recover it by an act of vol- untary recollection, but some association may revivify it enough to make it flash after a long oblivion into conscious- ness. Further, it may be obscured so far that it can only be resuscitated by some morbid affection of the system; or finally, it may be absolutely lost for us in this life, and destined only for our reminiscence in the life to come." The view, whose salient points with large omissions are here indicated, is analogical, is purely theoretical, is beset with internal difficulties, and is unable to explain the phe- nomena that call it forth. (1) It is analogical. It assumes that the laws of physical forces pervade mental facts. It does not even stop to inquire how far or with what modifi- cations this diffusion of forces takes place in the brain. It accepts at once the principles of physics, as applicable in the boldest way, to processes of mind. (2) It is purely theoretical, for its alleged facts all lie in the unapproachable region of sub-consciousness, whose existence is not estab- lished, much less the details of its phenomena. (3) It is vexed with difficulties of its own, greater than the difficul- ties it is brought forward to remove. It rests on a physical idea of force, but cannot consistently carry out that idea. If MEMORY. 131 no force, no activity can be lost, how shall an act of mind fade out of consciousness ? What is this fading away, if it be not a loss of force ? Or, again, if the mind have but a given amount of force to bestow, and each act takes a por- tion, how long will it be before its stock of power will be exhausted ? Or, if this power is divided up into a multi- plicity of acts, and previous acts, thsref ore, are weakened in their impressions, does not this imply a withdrawal in part of activity from earlier actions, and if a partial with- drawal is possible, what renders complete removal impossi- ble ? Again, what is meant by recalling an obscure cogni- tion ? Is it simply infusing more power into it, deepening the action already present, or, is it a new act of mind by which we direct attention to it, and bring it to the light ? Must this new act also, in turn, subsist forever, still farther sub-dividing the power of the mind ? These and many like questions are pertinent to this semi-physical theory, and show it to be unintelligible, not to say preposterous. It has no coherence and completeness in itself. (4) Nor does it explain the difficulties which the facts of memory present, and which call it forth. Indeed, these phenomena are every way more comprehensible than the solution of them thus offered. The act of recollection still remains, and is cer- tainly no more intelligible because we suppose somewhere, in some out-of -sight region of the mind, is lurking a previous act, which this new one fastens upon and brings forward. What relation do these distinct co-existing acts, the recalling and the recalled, the captor and the captive, bear to each other ? How do they together constitute memory ? Recol- lection seems to be as single, simple an effort of mind, as perception or thought in the first instance. There is no oc- casion, because memory is an act of recollection, to put either in the mind or out of the mind, in an independent self- exist- ent form, the exact thing recollected. A dead man can be 132 UNDERSTANDING. remembered as easily as a living one, a defunct thought as readily as one that has not passed away. Indeed, we do not see why any other needs to be recalled. So far as the act has not passed from consciousness, it calls for no recollec- tion ; so far as it has, it is lost to the mind, and the power to restore it involves the whole mystery. These words, re- store, recall, resuscitate, are not to be allowed to mislead us by their physical imagery. The state recalled exists anew in the primitive, simple, inexplicable act of memory ; a movement of mind as much of its own kind, and with its own force, as the first act of perception ; and as indepen- dent, save that the occasion for it is found in the existence of previous states of consciousness. If acts of mind could be shown to the fire-flies passing from light into darkness, and darkness into light, with patient and inexhaustible al- ternations, it might be to the purpose ; but if there must still be a distinct act of recollection, either to go in search of other acts and restore them, or when they are present to remind us of their previous presence, such an act involves the entire difficulty, and to be really anything, it must be a fresh handling of an old topic, differing from the first in that the mind knows it to be a second state of conscious- ness, and subject to the conditions of such a state. To re- experience sensations and recollect them, are quite different things. Much is written concerning the last, which at most would be applicable to the first only. Under this philoso- phy, we should be able neither to distinguish between (1) the lingering of perception in an irritable organ and mem- ory ; nor (2) between memory and repeated perception. A peculiar, primitive power is present in memory as in every other act of mind, and as a simple act, it admits and calls for no explanation. To foist on such states of conscious- ness, ultimate and complete in themselves, conjectural an- alogical explanations, is to make the simple and plain, com- ASSOCIATION. 133 plex and obscure, is to darken counsel with words. If we would explain memory by its own facts, it would be our true empiricism. To create difficulties by the introduction of physical analogies into a field alien to them, and to seek their explication by a farther importation of imaginery states is a palpable violation of the principle of original, simple induction in each department of inquiry. It is a most vicious a priori method, disguised under the form of empirical investigation. § 3. We need to distinguish memory from certain things with which it is in result allied. Association may restore facts to the mind with no direct effort of recollec- tion, indeed, in hours of idle revery, with scarcely a distinct observation of their previous presence. This indolent flow of thought, mingling past, present, future, blending the real and the fanciful, submitting itself to the native cohesion of events and desires, is remembering, precisely as it is thinking. It is neither the one nor the other, consecu- tively, tensely, clearly ; but is merely a succession of mental movements, holding on to each other, under a feeble im- pulse of pleasure, by accidental connections of thought, memory, fancy, acting the part of nimble servitors in this feast and repose of the desires. Association in large part rests upon memory, yet this easy natural movement of the mind, along certain trails of imagery which have been established by previous experience, serves to disguise the action of memory which underlies it. A certain sequence of impressions may be the result of many previous ex- amples, yet directly recall no one of them, when, in the lazy flow of thought, the mind passes this way, using once more groups of conceptions which the entire past life has been combining. Much therefore rests upon memory in which its action is so far from being prominent, that its presence is hardly discerned. Much is thought to be origi- 13 ^ UNDERSTANDING. nal which is not so, because the memory has restored it stripped of the time, place and circumstances of its acquis- ition. Habit, a permanent union by repetition of certain states with certain acts, often closely unites itself with memory. Words which have been very frequently uttered in a fixed order can be repeated with a rapidity and slightness of attention which hide the act of memory. We are said to recite them by rote. There is here doubtless muscular and nervous training as well as recollection. The facility gained in any lengthy process by repetition is of this double character. The memory itself, however, seems in most cases to require the lapse of a certain time, and a certain frequency of recurrence, to make its action rapid and spontaneous. We readily repeat in the morning what was recited with difficulty the evening before, and few can acquire a literary composition for easy, accurate re- hearsal in the period immediately preceding its delivery. The same effort, scattered through several days, is far more effectual. The growth of the mind is also to be distinguished in its effects from the action of memory. Mental phenomena are so blended that the predominant is by no means the exclusive element. Later movements of mind are not mere counterparts of earlier ones. A better grasp of premises, and more insight into them; conclusions more complete and decided, belong to the thinking powers, as they are strengthened and enlarged by use. This fact of growth is an ultimate one. We know it, and through familiarity it seems simple to us, without our understanding its grounds. It is something more than memory. We are not merely wiser, with more acquired knowledge ; we are stronger, able to make an increasingly effective use of what we know. Memory and growth are very closely related. The MEMORY. 135 accumulated stores of the mind are the condition of its ex- panded action, and this increased action gives new signifi- cance to its acquisitions. It is not easy to say how far present soundness and shrewdness of judgment are the product of increased strength, and how far of increased knowledge. Our reasoning powers, by easily evolving con- clusions from premises, by renewing, rather than by recall- ing previous processes of thought, may closely resemble the memory in their action. We may seem to recollect an argument, to remember a proposition, when in fact we are merely tracing again the steps of reasoning of which it is constructed. Historical facts also, as our information is enlarged, cluster together, and are held in the mind with less tension of memory than while they remained compara- tively few and scattered. A knowledge of their depend- encies enables us to reach one from another, to mingle reasoning with memory, and hold the entire group by the double ties of deduction and recollection. Memory is the simple power of recalling the past in our intellectual experience. We have no occasion for the double division of a conservative and a reproductive power. We know nothing of any conservation save as we choose to infer it from reproduction. The first, without the last, can give no ground of inference even wherewith to estab- lish its existence. Reproduction is the only process that comes under our observation. We do know that the mind recalls its previous states, but how this is done, or whence these states come, are inquiries either impossible of answer, or impertinent to the subject. Indeed, the tendency to ask them, we regard as an unphilosophical one, pushing back of simple ultimate action, and this under the analogies of the material world. Of course those who enter on the wholly theoretical ground of the manner of the mind's pos- sessing its phenomena may find occasion for a theoretical, 136 UNDERSTANDING. conservative faculty, to do the theoretical work assigned it. Of the presence and action of such a faculty, we directly know nothing, and find its existence a matter of inference. If then we confine our attention to actual phenomena of mind, and believe it quite as intelligible that the mind should repeat states in the interim inexistent, as to recall states that have hidden themselves in some region of de- funct ghostly impressions, we have only occasion for one, to wit, the reproductive faculty. What becomes of a thought after we cease to think it, of a feeling after we cease to feel it? From what quarter of the universe do they return to us when recollected? are inquiries whose only gleam of meaning comes to them from material fancies. A power, that should simply hold without being able to recall facts, would be an odd power, a power not powerful enough to show its own existence, an activity too indolent to give the least scintillation wherewith to indicate its whereabouts ; a gratuitous and ridiculous faculty. § 4. The two qualities of a good memory are said to be strength and quickness. These are thought to be separable, to exist in various degrees in different persons. Is not this conclusion somewhat akin to the double division of the power? and does it not arise from not directing atten- tion exclusively to the action of memory? A strong memory is a quick memory, and a quick memory is so far forth a strong, retentive one. We sometimes fully recall things which at first we could not remember, the mind struggling with obscure recollections till the facts one by one come to the light. This result is only partially the fruit of memory; it is largely reached by reasoning, by closely questioning the facts that are retained, and making them witnesses for the recovery of the remainder. When the reflective, philosophical habit of mind predominates, memory may have the appearance of retentiveness without MEMORY. 137 celerity ; but it is an appearance rather than a fact. The weakness of the memory is covered by the strength of the elaborative faculty, and results are at length reached which the memory vouches for, but could not alone have plucked from oblivion. The action of simple memory is aided by other powers and facts of mind. Our recollection fails us, and we strive to grapple the lost fact by inference. We say it must have been so and so, because these were the preceding causes, and these the accompanying circumstan- ces. A clue thus given to memory, the detached fact lays aside its disguise, comes forth from its hiding-place, and confesses itself found. Or the mind keeps in the region of the lost fact. It directs its attention to every resembling or adjunct object, hoping by some thread of association to restore to consciousness the furtive event. The mind thus, in the weakness of retention, avails itself of the logical cohesion of thought, betakes itself from one position to another, lingers in the neighborhood of the lurking impres- sion, to see if from some vantage-ground, from some sudden disclosure, the memory may not again seize it. This, how- ever, is not recollecting, it is trying to recollect, bringing other powers and attitudes of the mind to the assistance of memory. Such a memory is neither strong nor quick. The one quality, therefore, of memory is strength, indicated by quickness ; while what has been termed strength of memory, as opposed to quickness, is to be referred to re- flection. Memory presents different phases of power. Some per- sons recall one class of things easily, other persons another class. Some have a verbal memory, while others are very deficient in this respect, finding it perhaps much easier to retain figures than names. The idea alone is treasured by one mind, while the exact expression is borne away by an- other. These variations seem chiefly due to different de- 138 UNDERSTANDING. grees of strength, and the different degrees of interest at- tendant on diversity of powers. The memory that refuses to retain the precise language is relatively a feeble one, while the thought itself is lodged in the mind as much by the force of the truth, by logical connections, by the inter- est of the statement, as by mere recollection. The power of recalling words, especially proper names, is a good test of the strength of memory, since these, detached from all connection, are thrown as a dead weight on the mind. Weakness of memory may sometimes exist in connection with considerable ease in the retention of figures, since a mathematical habit of mind and general interest and power in this department may concentrate attention on its data, and increase the ability to retain them. The diverse forms of memory are chiefly to be ascribed to diverse tastes and habits, and the interest and attention which accompany them. A tendency once established toward a given pur- suit reacts strongly on all the faculties engaged in it, mak- ing them peculiarly vigorous and effective in that direction. Though memory looks for aid to all those mental pow- ers which unite and correlate ideas, it is by no means de- pendent on them. Its most vigorous and characteristic efforts are almost wholly independent of association. It is when its native strength fails, that association comes prom- inently forward. If the memory could act always with en- tire vigor, it would pick up at random, by any arbitrary, momentary law, the facts of past experience, not colloca- ting them by any of the accepted connections of thought. This, in cases of rare power, it freely does. A person has been found, who, after a single rehearsal, could recite hun- dreds of words thrown promiscuously together, could repeat them backward, could give every fifth, sixth, eighth word, could deal with them exactly as if they lay before the eye. Herein is the perfect, the typical power of memory, and it MEMORY. 139 derives no assistance from association. Even when the in- fluence of association is most manifest, it is only the order of the conceptions which can be accounted for by it, not their actual restitution to the mind. The power to do this work still remains simple and primitive. We need, there- fore, no doctrine of latent states to account for the remote character of two facts reported by the memory ; nor a belief in a great crowd of thoughts, always present to the mind, of only a small number of which we are distinctly con- scious, in order to explain the celerity with which memory produces an appropriate event, or matter pertinent to our state of feeling, or to the argument in hand. The concep- tion that the memory has already partially evoked from limbo a great crowd of facts, and is moving among them as so many personce dramatis, making ready by various laws of association to produce the next fit player on the 6pen stage of consciousness, entirely transcends the facts, is no more intelligible, is less simple, than the statement nakedly accepted, that memory, under the suggestion of a direct question put by a stranger, or at the intimation of the thoughts with which the mind itself is occupied, can di- rectly reach and repeat pertinent previous experiences, and thus enable us to regain, without constantly maintaining, former phases of activity. In many of the connections of association, there is no potency whatever wherewith to re- store a missing member, except as memory gives them that potency. Strength of memory depends much on original endow- ment, though this faculty is as readily cultivated as any of our powers. It comes to do what we patiently insist on its doing. The acquisition of a few names in botany or in or- nithology may at the outset be very difficult, yet in the end memory may retain many hundreds with comparative ease. In extemporary discourse the line of thought comes by 140 UNDERSTANDING. practice to be recalled with scarcely an effort ; yet when the occasion has passed, it at once and entirely slips from the mind. To insist early and strennonsly on the tasks as- signed the memory is necessary to its efficiency. Yet in spite of cultivation, there will be very striking differences in this power. Some will retain lengthy discourses after one or two readings, while others can scarcely repeat with ac- curacy the shortest production. A powerful memory is a great aid to other faculties, though its strength does not seem necessarily connected with the strength of any other portion of our intellectual endowments. Memory is liable to usurp the office of re- flection, and to overshadow the native growth of the mind with the luxuriant products of other intellects. Indeed, there come these compensations to a memory comparatively weak, that we are thrown back more habitually on our own resources ; that the thoughts find free play, the statements of others on the same subject, and their methods of treat- ment not being vividly present ; and that we make all ac- quisitions minister to the vigor and growth of thought, to its nutritive processes rather than to those formal posses- sions which are held in a somewhat lifeless way in the memory. We are thus compelled to enlarge and develop our strength by consumption and digestion rather than by retention. Yet with a truly vigorous mind, that cannot be overborne and burdened by the thoughts of others, a strong memory is a most valuable power. Memory is cultivated in several ways ; first, by persist- ency and vigor of purpose in its use, by requiring positively what has been distinctly committed to it. If its burdens can be made wise, definite, and reasonable, and the mind return patiently to the effort of bearing them, the memory will take up its tasks with increasing ease, and become more and more trustworthy. If, on the other hand, ma- MEMORY. 141 terial in large quantities, in an unanalyzed, vague form is put upon it, and then its delinquencies are passed carelessly over, it will become increasingly slovenly and unreliable in its work. The second method of increasing the strength of the memory is found in deepening the original impressions which objects make upon us. Lively attention, active thought, sincere interest, these are the conditions under which the mind truly receives, and so easily retains the matter before it. Inertness, sluggishness, and wavering attention, in weakening first impressions, weaken also the memory. As, then, our interests is not likely to be univer- sal, a concentration of it, an exclusion as well as an inclu- sion of topics, are needful, lest in dividing we weary and waste our forces, and make our faculties negligent. The third aid to memory is reiteration. Facts which we are determined to retain, we must return to frequently, till we thoroughly possess them. The paths of memory are to be made smooth and hard by use. If we pass on to new material without returning often to the old, a process of separation and disintegration follows rapidly upon growth. We need to integrate and re-integrate our material by re- sorting often to first principles. Thus, while we build layer upon layer around the core of knowledge, we are not to allow 'the circulation of thought to desert this heart- wood, till it is compacted in perfect strength, and lies with- in, an insensible presence of power. The memory above other faculties demands reiteration, repeated integration of its material. A fourth and yet more fundamental condition in the cultivation of memory is that our knowledge shall be made logical, coherent, and fairly complete in the departments it covers. The logical relations of truth greatly support the memory. If the topic is incapable of close connections, yet 142 UNDERSTANDING. many facts near to each other in one field, as in any portion of history or of science, are more easily held fast than a few. It may be doubted whether a wise progress in knowl- edge, notwithstanding the rapid multiplication of facts, is not a relief to the memory. One truth unites itself to an- other, and all are knit together by so many and so rapidly increasing relations, that we have various approaches to each distinct fact. Nothing is of more moment in reference to its own value than this interior coherence of knowledge, nor in reference to the mastery of the memory over it. Detached facts are like sand. We may fill the hand with it, and firmly grasp it, yet it begins at once to ooze out at every crevice till the palm is left empty. Moisten it, till it coheres in a ball, and the open hand will hold with ease twice the quantity. Mnemonics, or artificial associations as aids of memory, is not to be commended, as its application is at best limited, and tends to divert attention from those inherent dependen- cies an observation of which should accompany us every- where. § 5. The second faculty belonging to the understand- ing is that of imagination. By the imagination we mean the power which the mind has of presenting to itself vividly all phenomenal forms. Whatever has assumed, or is capa- ble of assuming, this phenomenal character, whether in the external or internal world, is an object of imagination. A landscape, a melody, a state of consciousness, a character may all be imagined, that is, vividly presented to the mind under their own appropriate forms. As sight is the most full, elaborate and distinct of the senses, giving many par- ticulars, and cutting them apart by sharp outlines, the pic- tures which arise under this form of perception are es- pecially clear and impressive, and hence have given the name imagination to the faculty which paints them, and IMAGINATION. 143 have furnished the general type of its action. Nothing however seems unapproachable to the imagination which is capable of phenomenal existence, that is, of appearing and hence re-appearing in consciousness. Thought, let it be ob- served, enters the imagination as it enters consciousness, merely as a phenomenon. The moment we begin to think, that is to judge, we renew thought as a fact, and do not re- store it as an image. Imagination is simply a general, representative power, and can not therefore work alone without working at ran- dom. The powers which direct it, which employ it in their service, are memory, appetite, desire, the aesthetic and the moral taste. By its aid we restore vividly, that is under a living form, the past ; we intensify the present, filling it with the image- ry of pleasure , we reach toward the possible, the future, in a higher conception of achievement and character. Im- agination is so blended with memory in a portion of its action, that we should hardly separate the two, were it not for other fields independent of recollection on which it enters. It, like memory, is instrumental, and waits the use and guidance of other faculties. § 6. A theory of the imagination accepted by philoso- phers so diverse as Hamilton and Bain, is expressed by the latter in these words : " The renewed feeling occupies the very same parts, and in the same manner as the original feeling, and no other farts, nor in any other manner that can he assigned" (The Senses and Intellect, page 344.) " The imagination of visible objects is a process of seeing. The musician's imagination is hearing, the phantasies of the cook and gourmand tickle the palate." (Page 352.) The statement of Hamilton is not so unqualified, and to that degree less objectionable. Both of them, however, go 1 44: UNJD ERSTANDING. much beyond our knowledge. When a statement so purely theoretical as this explicit, italicized dogma of Bain's, is made the foundation of a complete explanation of. the faculty involved, an explanation resting entirely upon its truth, we see that metaphysicians of the old school are not the only ones who can put foot in air, and mount to the stars. An act of imagination and memory thus becomes with the latter another — as indicated by the clause, "nor in any other manner" — unmodified perception, lingering or reawakened in the organ of sense. The proof of their explicit assertions is found by Bain and Hamilton in the fact, that the organs of action are evidently affected by the images present to the mind in imagination as they would be by the objects themselves, only in a less degree, and that with a loss of any of the senses the power of imagination disappears in a correspond- ing direction. The examples adduced under the first ar- gument are of a kind not leading directly to the conclusion in issue ; but are quite as explicable on other grounds. "A dog dreaming sets his feet a going, and sometimes barks." " Some persons of weak nerves can scarcely think without muttering — they talk to themselves." " Anger takes exactly the same course in the system, whether it be at a person present, or at some one remembered or imag- ined." Suppose our fancies to be pure intellectual acts, independent of the senses, should we not expect these results ? The nervous flow outward on the active, related powers would naturally be secured, though the senses were quiescent, if that state of mind were present which occasions this result. These examples furnish no proof, that the organs of perception are affected, and are the source of this tendency to movement. Farther examples are quoted from Miiller. " The mere idea of a nauseous taste can excite the sensation even to the IMAGINATION. 145 production of vomiting," We think the more correct state- ment would have been, the mere idea of a nauseous taste can produce vomiting. In this form, it loses all pertinence as proof. The active results follow from the idea, the action in the brain, and not from the sensation. We do not in such cases suppose that we taste the disgusting food, but only that we conceive its taste. " The mere sight of a person about to pass a sharp instrument over glass or porce- lain, is sufficient, as Darwin remarks, to excite the well- known sensation in the teeth." Now the setting of the teeth on edge, is an effect of nervous action, and may as fitly follow that action when coming in connection with the imagination, as when occa- sioned by the senses. The fact that fancy affects the ner- vous system, and hence the muscular system, in a manner allied to that of the senses, no more proves the identity of imagination and sensation than a fright . at a ghost proves the existence of a ghost. These examples do not reach deep-enough to do the work required of them. They only show the results to be in a measure the same, whether the object be imagined or perceived, whether the initiative is from within or from without : whereas, they ought to show the organs of sense so affected in what we call imagination as to be a sufficient cause of the effects which follow. Against this, mental and physical experience testifies. (1) We distinguish easily between acts of imagination and per- ception, both in the character and locality of the activity.. An impression that lingers in the organ, or renews itself, there, implies disease. The organ shows its power by a quick, exact response to external conditions only. (2) We observe, also, that the action occasioned by the images of, fancy in most men is slight and ineffectual when contrasted with the results of real perception. Great diversities in character hinge just here, on the predominance of the senses 146 UNDERSTANDING. or of the imagination; the latter type being peculiarly divorced from action. Action may arise directly from stimuli in the senses ; or indirectly from the states of mind evoked by sensations ; or again from states of mind called out by memory or by imagination. This partial agreement of effects does not imply an identity of the causes involved in them. Neither do we find that that which paralyzes the organ of sense necessarily and immediately destroys the power to imagine objects which enter through that sense. A deaf Beethoven can compose music, a blind Milton, blind by disease of the nerve, can write an epic. That there should be a slow decay of the imagination in connection with the early loss of a sense is natural, almost inevitable. The requisite material ceases to be presented to the mind; present possessions, impressions, fade out, and the objects of the remaining senses usurp the place of the lost sense. The doctrine, as stated above, would require that blindness, when an affection of the nerve, should be followed by the instant and entire loss -of the images of visible objects. The facts signally contradict the theory, and the theory fails. The blind man deals with all the imagery of the eye, walks the streets, and uses, to the full, the language of vision. Indeed, in the strict form in which it is stated, this dogma approaches an absurdity. If I imagine a visual object on the retina of the eye, " in the same manner "■ in which I see it, my imagination should be confined to the open eye, and be identical with the impression of objects actually seen. Otherwise it must be conceded, that in one case the agency affecting the retina acts from without, and in the other from within, in itself a grave difference. The imagination of feelings, tastes, odors, should also be as clear and deci- sive as the conception of the objects of sight. Quite the reverse is true, a fact entirely intelligible on the ground IMAGINATION. 147 that imagination is an intellectual power independent of the organs of sense ; as the intellectual element decidedly predominates in sight, while the lower senses are single and emotional in their character, and thus yield less matter to the fancy. We take a certain pleasure in drawing attention to the airy strides of one who so thoroughly sympathizes, as does Bain, with Positive Philosophy. Not having set to ourselves the task of preparing the way for the insensible growth of intellectual out of physical phenomena, we can accept the impression of consciousness, that an act of imag- ination is one of imagination, quite distinct and distin- guishable from every form of perception, clear or obscure. We feel no more interest in discussing imagination under perception, than perception under imagination. In honest induction, we can take what we find. Nor is the intelligi- bility of our philosophy any the less in thus regarding the mind as an independent first cause of its own action than in filling it with echoes, and mild vibrations, and the ling- ering, trembling, sobbing, swelling cadences of sensation, as of a harp unable to part with the harmony that has once run along its strings. These transferred analogies are the most inexplicable of all explications. Mind and matter present plain phenomena, each in its own way cognizable ; but mental movements that are semi-physical, and physical movements that rise into and are productive of thought, have no organ whatever for their apprehension. Seen with the eye, they become purely material ; known by conscious- ness, they become at once and completely transcendental. Two things known as unlike, each by its own faculty, are far better known than when affirmed to be alike, with no common ground or common origin for their comparison. § 7. Whatever is thought of the nature of imagination, its influence and office are not doubtful. It is a great in- tensifier of emotions. Acting under the impulse of desire, 148 UNDERSTANDING. it brings vividly forward the means of gratification, and kindles the passions into a flame. The mind occupied by furious lusts, becomes, till imagery is displaced by reality, the lodgment of a Tantalus. Unreal phantoms provoke the eye, stimulate the appetites, and, in the grasping, sink back into tormenting shadows. The mind is consumed momen- tarily in the red heat of its own passions, which it can neither quell by authority, nor quench by indulgence. Mis- ery in all forms uses the imagination as a means wherewith to irritate and exasperate itself. Discouragement and fear evoke troubles beyond the reality. Not only is the ship battered by the waves about it, the vista of a yet more angry ocean is opened up, and it plunges on from shock to shock, the heart sinking in despair more in view of what is to be, than of what is. Disappointment aggravates the evils it suffers by exaggerated pictures of the good to have been attained. One feels the heat of the desert, and thinks of cooling streams. On the other hand, pleasure owes its hilarity, its intoxi- cation very much to the imagination. It spreads the rosy, blithesome atmosphere of the present to the very horizon, and makes the distance gorgeous with a play of light, be- yond what approach will verify. The eccentricity, the boldness, the poetic inspiration, the enthusiasm of the mind find expression and play chiefly in the fancy. By it we cease to be roadsters along the regular route of existence ; we dart ahead, or fall behind, or turn to the right, or to the left ; we rise upward, tread paths of air, and return only at intervals to the actual, where the toot-sore senses and judg- ment are plodding on. It is evidently this faculty that is yoked to the car of the mind in sleep, and wheels it, in ranging fashion, through possible and impossible scenes, through weird imagery, recollections interlacing fancies in strange and monstrous IMAGINATION. 149 guise. The very fact that the senses find such complete repose in sleep while the imagination is so bold, dashing and wayward, would seem to indicate that the action of the two is far from identical. The same is true in reverie, in day-dreaming. The mind closes its senses, takes out these airy steeds of fancy, throws the rein on their necks, and gives itself up to the luxury of motion along ways in which the friction of ruts, the jar of collisions, and delay of wheels lifting slowly in the mud are not experienced. The imagination also greatly aids our thoughts. The judgment and the fancy are frequently regarded as faculties somewhat opposed to each other in their action. The ease and certainty of the first, in some of its most severe and logical processes, depend very much on the clearness and precision of the second. In solid geometry, in many branches of the higher mathematics, in mechanics, in as- tronomy, a first condition for the ready and safe movement of the thoughts, is a clear unwavering image of the solid, or of the objects and their relations, involved in the problem. If the subject of contemplation cannot be easily evoked, and quietly held in the field of imagination, the judgment is at once at fault in establishing its connections, and gropes in the darkness, like one blindfolded. Scientific inquiry also, the tracing of analogies, the observation of resemblances, are greatly aided by a vivid imagination, presenting distinctly to the mind a large circle of objects. The memory is but very partial in its action without this faculty, and the mind, in the weakness of representation, is compelled to take up objects singly, to the oversight of dependencies which might furnish the key of success. The imagination, then, is as essential to philosophy as to poetry. The difference, lies in the two cases, not so much in the number of objects pre- sented, as in the manner and purpose of consideration. Most immediate and powerful is the influence of the 150 UNDERSTANDING. imagination on action. The pleasures, disappointments, regrets, admonitions of the past, keep company with the mind in that living way which makes them effective coun- selors through this faculty ; and, as the wisdom of the pre- sent is chiefly the gleanings of the past, our immediate purposes its ripened conclusions, the pictures of the fancy are as the reflectors which gather the otherwise diffused, fugitive light, and pour it all in on the working-point. 1 But it is in the ideals of action and character, which are al- ways distinctly present in noble minds, and hardly wholly disappear even with the lowest, that the most constant and valuable function of the imagination is seen. Through a conception of that which is more desirable in ends, more skillful in means, more wise in action, more graceful and winning in method, more pure and holy in purpose, more benignant and beautiful in presentation, imagination fur- nishes an embodiment of the truth nearest us, becomes an angel of light running before us, guiding our steps, scaling for us every steep of excellence, dropping back upon us words of encouragement and hope. To be destitute of an ideal, is to want the best motive of effort, is to lose direction, is to lack momentum, is to be dead, passively preyed on by the forces that have clutched us. Evil and death admit this inertia, goodness and life do not ; and an imagination that looks out on fields of light that open vistas into the paradise of hope becomes an essential to all high resolve and cheerful effort. If the imagination is captivated with the past renewing its life with an art more cunning than that which first spread its colors, we have the poetical tem- perament ; if the imagination keeps close at home in the present, it gives the practical disposition ; or if it pushes out with a noble impulse of exploration and improvement into the future, it discloses to the full the spiritual force of man. IMAGINATION. 151 § 8. The strength of the imagination, aside from original gift, (1) depends on exercise. This faculty cannot fail to be called forth ; the point of interest is chiefly the direction and degree of its employment. When made to minister to the judgment chiefly, it seems to be somewhat overshad- owed by that graver power, and its action oftentimes ap- pears to be less than it really is. Philosophy may be as impassioned as poetry. When, on the other hand, the fancy is left to construct its imagery at the beck of desire, bound down to no useful artistic end, it leaves the mind extravagant in its conceptions, wayward and fickle in its purposes. Persons characterized by the unguarded, ungov- erned action of this faculty, are inefficient and visionary. The most perfect and exclusive training of the imagination, is found in the fine arts. Here it is put to its boldest, yet most restrained and governed efforts. The sense of the beautiful calls it forth, and guides it, and the combined vigor and poise of its action yield the highest works of art, the statue, painting, cathedral. The energetic exercise of our intellectual powers, especially elicit this faculty. All forms of expression seek its lustre. (2) The power of the imagination is also increased by a cultivation of the senses. These give it material. If physical and mental phenomena are appreciatively received in the first instance, the imag- ination will restore them with corresponding clearness. The imagination is thus closely associated with the sensitive force and sportive freedom of the mental life. There is, in this connection, a very misleading use of the word conception, to which we wish to draw attention. That an idea is conceivable or inconceivable, is constantly brought forward in philosophical discussion as a reason for its acceptance or rejection. There are other uses of the word to which we shall revert later, but the use which con- nects conception with imagination, and calls that conceiv- 152 UNDERSTANDING. able which can be imagined, and that inconceivable which does not respond to this faculty, is a frequent and deceptive one. As the imagination deals only with the phenomenal, to say that a thing is inconceivable, is only to say, that it is not of a phenomenal character, not presentable in its essence under a phenomenal form. This may very well be, and yet the idea be one that is to find acceptance. It may be of- fered and urged as one that is not phenomenal, but is of a direct, intuitive character. To say of such an idea, that it is inconceivable, is simply to restate what is avowed, indeed insisted on, concerning it. It is the essential character of an inner intuition, that it should not be an object of expe- rience, and therefore not capable in the fancy of assuming this form. In this sense of the word, the truth of a judg- ment even is inconceivable. The act of judging is conceiv- able, the objects to which it pertains are conceivable, but the truth itself of the judgment is inconceivable. If it were so, we should require no judgment. The act of con- ceiving or imagining, would be sufficient, and would in- clude in itself the entire process of reaching the truth. The judgment is superadded to the imagination for the very reason that new matter is and may be amenable to it. Not to be able to conceive a thing is simply not to be able to im- agine it, and the field of imagination is, in the outset, put down by us as a limited one. When, therefore we are by claim and concession talking of that outside of this field, the assertion disproves nothing, that the subject is inconceivable. Of course it is; if it had not been, we should not have offered it as an intuitive notion, a necessary and univer- sal idea, but as conforming to our observation. The true stroke of overthrow directed against such notions as that of liberty, of the infinite, would be that they are conceiva- ble, and therefore of a phenomenal character, not deeper nor more necessary. To say of such ideas, that they are JUDGMENT. 153 inconceivable, and therefore not true, is to make that a ground of inference for their non- existence, which is in fact the result of their peculiar and permanent character. The blind might as well say, colors have no existence, because they are neither tastes, odors nor sounds, nor are they con- ceivable as such. § 9. The third power of the understanding is judgment. This is, in some sense, the most fruitful and important of all our faculties. To it, the others seem especially to min- ister, and, in connection with it, to fulfil their purpose. By the judgment we rationally combine and use the mate- rial furnished in perception and intuition. It is that action of the mind, by which the phenomena of sense are taken up into the light of reason, there interpreted in their nec- essary relations, and presented as a system of things. The judgment is the power by which we unite subject and predicate under some appropriate regulative idea. The exact meaning and force of this language may not at once be obvious, but will be unfolded by farther discussion. Abstraction, generalization, conception, classification, an- alysis, synthesis, are all processes of thought, requiring no peculiar powers beyond those now mentioned. They are the results and the attendant methods of judgment, judiciously employed. The faculties of perception are not left to perceive all things promiscuously and indiscrim- inately. The judgment does not judge blindly, satisfied with the link of each copulation, no matter whether it lies apart, or is united into a chain with others. This power is set at work in the service of certain intellectual impulses, and works therefore consecutively with selection and rejec- tion, with directed and conjoined effort towards the desired results. Separate judgments are thus thrown into trains of reasoning, and those judgments sought which can be made the parts of such a train. Those objects are considered, lott UNDERSTANDING. > and those qualities in each object, which are, in the present connection, points of interest. Agreements are songht as links of thought, to the dismission of differences. Thus we have abstraction, the separation of one quality or relation in attention from every other ; generalization, the detection of one quality, one form of action, one relation in many diverse objects ; conception in its limited sense, the uniting of several qualities under one generic or specific word, to the exclusion of individual, distinctions ; classification, the uniting of conceptions into a complete system, covering some department of knowledge. Analysis and synthesis are the same processes differently expressed. The first process, that of abstraction, which is the foundation of the other three, is analytic ; while the remaining three are synthetic. By analysis we separate concrete wholes into their intellectual parts ; by synthesis, we recombine these parts in new ways for intellectual ends. These, then, as the various methods and fruits of a fertile judgment, re- quire no farther attention in a discussion of faculties, but belong to logic, which treats of the laws of thought, of the several forms of activity which the one power, the judg- ment, presents. The extent of field and the complex re- sults which belong to this faculty are evinced by the fact, that a distinct science is set apart to it, and the laws of thinking or judging are discussed in a separate and com- plete form as logic. Yery simple sentences are a plexus of judgments. They are primary statements in which many other statements are included. A brief affirmation has the same involution of judgments as a simple perception. The central assertion only assumes the full form of a judgment, while every adjective, adverb, conjunctional and propositional clause, every inflection, modify the primary affirmation by a sub- ordinate statement. The giving of a name, the applica- JUDGMENT. 155 tion of an adjective, the change of a prefix or affix are judgments. The abstract noun in its formation and in its use involves a judgment. One quality or relation is distin- guished from others, and affirmed separately in its associa- tions with them. Generalization is a similar judgment, the ascription to many things or actions of one quality or relation. Conception is the grouping of several qualities, as the differentia of species, genera, classes. Classification is a formation of groups in reference to each other in a field of knowledge ; these groups mutually excluding each other, and conjointly covering all the facts before them, and marking their relation to each other. It involves, there- fore, very many judgments, is made up of a series of judgments. Classification is the final result of inquiry into things, and if it rests, as it must ultimately rest, on the relation of forces, it is the summation of knowledge. Ab- straction leads to generalization, generalization to concep- tion, conception to classification. Analysis and synthesis accompany the entire movement, analysis being foremost in its earlier, and synthesis in its later stages. We should observe how thoroughly the action of the judgment is a construction of relations ; how exclusively the concrete fact is considered in its invisible dependencies. The whole process opens in abstraction, which is a breaking up of the concrete experience, a consideration of it in con- structive parts or elements which have no separate exist- ence. This abstraction is not only not an act of the senses, it is impossible to them, and must commence and proceed in a purely mental region ; one in which relations, not sensible qualities, are the objects of consideration. Each successive step involves this same abstraction in an increasingly ex- tended and complicated form. The conception is made up of a group of qualities which the mind is constantly shift- ing for purposes of more exact classification, and which 156 UNDERSTANDING. have no detached existence in experience. Classification involves an extended consideration of relations that abide only before the inner eye of the mind ; and thus the entire web of knowledge is spun out of a material wholly impal- pable to the finest senses. Analysis yields a tenuous thread of thought, and synthesis weaves it into the invisible con- nections of knowledge. The material of the senses remains after observation and inquiry precisely what it was before, but perfectly new relations have been disclosed between its parts. Supersensual forces and dependencies have been put back of the visible facts, and so they are understood. The judgment, then, can do nothing with mere phenomena. These are declared by the senses. It is not till a new ap- proach is opened up in a process of abstraction, that the relations of things are separately discussed, and the judg- ment combines them in an independent way for its own ends. A getting out of and beyond the senses into rela- tions abstractly considered, is the first condition of the ex- ercise of the judgment. But this condition involves two others, the presence to the mind through the senses of con- crete material, and the power to analyze this material by the intuitive discernment of its dependencies, its primary intellectual elements. Reasoning, a series of interlocked judgments, arises from the same clear and exclusive in sight into relations, and the same ability to deal with them, both independently of concrete things, and as associated with them. § 10. The mind of man uses the contents of the senses simply as the raw material of knowledge. It constructs for itself many systems of arrangement, and by analysis and synthesis works up these subjects of inquiry under them. Though the world is already an orderly and beautiful thing as offered to the senses, it is none the less treated by the mind as material roughly gathered into a museum, there to JUDGMENT. 1^7 be sorted and put in its place. Language is the medium of this great transformation under the constructive force of the judgment. ~No sooner in the use of language was a name annexed to an object, than, by virtue of resemblances discerned though not defined between this object and other objects, the word began to enlarge its application, and to include many things' within a vague circumference of likeness. This movement is so normal to the human mind, that many nouns doubtless never had a very distinct use as proper nouns ; and many proper nouns began at once to expand into common nouns. The fixedness of the proper noun grows up with the fixedness of the common noun. It is quite too much to assert that all common nouns have been proper nouns, simply because this expansion in the beginning would be the prevailing line of growth. Trees seen in a grove, birds seen in flocks, and animals encountered in troops, would be named collectively or rather as a class. As words enlarged their meaning, other words would be called for under the wants of men to designate things more nar- rowly, and so a counter movement of restricted definition would arise. Classification would thus proceed by accident, under the variable interests and inclinations of men. But no sooner is the mind awakened to its own pro- cesses, than all this is altered. Classes are distinctly formed and put in definite relations to each other. Among all com- parable objects proper nouns are extended, and each held fast ; and thus the purposes of thought and expression are subserved with increasing clearness and thoroughness. Much discussion has been had as to the precise significancy of this growth of knowledge by the mediation of language. How far are the results in the things themselves, and how far in the mind ? What are the things expressed by words ? If we take any common noun, as stone, the discussion is 158 UNDERSTANDING. brought to an issue by asking, What does it designate ? In an obscure form the controversy was included in the " ideas " of Plato, those intellectual prototypes of classes. Plato putting the constructive force of mind in the foremost position conceived of individuals as only the variable ex- pression of a controlling idea ; as the Madonnas of Raphael of a conception in the mind of the artist more perfect than any one of them. Aristotle, more empirical in his bent of mind, did not accept the idea as preceding the individual, but as found in it. He explained things from the position of the critic rather than of the artist. In the scholastic pe- riod this discussion gathered heat, both because of its own subtility, and because of its relation to doctrines such as the trinity and headship in Adam. It developed three leading opinions, with minor shades of difference under them. The realists, at first the more prevalent class, asserted that the generic name expressed real generic being, of which indi- viduals are variable manifestations. Thus the same water reappears in many different waves. The nominalists, whose earliest representative was Roscelin, and ablest representa- tive was Occam, affirmed that we have in classes only words and individuals. The conceptualists declared that we have the general word, the conception back of this word, and the individual things grouped under it. Though the controversy lingers to our time, the nominalists and concep- tualists prevail. The empirical school inclines to the first opinion, and the intuitional school to the second. The reasons are obvious. The empiricist works up his mental facts out of the impressions of things. Imagination, on its passive side, is a supreme fact with him. Things repeat their impressions, and these impressions grouped by association become the substance of knowledge. The common noun is simply a word backed by many images instead of by one image. JUDGMENT. 159 While the statement of the conceptualist seems to most minds simple and sufficient, the nominalist succeeds in obscnring the subject afresh by an appeal to the imagina- tion. When we use the word horse, it is said, we cannot realize the idea back of it, save under the image of some individual horse. The word then differs only from Buceph- alus in calling up many images instead of a single image. But this is the difference between the names, John Smith and Martin Yan Buren. It seems plain that the mind can and does proceed without the accompanying imagery of the imagination, and that the fact of images is not the entire fact of classification, nor the very gist of it. If we know things only by perception and imagination, the nominalist is correct; but the entire reconstruction of the world through language, by which the thoughts pass away from and transcend the senses, is in contradiction of the assertion. It is sometimes said, that we cannot think without lan- guage. This would be true under nominalism, since words and things make up the sum of being, and we must handle either one or the other. Yet it is not true, the thought always precedes by a little the word, and words follow on to hold the ground gained. There is no more certain dis- tinction than that between words and the meaning of words, the one giving occasion to the other. If common nouns have meaning, and that meaning is present to the mind, the conceptualist is correct. The by-play of the imagination does not alter the primary fact. The mind passes rapidly through page after page of abstract terms, and scarcely reverts once to any illustrative images. It has not, therefore, been unoccupied, or occupied only with words. The mind in judgment gets as certain a hold of single qualities, as it does in the senses of concrete groups. Take such a relation as that expressed by the conjunction in the sentences, You or I must do it, You and I must 160 UNDERSTANDING. do it ; the mind notes them at once, and shapes its appre- hension to them at once, without the aid of any image. Indeed, no image exactly covers them. The whole ques- tion resolves itself into the reality and validity of the proc- esses of mind, expressed in abstraction, generalization, con- ception. Have we here a power of mind, or only the grouping of images ? § 11. Before proceeding to speak more fully on the ex- act office of judgment, I wish to draw attention to one or two erroneous views becoming increasingly prevalent con- cerning it. Says Sir William Hamilton, "Consciousness, necessarily involves a judgment ; and as every act of mind is an act of consciousness, every act of mind consequently involves a judgment. A consciousness is necessarily the consciousness of a determinate something, and we cannot be conscious of anything without virtually affirming its ex- istence, that is judging it to be. Consciousness is thus pri- marily a judgment or affirmation of existence." These as- sertions are much too broad ; especially so for the philoso- phy of Sir William Hamilton, that we directly know the object of perception as external. All matter in conscious- ness may become a subject of judgments ; if it is thought about, it must become such a subject. But there is no ab- solute necessity that it should, by the judgment, be made an object of attention ; that this faculty should play upon it ; that it should more than quietly now through the organ of sensation without producing any action of mind beyond simple perception. To say that mere, pure perception is a judgment, that " consciouness is primarily a judgment," is an affirmation wrong in form, since consciousness is the con- dition of mental action, and not the action itself ; and erro- neous in idea, since it virtually merges all mental acts or powers in one. If perception is primarily a judgment, so is feeling, so is memory, since out of each of these acts, by JUDGMENT. 161 the same method, a judgment can easily and instantly be constructed. Perception as perception is distinct from judgment, and may exist without it. There is nothing in the one which necessarily involves the other ; though in the rational mind the one gives constant occasion to the other. Moments of perception may be moments in which objects come and go with no thoughtful attention directed to them ; they are left to expire in the sensual impression they are for the instant making. In the case of the brute, is not this the habitual attitude of mind, the field of consciousness occupied with sensations with no reflection on them, or interpretation of them ? "Why speak at all of the power of perception, if, in later analysis, we purpose to resolve it into judgment ? What may instantly spring from an act and the act itself are very different. What also becomes of Hamilton's doctrine, that " per- ception affords us the knowledge of the non-ego at the point of sense," under this farther assertion, that " con- sciousness is primarily a judgment or affirmation of exist- ence." Is such a judgment involved in the perception of an object ? If so, we have not the doctrine of direct, exter- nal perception, but rather the view given by us of the in- ferential existence of the outer world. The two views would be identical, save that we do not affirm, that each single perception compels, or in that sense involves, the formal or actual inference to real, outside existence. It only gives a ground or occasion for such conclusion, which may, or may not, in a specific case, be made. If, how- ever, we do perceive simply and purely " the non-ego at the point of sense," then that act of perception or of con- sciousness is not a judgment, does not include one as its primary element ; or the distinction between judgment and perception disappears, and we infer, and do not in the 162 UNDERSTANDING. ordinary sense perceive, the existence of the external world. If an act of perception, as such, gives us the " non-ego," we find no occasion for an act of judgment to do the same thing. The actuality and externality of the phenomena are already present as a fruit of perception. Does not the diffi- culty lie here, that Hamilton has given to perception a task impossible to it, and then, in later analysis, for a moment forgetful of previous assertions, has made it to involve a judgment, thereby easing it of its burden, though at the same time losing the distinction between these two acts of mind ? The simple content of a perceptive organ is know- able, for perception is a power of knowing. It is not, how- ever, while it remains in the sense simply, thinkable, because thought is an additional action of judgment, nor is it wordable, because words are the instruments of thought and imply it. The only confusion here arises from the fact, that the above assertions apply rather to the primitive data of perception than to its acquired elements. These have been added as the products of judgment. A very limited and objectionable statement of that in which judgment consists, has been much dwelt on by Her- bert Spencer, and distinctly enunciated by Alexander Bairn "What is termed judgment," says he, "may consist in dis- crimination on the one hand, or in the sense of agreement on the other : we determine two or more things either to differ, or to agree. It is impossible to find any case of judg- ing that does not, in the last resort, mean one or other of those two essential activities of the intellect." — The Sense and the intellect p. 329. Says Hamilton : " What I have, therefore, to prove is, in the first place, that comparison is supposed in every, the simplest act of knowledge : in the second, that our factitiously simple, our factitiously complex, our abstract and our generalized notions, are all merely so JUDGMENT. 163 many products of comparison : in the third, that judgment, in the fourth, that reasoning, is identical with comparison." That resemblance, or, stated on both sides, agreement and disagreement, is the sole ground of connection between subject and predicate in a judgment ; that comparison is the only act of mind involved in reasoning, are conclusions quite consonant with a philosophy that derives all the data and the conditions of thought from the phenomenal world, from perception and consciousness ; but is wholly at war with a philosophy that accepts those ideas which illuminate facts and make them intelligible subjects of thought, as of supersensual origin, furnished by the mind itself as adjuncts of its comprehending powers. If we deal purely with phe- nomena, we can only compare them, discover and assert their agreements and disagreements. If, then, we do more than this in judgment, this limited statement should recoil against the system that puts it forth, whose ultimate and consistent product it undoubtedly is. That all judgments do not rest on resemblance will appear in the analysis of the action of the mind in predication — in the office which thought performs. We believe a judgment always to involve the direct or indirect application of a regulative idea to the phenomena included under it, and this is its peculiar feature and oc- casion. Using an undesirable word, judgment is the ra- tionalizing of sensations, it is completing them in thought, through those ideas which the mind furnishes in making them objects of rational contemplation. The full force and proof of this statement can not be easily seen previous to a detailed statement and establishment of these native forms of thought ; yet a little analysis may render it intelligible. Every single perception admits of a judgment, which is the product of the first action of thought upon it. This state- ment has no two perceptions to deal with, and therefore no 164 UNDERSTANDING. ground for a comparison between them. It is simply an application to the phenomenon of a regulative idea. The finger is pierced. A single, sharp feeling is present. We say, it is painful ; a judgment which, restated to give its substance distinct expression, becomes, Pain is. Here the specific experience is taken under the general notion of ex- istence, and we call the result a thought, a judgment that may be offered to another mind. Between the idea of ex- istence and that of pain, there is no resemblance, for I could have as readily affirmed it of a pleasure, of a color, an odor. This judgment, the type of a large class, a step by which any experience whatever receives a form of statement and becomes an intellectual product, is bringing to a phenomenon one of the regulative, formative notions pertinent to it. But I might have said — The pain is one, The pain lasts, The pain is here. In each of these cases, I should have brought forward a different idea, and affirmed its applica- tion in a given form to the sensation. Now, if these ideas are themselves previous sensations, then the doctrine that resemblance forms the substance of every judgment holds good, but not otherwise. If for instance the idea of dura- tion be entirely distinct from the whole and every part of the sensation that evokes it, and is ready to be furnished by the mind to each of twenty or twenty thousand sensations that endure, in order that they may singly or collectively be made intelligible in this relation of time, then this judg- ment, It endures, is one whose predicate and subject are totally distinct in kind, received through diverse powers, and united in another relation than that of agreement by a third power. In this example we suppose a mastery of language, which does indeed in its acquisition imply com- parison. This fact, however, does not weaken the analytic proof, since we can suppose the judgment to be present without the words to express it, or a present mastery of JUDGMENT. 165 words independent of the training which leads to it, is essential to it, thongh not to the very act of judging. The belief which identifies comparison and judgment must make the notion of time derivable from a number of sensations *; something in the sensations themselves, rendered discernible and comprehensible bj repetition. It would thus follow that a single sensation could not be made the occasion of a judgment, since there is in it no opportunity for compari- son. It is a unit. The mind has nothing to bring to it, and it abides barren in the organ of sense alone. The feel- ing could then no more be said to exist, than it could be said to be unusually intense, since both assertions are alike relative. How Hamilton, who has given his authority to a statement so alien to the intuitive philosophy, would dispose of the fact, that the mind puts a single perception in the form of a judgment, a point he especially insists on, going so far as to say, that perception necessarily involves judg- ment, is not evident. In the first act of perception, there is no material present to the mind, between which to insti- tute the comparison said to be involved in the judgment, itself involved in the perception. To initiate such a move- ment Hamilton would be compelled to make his com- parison between the pain and the idea of number, the idea of time, the idea of space, the idea of existence, and affirm at this point a resemblance, a complete abuse of the word comparison. The objects compared are unlike in kind, belong to alien fields, and do not admit the notion of similar and dissimilar. In fact they must admit simi- larity if either, since the two are coupled in a conjunctive judgment. Only as we regard the time, the unity, the existence, as in some way in and a part of the sensation, and also in and a part of the other sensations present to the memory, can we make these judgments examples of comparison. That these ideas cannot be thus directly dis- 166 UNDERSTANDING. cohered as parts of sensation, as Spencer, Mill, and others affirm, will be further seen in a later discussion. In this class, then, of judgments, which are statements concerning single states of consciousness, it is evident that a regulative idea is united to a phenomenon, and the content of the lower organ, so to speak, taken up into the intellect. To this class also may be added those judgments in which the same idea, as existence, place, time or number, is affirmed of several phenomena ; and we have our first division of judgments, those which directly unite to facts a regulative idea. § 12. Another form of judgment unites two distinct objects under a regulative idea. Of this character is the statement, This apple is like that apple. Under the notion of resemblance, two objects, the products of distinct sen- sations, are united. Here the thought-process consists in bringing the two together under a comprehending form or rational notion. It is to this kind of judgment, that Bain and Spencer would analyze all thought, omitting even here the essential feature of the act, that a notion is in- tuitively seen by the mind, under which the movement goes forward. A sensation is complete and independent in itself, and does not necessarily lead to any farther state of mind. This it may or may not do according to its con- nections. In reflex action, so-called, an inward current, that may not affect consciousness, is followed by a physical force, by an outer motor current. This inward movement may, as a sensation, enter consciousness, and may thence go forth in certain, involuntary, automatic actions ; or, as a sensation, it may be taken into the processes of thought, be merged in the intellectual movement, and reappear as a voluntary act, a new and independent impulse. Now the sensations occasioned by the presence of two apples, may simply and directly, as in the case of a brute, draw forth JUDGMENT. 167 action ; or they may become the occasions of thought, and the inquiry be instituted, whether they are of one kind, or of different kinds. For the first result there are necessarily present appetites and senses ; for the second, rationalizing power, which is no other than the power to furnish an idea, in this case, that of resemblance, under which an inquiry can be instituted and a judgment formed. It is the exact office of the judgment to apply discriminatingly, in refer- ence to an end, these notions to the objects before the mind. The sensations, as sensations are complete. They are not halves ; they are not uneasy, nettlesome, looking out for mates; nor adhesive, linked, dragging something after them ; nor are they dove- tailed into thoughts, making their succession inevitable. They might lie forever perfectly quiet, nothing coming of them, were it not for the appetites below them, into which they sink by physical connections ; for the eye of reason above them, into whose realm of thought they rise, by the dropping down upon them of judgments, through tentative inquiries prompted by its own perception of invisible, unheard, unfelt relations. This working up of sensations, this vitalizing of them in pro- cesses of thought, needs solution, as much so as the activity of chemical elements previously dormant, when heat is applied. We know an object as red, as sour, as fragrant, through our respective senses of sight, taste, and smell. A judg- ment has nothing to do with this knowledge. The first ob- ject received in any sense imparts to it a form of knowing, in itself ultimate and inexplicable. When we meet with a second object of a like kind, we have no new sensational knowledge ; yet we have an occasion of a judgment, which we did not have, as regards the quality, the flavor, or odor, or color in the first case. We say of the two, They are the same. Now how happens it that the second sensation 168 UNDERSTANDING. has in it more than the first, to wit, this occasion of a judgment ? As sensations they are alike ; one is no more stimulating than the other and, to the sense, should yield no more than the other. The solution lies in the fact, that the mind is able to furnish an idea, that of agreement and disagreement, infus- ing rational order and relations into a plurality of objects, and brings it forward for immediate application, on this the first occasion. Here the judgment finds its function and office to run between phenomena, and marshal them under notions. Of phenomena alone it could make nothing. It must have its men, and its plan of rank and regiment, and then it can construct an army. Of the same character are the judgments, This is higher than that, This event is more recent than that. In each case objects of perception are thrown into relation with each other, by means of a regulative idea. Many, accepting the intuitive nature of the idea of space, would easily recognize the character of the judgment, This house is nearer than the mountain, who would yet fail to see the transcendental element in the kindred statement, This stone is like that rock. Evidently the mind furnishes the ground of the judgment, — the idea of the relation, as much in the one case as in the other. The present division of judgments in- cludes all acts of classification, and is a most numerous one. The statement, This action is right, may sometimes be one of classification, assigning the act by its form to a kind or class previously recognized as right. More frequently, however, it is a judgment of the first class, in which a single act is stated and interpreted under an intuitive notion. The notion right is not perceived in the action, but brought to it, discerned as a spiritual factor in it. If it had been red- ness that had belonged to the object, the mind must needs have waited for a second, third, fourth instance before it JUDGMENT. 169 would have said, This is red : and then the assertion would have been one simply of classification. The perception gives the quality, and the judgment remains quiescent till, by repetition, it is called to the act of classification. In the case of the right action, however, the action enters through the senses without this quality, simply and nakedly as an action, and the reason bringing forward a farther idea for its explanation as the act of an intelligent and free being, the judgment at once finds play in applying it, and says, It is a right action. This it might do should the mind never know another, if this act in its motives and consequences were plainly before it. In the first class of judgments, one limb of the predication rests on the phenomenal, the other passes over into the purely intellectual, the transcendental. In the second class, both abutments of the arch press back on phenomena, but the spring and crown of it rests in the air ; the connection strikes into and returns from the region above. There is a third class of judgments of which the expres- sion, The heat melts the wax, is a type. Here, under the notion of causation, we grapple by a judgment that which physically exists, yet never directly enters the phenomenal world. The mind walks as one who travels on a morass, the points of support are hidden a little below the surface. The foot, under the quick suggestion of the eye, and the inference of reasoning, dashes at the more stable ground, which it never sees, and is yet able to find. The mind could not move, did it not believe in causes, yet it never sees a cause, or knows causes save through effects constantly attributed to them, safely expected from them. It is not sufficient that the mind should weave the visible into a firm fabric of order by invisible connections it alone can grasp ; it is made to stand, and must forever stand, and all it beholds stand with it, on invisible, intangible supports 170 UNDERSTANDING. of forces, whose existence it can verify by no sense, and mnst leave with its own assured conviction. Deny these supports and it must yet seek for them, and believe in them, and talk about them every moment of its life. These judgments by which we spread the phenomenal over the actual, by which we search out the streams of force, and feel the under-flow of divine power, are among the most constant and radical of any we ever make. It is evident, however, that they do not rest on resemblance merely, since the cause is never in any way phenomenally known, save through its effects, and therefore furnishes no hold for a comparison. Of course we mean the actual cause, and not the phenomenal cause, that is the effect just previous to the effect under consideration. We mean the very heat and not the taper, which is itself in its visible form an effect, and not a cause. In the third form of judgment we unite the sensible and the transient to the insensible and permanent through a pure intuitive movement of mind. "What was understood to be, also understood in relation to other things that are, is now referred to hidden sources or causes. We do this under two notions. That of causation and that of spontaneity. In this third class of judgments we refer effects to causes and acts to agents. It is then the general office of judgment to unite the phenomenal and the intuitive, the perceptive and the purely intuitional elements of mind, in the rational apprehension and use of the former. Reasoning is the interlock of these judgments, a chain of these conclusions by which remote points are united, and discloses therefore no new power. § 13. Before passing from the judgment, we wish to mark a second use of the word conceive, leading to further obscurity. By a statement, that an idea, for instance that of infinity, is inconceivable, seems sometimes to be meant, that the judgment cannot grapple it, that it can not be JUDGMENT. 171 wrapt about sufficiently with logical relations, worked up as material in the processes of thought. Very well ; the judg- ment deals with the phenomenal under ideas, and therefore a notion not phenomenal, and not calling for the interpre- tation of a farther idea, is not material for the judgment. The judgment ought not to be able to handle it ; if it were able, a phenomenal element would therein appear, destruc- tive of its pure intuitional character. The query still re- mains, however, whether such an idea may not be validly presented to the intuitive power set apart for its apprehen- sion, given to perform this very service? The 'incogita- bility ' of a thing may be proof of its nature, though not necessarily of its reality or want of reality. We are now able to see something of the relation of the understanding to the entire mental furniture, and also of the three powers which compose it to each other. The understanding plays between the intuitive parts of our nature, the physical perceptions on the one hand and the spiritual intuitions on the other. With no absolute final comprehension of either, it interlocks them, and comes to a definite knowledge of their relations. This knowledge of connections seems to us more satisfactory than that of qualities in perception, or of ideas in intuition. We try to make a color, an odor or the notion of existence, an object of reflection, and can do little or nothing with it. As simple and primitive, it eludes those relations which we are so diligent in establishing between objects, and the mind, perplexed by its inability to fasten and weave the web of thought, is ready to feel that there is here no real knowl- edge ; forgetful that an organ of sense, or an intuition gives a new and final form of knowing. All knowledge is good and adequate, if we know enough to recognize and accept it. The understanding furnishes us a knowledge of relations. The judgment, like a busy shuttle, flies between the 172 UNDERSTANDING. loose, parallel, independent lines of phenomenal being, bears with it the interlacing thread of intuition, and shortly weaves all into a firm, coherent fabric, a system of things. The steadfastness and permanence of the work are secured by memory, while its brilliancy, the vividness of its color- ing, arise from imagination. We thus seem to see some reason why these faculties, and no others, are called for. The judgment, under the eye of reason, knits together facts into relations, which make them significant and intelligible, which show them to be a system of things ; the memory stands by to proffer the facts, and store the fabric ; while the imagination dips again in living colors these shadow products of the mind, as the sun saturates the cloud with its own hues. § 14. There are very important laws of thought in connection with the understanding of which but slight men- tion has so far been made, those of association. Association is brought forward to explain many pro- cesses of mind ; we believe that association is rather the result of these processes. Ideas, images are not associated in the mind otherwise than as it itself binds them together. Reflections in a mirror gain no coherence, have no power to restore each other. Aside from the analogy of physical habit, which is of narrow, indirect application to mental phenomena, we have no reason to believe that ideas as ideas can maintain a sequence among themselves, the first idea bringing with it the second, the second the third. It is judgment, memory, imagination that separately or in com- bination restore ideas and images ; in these faculties, not in themselves, are found the laws of coherence. Involution, causation, resemblance are the primary connections of judg- ment, and time and place of the imagination and memory. When Hamilton resolves the laws of association first into simultaneity and affinity, and then into the one law of re- ASSOCIATION. 173 dintegration he is not stating an ultimate force acting be- tween ideas, but is falling back on the memory and judg- ment, and later on memory alone. Things once together in the mind are restored again to this first relation by virtue of memory. This restoration is memory, and is not to be ex- plained by a power of images over the mind, but of the mind over images. Association, greatly enlarged as an explanatory doctrine by Heartly, has been increasingly brought forward by later writers as an all-inclusive law of mental phenomena. Taine goes so far as to speak of the anterior and posterior portions of an idea, and ideas are made to arrange themselves and adhesively drag themselves through the mind, quite passive in reference to them. Spencer, in his Data of Ethics, p. 109 makes this concise restatement of mental processes. " The first of these elements, originally an excitement, becomes a simple sensation ; then a compound sensation ; then a cluster of partially presentative and partially representative sensations, forming an incipient emotion; then a cluster of exclusively ideal or representative sensations, forming an emotion proper ; then a cluster of such clusters, forming a compound emotion ; and eventually becomes a still more involved emotion composed of the ideal forms of such com- pound emotions." Mental powers have disappeared in be- half of some assumed coherence among the molecular con- stituents and movements of ideas; and obscure physical images have taken the place of clear mental facts. This power of association among images can in no way be ex- plained except by referring it to cerebral states, that in some way collocate and continue themselves. But these alleged facts are unknown, obscure, inexplicable ; far more so than the facts they are brought forward to illuminate. We know of no distinct coherence of nervous states that can be plausibly offered as the exact equivalents of logical pro- 174 UNDERSTANDING. cesses. The connections between stimuli, sensations and muscular action are not of this order. Nor, if we were aware of such special dependencies, could these be pre- sented to explain the connections of thought. Ideas mani- festly cohere as ideas in the mind by the mind's action ; they are not seen to follow each other by a physical link. There are in this doctrine of association, as very generally held, a wonderful displacement of plain facts by obscure ones, an astonishing assumption of facts, and a strange in- sufficiency in the expositions offered. The law of association, now so omnipotent in philoso- phy, has grown up slowly in connection with that empirical tendency which more and more divests the mind of power, and accumulates within it a series of impressions that carry with them their own connections. This philosophy should be able, first, to show clearly and certainly that a given cere- bral state is the precise equivalent of a given mental one, and its immediate cause. Nothing approximating this has been done in a single instance. It should then show that these cerebral states have cerebral laws, by which they com- bine with and follow each other ; and, third, that these phys- ical connections are the exact equivalents and causes of the various relations of thoughts, feelings, and volitions to each other. This work has not been so much as entered on. The law of association is left regnant when no sufficient ground of connection has been disclosed between cerebral facts and mental facts, and even while the cerebral facts themselves are purely theoretical. If all this had been done, we should have simply a physical or organic philosophy in place of a mental one ; but it has not been done. Ideas, thoughts, feelings, volitions are intangible, have no power over each other or relation to each other save those intangi- ble ones which the mind itself, in evoking them, imparts to them. The mind makes a feeling to be a feeling, and so ASSOCIATION. 1T5 gives it its energy. The logical insight discloses the logical relation, and so welds the conclusion to its premises. Thus also the memory renews those impalpable, vanished rela- tions expressed by place and time, and, in doing so, sets things once more as they were. The primary powers in association are memory and judgment. Memory restores things as they have been phe- nominally offered to us under the connections of time and place. Judgment unites images and ideas by the relations of resemblance, causation, involution. The imagination, working with both these faculties, combines its images un- der all applicable ideas. We have thus, as expressing the relation of ideas to each other under the force of memory, the law of contiguity ; and under the force of judgment, the law of affinity or congenesis ; and under the energy of the imagination, the law of congruity, a combined and soft- ened application of the other two. The laws of association are convenient as forms of expression, as rendering relations on their objective side which are due to the hidden energies of mind. But when these laws are spoken of and applied as ultimate, they become another of those many disguises of words by which sequences are put for forces, and state- ments for reasons. The question involved is one again of the nature of mind. The seat of energy is transferred from mind to the products of mind. CHAPTER IY. The Reason. § 1. We have now reached an action of mind, a faculty, whose existence is strenuously denied. Most able and thorough thinkers, patient inquirers within the field of phi- losophy itself, with a host of scientific investigators who bring with them predilections and reasonings suited to other departments, regard this furniture of intuitive ideas as wholly fabulous, as an unnecessary assumption in the ex- planation of phenomena entirely intelligible without it. Yet there is in philosophy no point of more importance, of more wide-reaching influence than this ; and that, too, not merely in the department itself, but in its social and moral and religious bearings. It is vain to strive to disconnect social and religious issues from mental science. The in- stitutions of society and the commands of God have man for their subject, and neither their defects nor their excel- lences can be understood without a knowledge of his nature. Indeed, the very character of those notions on which duty turns— of right and of liberty — are here brought under dis- cussion ; and also the validity of those conceptions and that reasoning on which the existence and government of G-od repose in our thoughts. The past attachments of our na- ture, its present powers and future hopes, are all involved in these investigations of philosophy, and more especially in that branch of them which settles the original endow- ments of mind, and the degree of its dependence on the external world. Indeed what is meant by the external and DIVERSITY OF OPINION. 177 internal worlds, and whether either or both of them can furnish a valid proof of their being, are inquiries that are now to find settlement, or to be left unsolved doubts, unex- plained fears, ultimate mysteries, drifting athwart the mind, restricting its spiritual vision, and displacing its cheerful surface-life with the shadows of deep, despairing clouds. Yet these discussions are as subtile and perplexing as they are important ; and moreover are looked on by patient plodders amid facts — most influential and serviceable men — as hopeless and futile. They regard these labors as the mere money-maker would regard another expedition to the North Pole. The whole region, to the purely scientific mind, seems one of chimeras, not dire only because in- creasing wisdom enables us to laugh at them. Ghosts are al- ways unproductive, and to men ridiculous. The only touch of kindly sentiment that the student of natural science has on this subject, is the regret that so many are still found to waste a hope or a fear on such airy existences ; are yet unwilling to confront daylight with open eyes, instead of owling in invisible regions for invisible things. In these fields of difficult and abstruse inquiry, we shall need to work our way slowly and patiently, confronting our adver- saries fairly, ourselves convinced of the importance of the truths here hidden ; sanguine as to the power of the mind to push and answer the questions most intimate to its own destiny, and repelling the scorn of ignorance with the silence of settled conviction; knowing that if ours or another's keel shall ever touch the distant shores of truth, shall ever add to the hemisphere of matter that of mind, the question, Who are fools ? will be easily settled. The ideas in dispute have received various designations. They have been termed innate ideas, regulative ideas, intui- tive ideas, a priori ideas, categorical ideas, and also have been regarded as forms of thought, entirely independent of 178 REASON. the objects or matter of thought. Some of these are very faulty methods of expression, especially if adhered to as complete in themselves to the exclusion of other methods. Indeed, no one word or expression is perfectly applicable to any one of these ideas, and the relation of the mind to it ; much less is such a word sufficient to characterize all of them, varying as they do, intrinsically, and in their connec- tions with the phenomena explained by them. One seems to inhere like a quality, as right in action ; another to be a condition, as space to the objects in it ; another to be the manner in which the mind regards the things to which it applies, as number in connection with the objects num- bered. No single expression, therefore, can be analyzed, no particular words tortured to disclose more exactly what is meant by an intuitive idea. In each case the relation itself must be contemplated, and the word be shaped to the fact, rather than the fact be learned by the word. "We do not understand by this doctrine of innate ideas, that the mind finds in itself a notion as a realized mental product, and applies this to the facts before it; nor that there is in thought certain forms or directions of movement from which it cannot depart, and under which it works up the material brought to it. We understand rather that in the facts, on the occasion of the facts, the mind, not the senses, discerns relations by which it is able to explain them, to think concerning them, and this by means of a certain rational power which it brings with it, or finds evoked under the conditions of the problem before it. Nor do we affirm that these notions come necessarily and at once to every mind on every occasion intrinsically fitted for them ; but that they each and all do find, sooner or later, an occasion on which they do arise, and that there is in them a furnishing by the mind itself of other and higher material of thought than the senses alone can supply. In EXISTENCE. 179 other words, there is called out, in the intellectual handling of the facts of the world revealed in perception and con- sciousness, a new power of mind, which we term the reason, furnishing rational ideas and grounds of procedure, and en- abling the judgment to operate on the otherwise stubborn, irreducible sensations present. The existence and office of a portion of these ideas all philosophers admit ; they are at variance only as to their source and nature — a variance which leads to the denial of the remaining and most es- sential ones. The notions of space and of time, traced to an empirical source, prepare the way for a denial of right and liberty in their transcendent character. The term, innate, was an early and unfortunate one, as it seems to imply an inborn product rather than a primitive power. The word, regulative, as also the word, categorical, find application to these ideas as bringing with them into knowledge fundamental lines of order. The adjective, in- tuitive, indicates the directness with which they arise, like an object of sense ; while they are termed a priori as logi- cally preceding the experience they expound. An occasion being given, each idea arises at once as the condition of its apprehension. These ideas are either denied by the empiri- cist, as that of liberty ; or differently referred, as that of space ; or modified, as that of right. We shall now proceed to take up these ideas one by one, both to establish the whole class and each member of it singly. In doing this, we shall not hesitate to repeat the argument in each case, so far as it presents any new features. The general doctrine of intuitive ideas is maintained if any one of them holds its ground, though for its successful and thorough application, the exact number and nature of these notions must be known. § 2. The first of them is that of existence. This has drawn forth less discussion than some others, and does not 180 REASON. therefore afford the best ground on which to meet the op- posing views. The affirmation is, that in the presence of sensations, perceptions, the mind comes at some moment to say, These are ; or, involving another idea, that of causa- tion, to say, The object occasioning them is. When this act of mind does take place, there is proof in it of a double activity aside from that of the judgment — an activity fur- nishing the perception, and a second activity supplying the predicate. Can the judgment be made without both of these conditional activities ? Can the three be resolved into two, or one ? We answer, no. The judgment can do noth- ing with a naked sensation. It is to this higher faculty, lumber without tools. The sensation can yield nothing but mere feeling. Feeling, as feeling is complete in itself, and may as well repose in the sensational structure of an oyster, as in that of a man. The judgment alone can add nothing to that which it is to handle ; for if it does, you therein as- sign it a double office, that of reason and judgment, that of calling forth the predicate, and of coupling it with the an- tecedent. A sensation and the notion of existence involved there- in, or better, evoked thereby, are very different. I see no reason why the one may not be experienced indefinitely without, in and by itself, giving rise to the other. Indeed we, with our rational powers even, are constantly enjoying or suffering sensations without affirming, or thinking of their existence. This notion is present only as the mind from time to time is brought directly to contemplate them. There is no latent judgment of their existence in clearly ex- perienced, but not definitely thought of, sensations, in any other way than that the mind may, at any moment, have its attention directed to them, call them before itself for con- templation, and then be led to affirm their existence, under this mode of regarding them. A cloud is above the earth, EXISTENCE. 181 and the mind may so decide at any instant : but there is no latent decision to that effect in the simple act of seeing a cloud, only the possibility of one. As the opposite view has not here received that com- plete and exhaustive statement which we shall find of it under space and time, we can not to the best advantage, con- trovert it. We merely remark, that it seems to confound the sensation with the idea. This it does partly perhaps through the ambiguity of the word consciousness. It is not an unusual or very harsh form of expression to say, I am conscious that the odor exists, while the affirmation, I smell that it exists, is obviously inadmissible. Yet for philoso- phical purposes the last expression has all the breadth that can be allowed the first. Consciousness only reports the sensation, is as broad as the sensation, and this is fully ex- pressed in the verb, smell. We are not then conscious of the being of an odor, unless we smell that being. The words conscious, consciousness have so enlarged their mean- ing as to be regarded as the ground of that which is known, when that knowledge springs from a judgment, and is thus referable to a reflective and not intuitive faculty. We are conscious of a sensation in no other sense than that we are conscious of the sensation, and also of the intuition and judgment by which existence is referred to it. These three acts are separate sources of separate elements in the joint product, This odor is. Consciousness is nothing in itself, nothing additional, but is the common and pervasive condi- tion of each of these acts as of every act of mind. No knowledge can be referred to consciousness which is not farther, more explicitly, referable to some given specific power of mind, and the power of mind yielding the notion of existence is the one here insisted on, the reason. Says Bain, " The sum total of all the occasions for put- ting forth active energy, or for conceiving this possible to 182 REASON. be put 'forth, is our external world." (The Senses and the Intellect, page 380.) In this and the accompanying pas- sages, the sensations of resistance, rather than the sugges- tions and interpretations of those sensations, are kept upper- most, and thus the action of the reason concealed under that of the senses and the judgment. But " occasions " are occasions to the mind, and thus become the conditions of a knowledge not found in the simple sensations which compose them. The judgment of existence does find its chief signifi- cance in connection with the experience and exercise of force, since here, united with that of causation, it leads to the telling affirmation of the noumenon, the permanent being, underlying the phenomenal, material world ; and also of the spirit, the abiding source of changed and changing mental states. To affirm phenomenal existence seems a merely formal act beside the doubly pregnant one by which we go deeper than consiousness inward, farther than con- sciousness outward, and fill supersensual regions with super- sensual forms of being. Phenomena are known directly, and thus directly yielded in consciousness ; but now the conditions of a judgment are found which penetrates be- yond appearances, and affirms permanent and unphenom- enal existence, a fact incapable of experimental verifica- tion, and thus of appearing directly in consciousness. We are conscious of judgments, not of their truth. We refer then this idea of existence to an independent faculty, the reason ; because it is not in the sensation as a sensation, nor to be secured by a passive flow from sensa- tion to sensation, each equally destitute of it ; but is found first and fully in the incipient action of mind, when it begins to deal with and handle its hitherto unobserved experiences. This judgment is too habitually involved in our experience to be ordinarily significant, but to one NUMBER. 183 waking from syncope, it may become a most momentous conclusion. § 3. The second regulative idea is that of number. This, like that of existence, is so simple and direct, so con- stantly merged in the very perception to which it is at- tached, as to have called forth little discussion, and made but slight claims for explanation on sensualistic schools of philosophy. Language also favors this oversight. I see one apple, I hear several sounds, I feel three distinct points, are examples of familiar expressions. We cover directly by verbs of sensation, their objects and the numerical rela- tions of those objects. Yet it is evident, that we do not see an object to be one. The numerical notion is brought to the mass of colors before us as one of the ways in which the mind may regard it. Indeed, the same object differently contemplated, yields a great variety of numerical relations, the sensations remaining exactly the same. It may present several colors, and while, therefore, we call it one in cohe- sive connection, we may separate it into a multiplicity of parts by diversity of shades, or by outstanding members, or by relative position. An object of regular outline and uniform color may still yield a plurality of parts through the unit of measurement we apply to its lines, angles, surfaces. The mind plays upon it with standards of its own, divides it with various linear and solid measurements, finds with each a diverse numerical expression, and terms it now one, now many, as suits the purposes of thought. All this is not a simple action of the senses ; nor any more is it when the incipient step of the process is taken by roughly calling the whole one thing. The color is seen, the hardness is felt, the odor is smelt, and the sources of each are regarded as one object, or more than one, as the mind chances to contemplate it, bringing to it one or an- other of various combining ideas. There is no object of 184 REASON. sense which is not in some relation one, as a tree, a grove, a forest, a world, a universe ; and none which may not be divided and thus yield plurality. Now this action of the judgment and attention must all go on under the notion of number, and, till this is furnished, all objects must remain undistinguished either as single or manifold. Objects of sense may reach the mind without drawing from it a numerical estimate. One may gather berries without re- garding the number taken or left, though both be clearly seen. Distinction in the senses is not distinction in the intellect, and does not necessitate it ; any more than dis- tinction in existence is distinction in thought. A dozen calls may bring a dog, though he has taken no note of them as a dozen. Articulate sounds may convey the designed thought to the mind, a thought dependent on the exact number of elements, without attention directed to them as twenty, less or more. This separable character of number from the objects perceived, is seen in the fact, that two impressions on the senses, as on the eyes of men, or many, as on the eye of an insect, become one object in the intellect ; and still more strongly in the fact, that numbers are treated independently in arithmetic and algebra, are accumulated in amounts entirely beyond experience, and are divided and com- pounded by processes not founded on observation, or proved by it ; but which belong to the necessary character of numerical conceptions. Our powerful algebraic solvents are general formulae, are wrought out wholly independently of things, and are brought to explain outside facts other- wise numerically unintelligible. Thus most evident is it, that in the more abtruse application of numbers, as to curves and to complex motion, phenomena receive their so- lution from the numerical conception and do not, through the senses, yield it. Moreover, these estimates are reached RESEMBLANCE. 185 by an arbitrary supposition of an equality of units never found in experience. One pound is regarded as absolutely equivalent to every other pound of the same denomination ; one foot, one mile, to the like measurements elsewhere. To fix on standard units, in which the approximation to equality is sufficiently close to enable us safely to neglect errors, is a large share of the difficulty in mixed mathemat- ics, and only when we deal with pure conceptions, as with that of space in geometry, do our numerical processes show their full power, stretching an unimpeded wing in realms as airy as themselves. Existence and number are among the most general of our notions, finding inherent, and, to a rational mind, necessary, application everywhere. We regard the idea of number as one brought by the mind to things, (1) because the thing considered remain- ing the same, numbers may be applied to it in many differ- ent ways ; (2) because distinctions in the senses which are naturally an occasion for the notion of number do not nec- essarily call it out ; (3) because it is applied to things not objects of sensation ; (4) because it assumes an absolute equality of units ; (5) and because abstract numerical pro- cesses wholly transcend experience. § 4. The next regulative idea we offer is that of resem- blance. This idea, though recognized by Plato, has been very frequently overlooked, and with great injury to the arguments sustaining the Intuitive Philosophy. It has been quietly assumed that resemblance is a matter of sen- sation only, that in it exclusively are given the data of this category, that one color is seen to be like or unlike an- other ; one taste tasted as like or unlike a succeeding one. We might as well claim the judgment in which this rela- tion is expressed to be an act of sense. Green, red, sweet, sour, are known as qualities by sensation, and here the sense pauses. The eye sees a green color once, twice, thrice, 186 REASON. but it makes no comparison, institutes no judgment, recalls no impressions. These, the labors of other intellectual powers, must commence and go on in the light, and this light is that of an interpreting idea. "What is resemblance % It is not the red in the apple ; no more is it the red on the leaf ; no more is it these two sensations united in time and place. It is a specific relation between the two, intelligible as a given case of a general notion. Can the specific rela- tion be first reached, and the general idea be deduced from it % ~No ! As a relation it is an intellectual product, an in- tuition, two sensations explained in their bearings on each other under an idea. The sensations alone do not contain in their sensational matter the relation, — if they did, each should contain it entire, or each a part of it— and can not furnish it, nor can the intellectual movement proceed with- out the forecasting apprehension, the head-light. Moreover, the specific relation must express the general relation, or that relation cannot be deduced from it. Resemblance is intellectually involved in the first instance of it ; and, as it is not a sensation, it must be involved for the direct appre- hension of the mind there present for its interpretation. It is not the result of the judgment which expresses it, but an element and ground of that judgment. There are sensible and supersensual data for the declaration, The leaf is like the apple. The frequent over-sight of this fact has greatly embar- rassed the discussion between the two schools of philos- ophy. The idea of resemblance has been quietly appro- priated. The observation of agreements and disagreements has been allowed to proceed as if it were purely a matter of perception, and thus a play of mind has been secured, a germ of judgment, a nucleus of thinking, with no recog- nized a priori material. From the elements of intellectual action thus secured, it has been comparatively easy, by pa- RESEMBLANCE. 187 tient composition and slight oversight, similar to that which characterized the first step, to broaden the grounds of thought, and to surreptitiously include one after another of its essential conditions. This process is arrested at the out- set, if we reclaim, as we should, the idea of resemblance. No generalization can go forward without it, and the fic- titious growth of regulative ideas is checked at once. We can not, for instance, compare sensations as co-existent or as successive, and under the one agreement smuggle in the idea of space, and under the other that of time. We are left, as we should be, standing on sensations alone ; knowing color, odor, taste, but with no opportunity for comparison, classification, generalization, as we have no luminous idea under which a movement of thought is made visible. There must be a little play given to thinking some- where, in some direction, under some notion, before it can work out anything whatever ; before it can acquire momen- tum, institute a process, and, in the superficial movement established, give apparent ground for the true connections of thought. The sensations are indeed present as the ma- terial of thought, the judgment is waiting as the agent of thought ; but there is no plan of thought, no direction of thought, no space or orderly way wherein thought can find exercise, till some notion, most frequently this of resem- blance, is furnished. The axe cannot cut while it is pressed close against the timber ; tools are of no avail packed tightly in a chest. Give the hatchet the play of an inch, and with patience and an increasing sweep, it will at length hew for itself a broad path. Scope must be granted for wielding a weapon. One after another the implements must be loosened from their lodgment, and to initiate this move- ment, room, the ground and condition of effort, must be granted. So must room, an idea under which to move, be given to the very first judgment, before generalization is 188 BEASON. possible, and the one stolen for this purpose, is that of resemblance. Resemblance is an intuitive idea, (1) because the like- ness in two things compared is not in the first, nor yet in the second, as objects of sensation - T (2) because the direc- tion of the comparison must be indicated before the com- parison is made ; and (3) because each case of resemblance, though holding between specific qualities, is yet only a particular application of a general idea. The general here expounds the particular, and not the particular the general, as in generalization under the senses. § 5. A fourth intuitive idea is that of space. This has drawn much attention and been one of the centres of dis- cussion between the different schools of philosophy. Space, as immaterial and exterior to the objects of perception, can npfc be directly referred to the senses, or lost sight of in that which is furnished by them. It is not, like existence,, the very thing itself as it were, or like number, the in- separable form of it, but stands an antecedent and inde- pendent condition of the objects it contains. The deriva- tion of this idea has therefore been assiduously labored over by philosophers who accept no intuitive faculties beyond those of perception. Herbert Spencer has given this subject a statement considered highly satisfactory and conclusive by those who share his general view. We will take from his Principles of Psychology sufficient matter fairly to present his conclusions. Those who wish the entire argument by which they are supported we refer to the above work. It is impossible for us to do more than present its initial features. " Imagine that an immense number of fingers could be packed side by side, so that their ends made a flat surface ; and that each of them had a separate nervous connection with the same sensorium. If anything were laid upon the SPACE. 189 flat surface formed by those finger ends, an impression of touch could be given to a certain number of them — a num- ber great in proportion to the size of the thing. And if two things successively laid upon them, differed not only in size but in shape, there would be a difference not only in the number of the finger-ends affected, but also in the kind of combination. But now, what would be the inter- pretation of any impression thus produced, while, as yet, no experiences had been accumulated ? Would there be any idea of extension? I think not. To simplify the question, let the first object laid on these finger-ends, be a straight stick ; and let us name the two finger-ends on w T hich its extremes lie, A and Z. If now it be said that the length of the stick will be perceived, it is implied that the distance between A and Z is already known, or, in other words, that there is a pre-existent idea of a special exten- sion, which is absurd. If it be said that the extension is implied by the simultaneous excitation of B, C, D, E, F, and all the fingers between A and Z, the difficulty is not escaped; for no idea can arise from the simultaneous ex- citement of these, unless there is a knowledge of their relative positions ; which is itself a knowledge of extension. By what process, then, can the length of the stick become known ? It can become known only after the accumulation of certain experiences, by which the series and fingers between A and Z become known. If the whole mass of fingers admits of being moved bodily, as the retina does ; and if by virtue of its movements, something now touched by finger A is next touched by finger B, next by C, and so on ; and if these experiences are so multiplied by motions in all directions, that between the touching by finger A and by any other finger, the number of intermediate touches that will be felt is known ; then the distance be- tween A and Z can be known — known, that is, as a series 190 REASON. of states of consciousness produced by the successive touch- ing of the intermediate fingers — a series of states compara- ble with any other such series, and capable of being es- timated as greater or less. And when by numberless repetitions the relation between any one finger and each of the others is established, and can be represented to the mind as a series of a given length, then we may understand how a stick laid upon the surface, so as to touch all the fingers from A to Z inclusive, will be taken as equivalent to the series A to Z — how the simultaneous excitation of the entire range of fingers, will come to stand for its serial excitation — how thus, objects laid upon the surface will come to be distinguished from each other by the relative length of the series they cover, or when broad as well as long, by the groups of series which they cover — and how by habit these simultaneous excitations, from being at first known indirectly by translation into the serial ones, will come to be known directly, and the serial ones will be forgotten, just as in childhood the words of a new language, at first understood by means of their equivalents in the mother tongue, are presently understood by themselves ; and if used to the exclusion of the mother tongue, lead to the ultimate loss of it. The greatly magnified apparatus here described, being reduced to its original shape — the surface of the finger-ends being diminished to the size of the retina, the things laid upon that surface being under- stood as the images cast upon the retina, and its movements in contact with these things as the movements of the retina relatively to the images — some conception will be formed of one part of the process by which our ideas of visual extension are gained." — Pages 221-2-3. The difference between the view we wish to enforce, and that presented in this passage, lies here : Do we inter- pret the experience here detailed by a notion of space, of SPACE. 191 extension — for the one involves the other — at some instant evoked by it, or do we, at its conclusion, as its result, finally eliminate such a notion ? This may seem a slight differ- ence, yet it is a fundamental one. We give a further quo- tation in completion of the above. " How, through experi- ences of occupied extension or body, can we ever gain the notion of unoccupied extension or space ? How from the perception of a relation between resistant positions, do we progress to a perception of a relation between non-resistant positions ? If all the space attributes of body are resolvable into relations of position between subject and object, dis- closed in the act of touch — if, originally, relative position is only thus knowable — if therefore position is, to the nascent intelligence, incognizable except as the position of some- thing that produces an impression on the organism, how is it possible for the idea of position ever to be disassociated from that of body ? How can the germinal notion of empty extension ever be gained? This problem, though apparently difficult of solution, is really a very easy one. If, after some particular motion of a limb, there invariably came a sensation of softness, after some other one of roughness, after some other one of hard- ness — or if, after those movements of the eye needed for some special act of vision, there always came a sensation of redness, after some other a sensation of blueness ; and so on — it is manifest that, in conformity with the known laws of association, there would be established a constant relation between such notions and such sensations. If positions were conceived at all, they would be conceived as invariably occupied by things producing special impressions, and it would be impossible to disassociate the positions from the things. But as, in our experience, we find that a certain movement of the hand which once brought the finger in contact with something hot, now brings it in contact with 192 REASON. something sharp, and now with nothing at all ; and that a certain movement of the eye, which once was followed by the sight of a black object, is now followed by the sight of a white object, and now by the sight of no object ; it results that the idea of the particular position accompanying each one of these movements, is, by accumulated experiences, disassociated from objects and impressions, and comes to be conceived by itself ; it results that as there are endless such movements, there come to be endless such positions con- ceived as existing apart from body, and it results that, as in the first, and in every subsequent act of perception, each position is known as co-existent with the subject, there arises a consciousness of endless such co-existent positions ; that is, of space." — Pages 233-4. We find, in our criticism of this passage, fatal defects of method involved (1) in the assumption of the idea of resemblance, and (2) that also of time, the experiences con- sidered being known as serial ; also (3) in the fact that in putting these experiences on different finger-ends, Spen- cer uses a significant feature which is yet without signifi- cance in the experiment contemplated since the mind is sup- posed to be unable to note the relation involved ; and (4) in the fact that in making the simultaneous excitation stand for and represent the serial excitation, he is merely substi- tuting a section of time, already assumed, for that of space. The force of the third point will be seen, if we consider a circle of an inch between the shoulders. Touching succes- sively such a circle in various parts, we not being able in sensation to distinguish its parts, would give us no clue to the extent of the surface involved. We hold that these experiences must call forth at some point the idea of space, as the light under which compre- hension must commence and proceed, and that they can not close with a half-formed generalization waiting farther ex- SPAOE. 193 perience to grow into knowledge. Till this idea is evoked, every movement will, in its special relations, be utterly un- intelligible, provoking indeed no attention ; after it is evoked, these movements will but make it the more definite and precise in its application. Take the illustration offered by Spencer. Let a stick rest on imaginery finger-ends, by its two extremities, designated A and Z. Can that fact alone call forth the idea of space ? We think it may, pro- vided the mind is ready to know it as a fact, and to recog- nize two mutually excluding positions in sensation. It would evidently be thus interpreted at once by the adult mind, and a farther movement of the fingers would only be be sought after as giving confirmation to the fact of two mutually exclusive sensations, and as furnishing a distinct estimate of the distance between the two points. The ob- jection expressed by Spencer in the words, " If now it be said that the length of the stick will be perceived, it is im- plied that the distance between A and Z is already known ; or in other words, that there is a pre-existent idea of special extension, which is absurd," has no particular force ; for it only holds against the assertion, that the space A Z is not merely recognized as a space, but accurately known in its dimensions. This knowledge, our latest adult experience fails to give us, and certainly a general notion of some space must go before this, its careful estimate. Spencer con- founds intuitions with generalizations of the senses. In these the particular does precede the general, the sweet- ness of honey or of sugar the notion of sweetness. It is not so in intuitive ideas ; here the general precedes the par- ticular, the notion of space that of a given distance, the no- tion of time that of the time of day. Here is a fundamen- tal difference between the two, which Spencer overlooks. If the points A and Z are recognized as distinct, accord- ing to the comparison on distinct finger-ends, or in the 194 REASON. sense of sight, which these multiplied points of touch are intended to illustrate, at different parts of the retina — then this simple experience of sensations, at diverse positions excluding each other, can only find apprehension by and through the comprehending idea of space. Only under this fact of space can the phenomena occur ; only by it can they be understood for what they are, and there are no possible steps toward their solution, till this first idea is present, as an apprehension of the conditions of the problem. If the sensations are not known as in position, and in distinct posi- tions, then there is not yet the germ of suggestion, the rudiments of inquiry; if they are so known, there is al- ready present the initiatory knowledge of space. Let us suppose, with Spencer, this notion to be wanting, that we have sensations at A and Z, and at such other inter- vening points as we choose, and yet have not any sugges- tion therein of position or extension. The mind remains perfectly quiet. The sensations as sensations merely lie in consciousness, but in their space-relations no attention is directed to them, or evoked by them. Exactly the same mental state might remain when the sensations should change by becoming serial, by alternating backward and forward on successive finger-points, by furnishing in any way farther data of that exact knowledge which the first data had done nothing to call forth. The images in a mirror may lie still, or move among themselves, and in neither case is any comprehension of them made necessary to the mirror. No more would there be if the mirror were simply and permanently con- scious of them as sensations. Suppose, however, the atten- tion of the mind is awakened and directed to this move- ment. How alone can it begin to understand and explain the facts before it, except by applying the notion of space, iinow so strongly plucked at among its comprehensive sol- SPACE. 195 vents ? If " numberless repetitions " are requisite, that is, if an entire series of movements can be closed and the mind still remain without the idea, — remain quiescent, dead, mir- ror-like, holding distinct sensations in distinct special rela- tions without knowing them as distinct, reaching no judg- ment — then a second, a third, a fiftieth repetition, as mere repetition, having the light of no new idea cast upon it, may leave the mind, nay, must leave the mind, unless at some point it be awakened to a new method, as quiescent and dark as at the outset. The only ground on which any other conclusion is pos- sible is, that space is not an idea, but literally a series of sensations ; or at least a sensational fact or quality general- ized from a series of experiences, as sweetness is a quality separated clearly by repetition from other qualities, red, a color distinguished by repeated observation from other col- ors. In these cases the reiterated sensation enables us to distinguish and abstract its peculiar quality. Absurd and impossible as this view of space, that it is a quality of sensation, seems to us to be, we believe that it lurks in the arguments and statements of the sensational school. Thus, in the passage given above, it is u the serial excitations," which are identified with the notion of space, and are made by association to underlie and explain " the simultaneous excitations." In fact, however, the one set of phenomena no more requires the explanation of the idea than the other, no more contains it than the other. It is merely because there is in the first a variation of the sensa- tions, that they give or rather seem to give, a foothold to explanation not found in the second. Yet this change must be observed in the very quality of the sensations and not in the relation of the sensations, or no ground of exposition is afforded by Spencer. Kelations are intellectually seen, the qualities alone are a matter of perception. Elsewhere Spen. 196 REASON. cer speaks of the " sense of ability to move" " the sense of freedom for motions" as a constituent in our idea of space. Observe that this ability, this freedom, are not spoken of as something explained under the idea, but as a constituent of the idea. Bain says yet more explicitly : " Extension or space as a quality has no other origin and no other meaning than the association of these different sensitive and motor effects." Mark the words quality and no other meaning. Again, " The mental conception that we have of empty space is scope for movement, the possibility of potentiality of move- ment ; and this conception we derive from our experience of movements." — The Senses and the Intellect, p. 378. How is it as to the interstellar, or the intermolecular spaces ? What has experience to say concerning these ? Do we in them derive our belief of space from the changed sensations of motion ? Bain proceeds still farther. " By such steps as I have endeavored to describe, we derive our notion of extended things, of extension in the concrete. And from this we can obtain an abstract notion of the extended in the same manner as we gain any other abstract notion, as color, heat or justice." This can only be true if our knowledge of space, like our knowledge of heat and color, is a sensa- tion; and this belief, not explicitly stated, underlies logi- cally the sensualistic philosophy. The doctrine that space is a sensation or " quality " of sensations, or a series or con- catenation of sensations, or in any way an immediate prod- uct of sensation, we are willing to leave without argument to the refutation of simple statement. It would thus sink wholly from the intellectual field, and, if allowed to drag other kindred ideas with it, would leave neither occasion nor opportunity for any other faculty than that of percep- tion. Sensations lie together, and need no conjunction by the judgment ; and as for any notions wherewith the mind SPACE. 197 is to comprehend and classify them, there are none ; those thought to be such, are themselves sensations. Feel your way, feel on and feel ever, would be the comprehensive di- rection to a being — we can scarcely say mind, for the mind is now resolved into mere sensibility — so formed. Feel space, feel time, feel number, and look to your finger-ends for liberty and right, or eternally lose them. Let us carefully guard against one point of misappre- hension. We say nothing as to any definite time in the progress of the infant in which the idea of space will arise. Sensations as sensations may come and go, we know not how long, without evoking the idea; but when it does come, it will come at once from within ; not in an abstract, discriminated form, but in a concrete, obscure application, and prepare the way for a new series of intellectual actions. All precise estimates and measurements are, of course, the sole fruit of experience, and give the infant mind abun- dant occupation under this regulative idea. One may study geometry with little or no abstract con- sideration of space as space, yet the idea is tacitly present everywhere. The child may come to a knowledge of the position and dimension of its own members, with no direc- tion of the mind to the notion of space as such, though that idea quietly informs the whole process. In the second of the longer quotations above given from Spencer, we have the notion of space, empty space, de- rived from a vacant organ of sense. Direct the mind steadily to this point. An organ, as the finger-end, or the eye, with no content of sensation in it, a simple blank, is one thing ; and this fact accounted for and explained to the mind by the idea of empty space is quite another. If the first, generalized in any way he pleases, is Spencer's idea of space, then that idea consists in the mere absence of sensation, and should exist in the highest degree in con- 198 REASON. nection with paralyzed organs. A recognition of blindness, or even deafness, should be one of space. If, however, the fact of a vacant organ becomes significant only in connection with a process of mind, we wish to know under what guiding clue that process proceeds. What is brought to the explanation of the fact of motion without sensation ? It seems to us that but one answer can be given— space. This conception as absolutely simple must come, when it comes, in (1) a complete form ; coming completely, it must come (2) suddenly ; and, (3) till it comes, no experience is intelligible under it. Spencer, with the marked approval of Bain, makes, in another phase of the argument, the notion of space depend- ent on co-existence, and co-existence the fruit of experience. " Not only is it that the idea of space involves the idea of co- existence, but it is that the idea of co-existence in- volves the idea of space. Fundamentally space and co- existence are two sides of the same cognition." " On the one hand space can not be thought of without co-existent positions being thought of; on the other hand co-existence can not be thought of without at least two points in space being thought of. A relation of co-existence implies two somethings that co-exist. Two somethings can not occupy absolutely the same point in space. And hence co-existence implies space. Space can be known only as presenting relations of co-existence ; relations of co-exist- ence can be known only as presented in space." " If now it should turn out under an ultimate analysis — that a relation of co-existence is not directly cognizable, but is cognizable only by a duplex act of thought — only by a comparison of experiences ; the question between the trans- cendentalists and their opponents will be set finally at rest. When after it has been shown as above, that our cognition of space in its totality is explicable upon the experience hy- SPACE. 199 pothesis, and that all the peculiarities of the cognition cor- respond to that hypothesis, it comes to be shown that the ultimate elements into which that cognition is decomposable — the relation of co-existence — can itself be gained only by experience — the utter untenableness of the Kantian doc- trine will become manifest." — Pages 243-4. Herein our author hardly agrees with himself, having insisted that the co-existent points, A, B, Z, can not give the idea of extension, though it now turns out that a knowl- edge of their co-existence would have been essentially a knowledge of space. We believe (1) that the notion of any simple position involves that of space, is explained under it, and therefore that a single sensation of touch, complex indeed, yet regarded as simple, might call forth the idea. This we do not care to dwell on, as it is doubtless in con- nection with many simultaneous and serial phenomena, presented in several senses, that the notion actually does arise. We can not accept the statement that the ultimate element into which the cognition, space, is decomposable, is co-existence. On the other hand, (2) the notion of ex- ternal, material co-existence is subsequent in the order of thought to that of space. Nor are the two by any means the same. I may have the idea of empty space'. I may put one object in it, or two or three objects in it, but the idea of space has preceded each and all before they became to me external objects, or the images of such objects. Indeed, (3) simple co-existence, as of an act of memory and a thought, of a thought and a feeling, does not involve the idea of space. The contrast of the inner and the outer, of the ego and non-ego, may or may not go forward ; but the first step in such a contrast, the initial stroke of light in handling a local sensation, is the localizing idea of space. How often, and how long I may have one, two, three sen- sations, and not expound them, is simply the question, How 200 REASON. long do the senses ante-date in development the other intellectual powers ? When these come, they come thus, not otherwise. The fact of co-existence is a mere blind datum of sensations, until contemplated under the idea of space. The (4) actual co-existence of two things is not involved in space, but only its possibility. The extension of space as the possibility of such a co-existence is the no- tion of space, is in and of the very idea. Actual co-exist- ence alone rests on sensation, the possibility of it on the intuition. Mr. Spencer is not to think and speak of the co- existence of two positions as if it were identical with the co-existence of things. The first is in no w r ay a datum of sensation. If he tries to make it so, he is thrown immed- iately back on to his former proof, and loses his present foothold. We have made no distinction between extension and space. We regard the first only as a specification under the second. The extension of particular objects, and the duration of particular events, are forms under which the mind applies the intuitive ideas of space and of time. A knowledge of actual spaces, a measurement of material objects, are the fruits of experience ; but these estimates proceed always under the prior notion of space, which makes them intelligible. Space, in its analytical contemplation, furnishes a variety of intuitive conceptions which are the basis of the demon- strative reasonings of Geometry and Trigonometry. Such a notion is position, a line, a surface, perfect curves, figures, solids. A circle, in its accurate form as a ground of demon- strative truth, is an intuitive conception, as are the proposi- tions which flow from the immutable relation of its parts, and of the lines which define, and are defined by them. A surface without thickness, a line without breadth, a point without dimensions, are all intuitive conceptions under the CONSCIOUSNESS. 201 primitive idea, and are the elements of a purely intuitive science. The most marked of those secondary conceptions is that of position. It is to be entirely distinguished from an infinitesimal body ; as the infinite of the metaphysician wholly transcends the infinite of the mathematician. Posi- tion is not arrived at by the futile sub- divisions of the fancy, is not the result of the dogma of divisibility. There is here absolutely no length nor breadth, and the idea is reached directly by the grasp of the reason. The imagina- tion may falter in struggling by additions to reach the in- finite, and by subtractions to arrive at pure position ; but the reason easily and at once accepts both notions, and rids them of those measurable parts by which the imagination baffles itself in the pursuit. Position is absolutely without measurements, and hence without parts ; the infinite is ab- solutely beyond measurements, and hence also without parts. There is no whole, therefore no division of that whole. § 6. The ideas of existence, number and resemblance belong equally to physical and mental facts. Space, on the the other hand, is the peculiar formative idea of physical phenomena alone ; while the idea we have now to offer, that of consciousness, is the exclusive characteristic of mental experiences. There has been no debate concerning this idea, because it has not been presented as belonging to this de- partment of our intellectual furniture. If, however, it shall appear that consciousness exists as a form rather than as a substance or quality, that it is therefore directly arrived at by the mind, and also that it furnishes the distinctive feat- ure of a class of phenomena, the transcendent predicate of a series of judgments, it will be plain that it belongs properly to the class of regulative notions. Consciousness is often spoken of as if it were a faculty, a form of knowing ; yet a little thought at once shows that it is not. I see a ball. I say in farther enforcement, I know that I see it. This Ian- 202 REASON. guage lias divided the first simple act into two, an act of per- ceiving, and one of knowing directed toward that of percep- tion. Yet this is merely a convenience of expression. The one single act of seeing the ball is all that is present. If there were a second act of knowing, this also would require sub-division in order to reach the element of consciousness in it. Thus analysis must go on indefinitely, unless we finally accept an act of knowing which is simple and indi- visible. \There is no double faculty, or double movement of one faculty, in thinking, feeling, willing.\ A thought is a thought only as it is known ; a feeling is a feeling only as it is felt. They do not first find existence, and then an added quality or element of consciousness ; but conscious- ness is the condition and form of their existence. Con- sciousness, then, is not, like judgment, a power; nor like pain or pleasure, a quality of certain states ; it is not a feat- ure or a relation of a sensation, but involved in the very notion of a sensation. This idea, therefore, as neither a faculty to be known by its exercise, nor a quality of mental states to be learned by observation, — indeed every act of observation must itself contain it — must be evolved by the mind as an explanatory idea, or conditional notion in con- sidering the phenomena to which it is applicable. It is not' only unphenomenal itself, it is introduced as the antecedent condition of a large class of phenomena, to wit : those of mind. What space is to material facts, conscious- ness is to intellectual facts, the interpreting light under which they occur. The words we constantly apply to it, recognize this relation. We say, " the field of conscious- ness," " transpiring in consciousness," " coming up into the light of consciousness," " the flow of consciousness," — that is of thought and feeling in consciousness. These and like expressions are shaped under an image in which conscious- ness is presented as an arena of mental movements, as is CONSCIOUSNESS. 203 space of physical events. The peculiar nature of knowing, feeling, willing, is not understood till the idea of conscious- ness is present : yet these facts remain in their integrity possessed of all the elements that analysis discloses in them, without accrediting any one quality to consciousness. Con- sciousness thus shows itself to be to the inner, invisible world, what space is to the outer, visible one ; the condition of its existence, the only canvas on which its colors can ap- pear. To occupy space is to have physical existence, to occupy consciousness is to have an intellectual existence, to occupy neither is not to exist, is to present no one of the known forms of existence. The idea is seen to be regula- tive in the large class of propositions which arise under it. I know ; I see the book ; I feel the pain, are of this sort. Each of them is comprehended by virtue of the notion of consciousness, which expounds their several predicates. This view also finds support in the difficulties which at- tend on the ordinary explanations of consciousness. What is it ? is a question that has greatly perplexed philosophy, and has seldom received a very definite answer. Some have striven to conceive it as a faculty, yet this faculty must be present in the action of every other faculty, and that other faculty would be absolutely null and void with- out this. To divide an act of knowing into one of knowing, and one of consciousness, each taking a distinct moiety, is impossible. Hamilton has said, " Consciousness is the ge- nus under which our several faculties of knowing are con- tained as species." But our faculties of knowing, no more require it than those of feeling and willing; and what exactly is a genus in distinction from the species it con- tains ? Nothing but a word. Certainly an effort to make definite this view, prepares the way for regarding conscious- ness as a general idea, under which all specific acts of mind, in themselves complete, find recognition. J. J. Murphy, 204: REASON. in his work on Habit and Intelligence, has united con- sciousness to feeling, and made it and sensation species under that generic term. The conclusion is on the oppo- site side to that of Hamilton, but no better than his ; since consciousness belongs equally to thoughts and feelings. Others figure consciousness under the image of an internal light. This is virtually to decline the inquiry, What is it ? since the illustration can reflect no explanation on this point. Others speak of consciousness as that power by which we refer acts to ourselves. This is to let the eye wander altogether from the subject, since the explanation overlooks the facts to be explained. Consciousness has also been divided into common consciousness and philosophical consciousness, and into consciousness and self -consciousness. These are divisions which pertain to the phenomena in consciousness and not to consciousness itself. It occasions confusion in some minds that consciousness should be spoken of as an intuitive idea, when it is ob- viously something more and other than an idea. The lan- guage merely indicates the manner in which the mind arrives at the relation expressed by consciousness. In the same way we speak of space as an idea, and of a landscape as a perception, and of a general term as a conception. A man may have an idea of himself, that is be an idea to himself ; yet he is something more than an idea. The recognition of consciousness as a distinct category cuts apart physical and mental facts with the deepest possible division. They lie in two incommensurable and incomparable realms, that can never overlap each other. § 7. We now pass to time, a regulative idea, like that of space, which has attracted much attention as obviously open to a super-sensual reference. It is the idea which unites all events, whether physical or mental. The sensations occasioned by phenomena into which the idea of time most TIME. 205 obviously enters are diverse in their relations from those chiefly suggestive of space : or rather, things are viewed in distinct bearings in the application of the one or the other notion. In each case, nevertheless, the diversity is only understood by an a priori recognition of the controll- ing idea of the relation under which it arises. Some objects can be contemplated indifferently in one order of succession or in a reverse order. We may move from A to Z, or re- turn from Z to A. Others, transpiring in time, confine the attention to one direction. We pass from A to Z, but can- not retrace our steps. The cars enter the field of vision at the left, and pass out at the right. In these facts there is an occasion, though not an explanation, of the notion of time. The mind cannot, under the influence of a mere se- ries of sensations, discover this relation ; since it is not in and of the sensations, but that which expounds them. Eor can it institute a comparison between the two relations of objects which shall issue in any comprehension of them, without itself supplying the essential conditions of that comparison — the notions of space and time. We must either hold that time is an order of sensation, or we must ad- mit it to be that transcendent idea which expounds that or- der, and is therefore supplied by the mind. Says Spencer, " As the ideas of space and co-existence are inseparable, so also are the ideas of time and sequence. It is impossible to think of time, without thinking of some succession ; and it is equally impossible to think of any suc- cession without thinking of time. Time, like space, can not be conceived except by the establishment of a relation between at least two elements of consciousness, the differ- ence being, that while in the case of space, those two ele- ments are, or seem to be present together, in the case of time they are not present together." — Principles of Psy- chology, page 247. 206 REASON. This statement, so far as it is admissible at all, is so as a statement of the circumstances under which the idea of time arises, and not of the nature of that idea itself. Used for the latter purpose, the author legitimately reaches the conclusion that time is " relativity of position among the states of consciousness." The process of arriviug at this result is farther explained thus : " Gradually, as by the accumulation of experiences, there are found to be like and unlike sounds, tastes, smells, sizes, forms, textures ; the re- lationship which we signify by these words, like and unlike, will be more and more dissociated from particular impres- sions ; and the abstract ideas of likeness and unlikeness will come into existence. Manifestly, then, the ideas of likeness and unlikeness are impossible until multitudes of things have been thought of as like and unlike. Similarly in the case before us. After various relations of position among the states of consciousness have been contemplated, have been compared, have become familiar ; and after ex- periences of different relations of position have been so ac- cumulated as to dissociate the idea of the relation from all particular positions ; then, and not till then, can there arise the abstract notion of relativity of position among the states of consciousness — the notion of time. Thus so far is it from being true that time, as conceived by us, is a form of thought ; it turns out contrariwise, not only that there can be thoughts while yet time has not been conceived, but that there must be thoughts, before it can become conceivable " — Page 252. Our objection to the above conclusion is double. The comparison itself cannot go on without a regulative idea, that of resemblance, under which it can be instituted ; and that in which it is said to issue is not the notion of time. That which is explained by time is very different from time itself. If the first were the second we should have no need TIME. 207 of an independent explanatory notion, the phenomena would be complete and intelligible in themselves. The sequence of events provokes the notion, but is not that notion. Se- quence and time do not mutually contain each other, but (1) time is that idea without which the fact of sequence is unintelligible. That time is not identical with succession is seen in our measurement of it. (2) A succession of events may be completed in a shorter or longer period, and if time to us were their mere relation in sequence, we should insist on its identity in the two cases. (3) We dis- tinguish time from any given sequence, indeed from all se- quences, longer or shorter according to the forces at work. "We do not identify it with that series of events even by which we measure it. The conditions for its exact estimate and general apprehension are different. (4) The notion of time, with no actual events transpiring in it, is quite con- sonant with thought. Moreover, many sequences are si- multaneous. (5) The relativity of which one of these is it that constitutes time ? It cannot be one to the exclusion of the remainder, for no one has such a pre-eminence over every other. Neither can it be all, since they are constantly varying among themselves. What effect has it on time, that one drives faster than he has been driving, that a rail- road train has stopped at a station, that the thoughts have been quickend by danger ? The quality of sweetness may exist in many things, and have shades of diversity in each ; is this also our conception of time ? (6) The prior notion of time, moreover, imposes se- quence, when there is no sequence in the states of conscious- ness, but rather alternation. The mind may pass from A to B in contemplation, and back again; it may vibrate between the two in alternate thought : yet it does this as certainly under the idea of time as if it had simply passed on to Z. Motion in a circle is felt to be motion as much 208 REASON. as movement in a straight line. Bare contemplation with- out conscious progress is felt to occupy time ; it is for the measure, not for the fact of time, that we revert to external events at the expiration or change of a single absorbing feeling. There is doubtless some succession in every phase of mind, but it is not necessary for us to contemplate these minor and obscure transitions to be aware that every act, the very act of attention, occupies time. We might as well endure an intense, absorbing pain for an hour as for an instant, if we were not able to distinguish between the two cases. That which we urge is, that the notion of time imposes the sense of sequence where there is no proper sequence in the sensations as sensations, and the alternate consideration of A, B, like the beat of a clock, marks dis- tinctly the flow of time. Indeed all consciousness is made sequential, no matter what the order of its states, by the very notion of time in which they transpire. We can not escape the inner succession of impressions, because we can- not elude the interpreting idea, that of time. The position of " states of consciousness " can be only that of succession, whatever their character or the number of times they are repeated. The inner law overrules the outward appear- ance, and imposes the notion of sequence. Suppose, on the other hand, with Spencer, that we could pass from A to Z, without the idea of time. In that case we should not only be destitute of it, but have made no progress towards it. We should simply have experienced sensations without interpretation. No repetition of this process, however frequent, could make it fruitful of a new notion. The simple idea must be present to open the inquiry. Time must be a sensation, like that of green and red, or its distinct abstraction can not follow from repeti- tion. The sensation green is given in each particular instance, and then, by distinguishing attention, assumes the TIME. 209 abstract form. This process is possible only to sensations, — and even then involves more than sensation — not to rela- tions ; since a relation is addressed to the intellect, and not to the sense, and can only be understood in connection with an idea under which it arises and is defined as a rela- tion of place, time, dependence. On no supposition is the closing statement of Spencer admissible. " So far is it from being true that Time, as conceived by us, is a form of thought, it turns out contrariwise, not only that there can be thoughts while yet time has not been conceived, but that there must be thoughts before it can become con- ceivable." As a sensation, time must be experienced in each sensa- tion from which it is to be abstracted ; as a relation also it must be discoverable in each series or it can not be general- ized from all ; and as an idea disclosing a relation it must come at once. Time must be a sensation, or it must be a specific relation under some general idea, or it must itself be a primary idea, the condition of actual, individual connections. The first supposition is plainly false, while the second is as unacceptable to the empirical school as the third, since it also implies original, intuitive action of the mind. Yet I see no escape except in the assertion that a relation of no specific order or kind can be discov- ered, and this is not an escape, since such a relation could not be generalized into one of a specific order or kind, to wit, that of time. § 8. The seventh regulative idea is that of cause and, effect. This is one of the most undeniable of them all, and! is either greatly restricted in its statement, or entirely re- jected by those who refuse to accept the reason as a source' of knowledge. Indeed an adequate presentation of the notion as it lies in the general mind, shows it at once to be beyond sensation, generalization, or any action that these, 210 REASON. processes can verify. The convenience of expression has led to the extension of the term cause, not merely to re- mote agents, but even to the conditions of their action. Any one of all the circumstances necessary to an effect is spoken of as its cause, though no direct efficiency proceeds from it. In a stricter sense, the word cause includes only those antecedents which are active in the effect, and in a yet closer sense, the sense which belongs to it in the present discussion, the forces immediately operative in the fact before us. The cause is strictly contemporaneous with the effect, underlies it, momentarily occasions it. The antecedent effect had its antecedent cause, and though this cause may have been identical with the cause now operat- ing, it remains a cause by virtue of its present activity. The effect is the immediate evidence of the cause ; and though the last is prior in thought to the first, neither can exist an instant without the other. The sound of the steam-whistle is remotely attributable to the distant loco- motive, is more immediately to be referred to the move- ment of the air and the tympanum, but finds its causes exactly in the forces which sustain the movement, and the living powers which receive and interpret it. In this sense the cause is always and necessarily transcendental, out of the range of the senses, incapable of verification by any other than the very faculty which in the first instance yields the idea. The statements of empirical philosophy are quite differ- ent from those now made. Says Bain, " The successions designated as Cause and Effect, are fixed in the mind by contiguity. Belief in external reality is anticipation of a given effect of a given antecedent; and the effects and causes are our own various sensations and movements." More clearly still does Mill speak of the notion as one of simple antecedence ; while Spencer treats of it under the CA USE AND EFFECT. 211 caption, " The relation of sequences." If these and other kindred statements are correct, then there is no veritable idea of cause and effect in the precise, intuitive sense, since a sequence finds explanation under the notion of time, and requires for its statement no other form of thought. That there is any sufficient knowledge or idea of the ground of such a sequence is simply denied by this class of philoso- phers. There is in this attitude an abandonment of the idea of causation as irresolvable into experience, and a substitution for it of a certain application of the notion of time. The ground of debate, therefore, is narrowed down to the correctness with which the phenomena under dis- cussion are stated by the respective parties. If it be shown, that simple sequence does not, in the common mind, cover the entire ground of causation, there is in the empirical philosophy a .surrender of one actual, universal idea as inex- plicable ; a superficial substitution for it of the fragment of another, and this for no other reason than that its own theory can find no place for this fixed conviction. It is the facts of the mind's action that philosophy inquires into, and the above proof being given, there will here be left a form of action as universal and persistent, and hence as ul- timate and authoritative as any, uncovered by materialism. The universal conviction, if it can be arrived at, is not to be pushed aside, to be left unexplained, nor to be re- garded as an accidental, invalid movement of mind, with- out a reason rendered, distinguishing this from kindred affirmations of our faculties. We may not impeach the ac- tion of the mind at one point, without at least separating this point broadly and decisively from every other. These universal convictions are not to be obnoxious, merely because, as otherwise inexplicable, they demand the intui- tive insight claimed for them. There is here really a priori conviction brought to disprove a priori truths ; for what 212 REASON. but an a priori bias of mind, is this antecedent reluctance to admit the possibility of regulative ideas % What then is the fact? Which statement best con- forms to the universal conviction, that of fixed antecedence, or of present underlying power ? There can be no doubt on this point. The case is a plain, almost an admitted one, against empirical philosophy. Language is full of this notion of inherent, sub-phenomenal connections between events. The word force distinctly expresses this causal link, and few words are more familiar, or play a more im- portant part in speech. Of the same kind, are the words power, influence, energy, strength, and more or less markedly most of the words which express physical action. Pull, pu'sh, press, pry, lift, lug, labor, the entire vocabulary of effort are saturated with this causal notion of an invisible efficiency, which expends itself in all forms of activity. Behold any striking display of force, the blasting of rocks, and every mind is impressed with the power of the invisi- ble agent. To look upon the lifting of detached masses, the seaming of the solid bed, as a mere sequence of dis- connected events, is impossible to any mind, in its sponta- neous action. No descriptive language was ever applied to such events, that regarded them simply as a sequence. The popular and the universal conviction is unmistakable, that here is force, invisible power. Equally present is the idea to all science. Gravitation, cohesion, chemical affinity, the correlation of forces, the various theories of physical facts, like Darwin's theory of gemmules, or Spencer's of physiological units, involve the notion of inherent power, working the results under con- sideration. Science could not carry forward its investiga- tions without this recognition of force. To discover the traces of its presence, and the lines of its action, is the con- stant triumph of knowledge. To confound fixed antecedents CA USE AND EFFECT. 213 with efficient force is impossible to successful inquiry. The shadow of an object approaching us from the light, would thus be its cause ; the effervescence of lime and water, the cause of the heat; the dissolving of salt in the water, the cause of the cold. The first fact, in each series of associ- ated effects, would be the source of the remainder. No se- quences are more fixed than those of day and night, summer and winter, yet there is no direct, causal connection be tween them, and no one ever so conceives the dependence. Philosophy likewise reverts constantly to insensible, un- approachable causes. A large share of philosophers admit their existence and the grounds of it ; and those of them who through their denial of the latter are content to sacrifice the former do not, and can not, use language, except in a f ew guarded passages, consistent with their own statements. They must, with Bain, in each inadvertent moment, speak of " active energy," of " mechanical powers," of " rousing the dormant energy," and to deny themselves these and kindred expressions, to forego the ideas back of them, would be to take away the opportunity of composition, or to make language most cumbersome, and untrue to our con- victions. The generally accepted dogma, that the mind can not know anything beyond its own modifications; a dogma insisted on by many of the empirical school, finds its ultimate support in this notion of cause and effect. The existence of the object perceived outside of the perceptive organ, independent of it and removed from it — at least by insensible distances — has determined the large majority of philosophers to deny the possibility of direct perception. If, however, the connection of cause and effect is one of antecedence merely, then this separation of the object per- ceived from the organ perceiving it, should oppose no obstacle whatever to direct perception. A fixed sequence can be established between things remote and wholly un- 214: BEA80N. like, as easily as between things, like and occupying com- mon ground. If, therefore, this connection of sequence is the deepest, nay the only connection between things thought to act on each other, it would seem to suffice for knowledge at all distances ; or if not, to make knowledge impossible. How shall even successive states of mind lie fruitfully to- gether in simple sequence, if sequence after all is a barren connection. If it fails to unite remote, how can it unite proximate objects ? If one set falls apart, all must. The general point is too plain for farther statement. Evidently the doctrine of simple antecedence does not ex- press the universal conviction, does not cover the phenom- ena under explanation, does not accept and expound the affirmations of knowledge which every mind is constantly making. Quite a different explanation of cause and effect has come from another quarter. Sir William Hamilton applies to it what he terms the law of the conditioned. The no- tion of causality is thought by him to arise from the weak- ness of the mind, its inability to conceive a beginning. The mind, he affirms is unable to conceive events without a be- ginning, nor yet with a beginning. "We can conceive neither the absolute commencement, nor the absolute ter- mination of anything that is once thought to exist ; nor any more the opposite alternative of infinite non-commence- ment, of infinite non-termination." Herein is given the the principle of causality : " When an object is presented phenomenally as commencing, we cannot but suppose that the complement of existence, which it now contains, has previously been ; in other words, that all that we at present come to know in it as an effect, must previously have ex- isted in its cause." This is a most inadequate explanation for several reasons. In the first place, it inverts the order of dependence iu our mental action. We can not conceive CA USE AND EFFECT. 215 of anything as absolutely commencing, because of this no- tion of cause and effect. The existence of the notion is the ground of our embarrassment, not the embarrassment the occasion of the notion. What would be simpler, were it not for causality crowding us backward, than merely to conceive any landscape, any personage, any event, with no thought of what has preceded it \ The present act of the imagination is not conditioned on the past, neither should we be compelled to evoke the present from the past, any more than to carry it forward into the future, were it not for causation. Any cross-section of the events of time could be as complete as a single pebble on the shore. Thus, often in dreams, when the imagination finds un- restrained phenomenal play, the judgment not being suffi- ciently active to impose the check of this purely sub-pheno- menal idea of causation, we in a great measure disregard it, and, with no sense of jar, suffer the unexplained presence of unexpected persons, and an incongruous order and issue of events. The- ideas regulative of space and time relations are present, while cause and effect, regulative of consecutive thought, is in whole or in part overlooked. A city makes its appearance suddenly, the ship moves unobstructedly across the land, the facts and figures of fancy come and go freely, bound to no ordinary sequence. A second objection is found in the fact that the theory affords no explanation of the alternative adopted by it. We can neither conceive, it is said, the commencement, nor the non-commencement of anything. Yery well, but how is this dilemma to be escaped by the present notion of cau- sation. The conclusion accepted under it of "infinite non- commencement" remains as inconceivable as ever, and, there- fore, as far as conception is concerned, presents as many dif- ficulties as would the opposed alternative of an immediate, independent beginning of events. If the mind is as ojDen 216 REASON. to one cf these conclusions as to the other, and can properly be satisfied with neither, what reason has it for preferring one to the other ? The difficulty is met only when causa- tion is made a positive notion, compelling us in the one di- rection. Accompany this acceptance with a denial of our right to direct the imagination in explanation to that which, according to our very notion of it, is sub-phenomenal, and we have at once the ability and the inability of the mind explained. We have a reason for its convictions, and also for their inconceivable character. Again, this theory is a concealed theory of antecedence, and fails to cover the strict idea of causation. The efficient cause is present with the effect, immediately underlies it, and sustains it. Thus the substance, the force, which con- stitutes matter, each instant gives occasion to its qualities. The power which is in mind, is the groundwork and source of its thoughts, feelings and volitions. The stream of causation flows under the stream of events, and moment- arily floats them, as the surface of the ocean is supported by its invisible depths. Simply to insist on an antecedent event to every event, is to throw up the phenomenal path along which the imagination travels, but is not a recogni- tion of the true force and nature of causation. The- imagi- nation, exploring the past, does indeed require that distinct, tangible foot-stones should, in due order, link its steps ; but that which impels the mind in thus sending it to search its way backward, is a sense of an unbroken series of causes, and that which the mind finds everywhere beneath the phe- nomenal supports of the imagination is the permanent power and flow of forces. This theory of the weakness of the human mind signally fails to account for so positive and pervasive a notion as this of cause and effect. If we accept this notion in its full, universal applica- tion, leading us to those invisible energies which thread to- CAUSE AND EFFECT. 217 gether the phenomena of the universe ; if we do not deny or limit the facts presented to us in our own spontaneous beliefs, in universal action and universal language, it is at once evident, that this idea must have an intuitive origin. Admittedly it transcends all experience, is wholly unap- proachable by the senses. The presence of such a notion, evinced by language, by science, by philosophy, by our spontaneous and inevitable interpretation of events, is un- deniable ; to discard it as a purely fanciful notion superin- duced on the facts, is to deny and not to explain the phe- nomena of the mind ; is to construct our theories in neglect of facts too broad for them ; is to invalidate an action of mind, as universal, as strong in the confidence and spontane- ous trust of the mind itself, as any of its processes. This is to make our several forms of activity, intuitive and rationative, contradictory and self -destructive ; is to bring one form of knowing from its own field into that of another faculty, and, because it fails to understand the diverse action of a power given on purpose to do a work different from its own, to expel, as fictitious and fanciful, conclusions wrought out by a native force of mind. Daily life pursues its hourly labors ; natural science accomplishes its great achievements, following the clue of causation; and yet a speculation termed philosophy steps in to declare the light under which these processes proceed, false and deceptive. It seems to be light, and does marvelously well the work of light, and all men insist on using it as light ; yet evidently it is not the waxen taper we are after. This, then, can not be our pre- determined light, and as there is no other, it follows plainly that this is not light, but darkness rather. Correct ideas must come from experience, and be capable of its verifica- tion ; this is not so reached, and cannot be so explained ; therefore it is no valid notion. In all this there is a fla- grant begging of the question. We thus put the grounds 218 REASON. and tests of validity in the faculties that directly concern experience, and then deny validity to ideas that must con- fessedly, if they exist at all, transcend experience and the judgments which unfold it. The notion of causation is well fitted to be a test of the two schools of philosophy, because of its plainly transcen- dental character, and because knowledge can not be gained without it. Deny causation and we deny reasons of all sorts. Mere continuity of impressions in the mind can not be the ground of any expectation, unless this succession tends to repeat itself, that is unless there is in it some causal energy. All thoughts equally with all things must fall apart, if there is no coherence between them ; apparent order would be no more significant than apparent disorder, as both would be accidental. A cause is a transcendental, co-existent, co-extensive force, underlying its effect. The effect is its only proof, its only measure, its only expression. The cause is neither less nor more than the effect ; no portion of a cause can exist without an effect ; no portion of an effect can exist without a cause % "While the word in its most explicit use is equivalent to the noumenon underlying the phenomenon, and is the source of the conviction of real being, the nou- mena and phenomena, at any moment present, imply pre- vious facts, facts from which they have sprung. This second application of the notion is of quite as much interest as the first, since events and our control of events turn upon it. Valid being is the statical force of the idea, and evo- lution its dynamical force. Every section of a river is referred to the one next above it. Language suits itself to the convenience of use as certainly as to the facts expressed. In speaking of a river, we divide it, though it is in itself continuous, into portions easily designated by changes in its banks, or by CA USE AND EFFECT. 219 features which interest us in the stream itself. Thus we refer to the bend, and the stretch above the bend ; or the rapids and the level below the rapids. In like manner in the coherent flow of events, we select the causes which in- terest us, and we name them by the effects through which they are expressed. Thus the death of a man is said to be due to the premature explosion of a charge, and this to the hasty firing of the fuse, and this to a reckless temper. Thought lays hold of those causes and conditions which are of moment in reference to the control of events. The mind is, therefore, even more likely to be attracted by sec- ondary and variable conditions than by the fixed lines of forces, since it is through these conditions chiefly that it modifies events. An engine on a railroad fires the barns along its track, it is complained of as being constructed without a sufficient defense against scattering sparks. The defect is said to be the cause of the conflagrations. Causes are divided into efficient causes, conditional causes and final causes. Strictly efficient causes are the forces which, interesting each other in the effect, occasion it ; or these forces in some previous form. Thus the flying of the rocks is due to the expansive force of the gases im- prisoned in the charge. The gunpowder is an efficient cause. The food which a workingman has eaten, and the weight and temper of his axe, are efficient causes in the felling of a tree. Evidently many lines of force may unite in an effect. Conditional causes are those things, actions, relations, which, without constituting a part of the efficient forces, determine the presence of these forces in the effect in an efficient form. Conditional causes are of quite as much practical interest as efficient causes, and may not be easily distinguishable from them. Drilling the hole, charging it and firing the fuse are conditional causes of blasting the rock. Final causes are not causes. We mean by a final cause 220 REASON. the motive which prompts an action. Thus the final cause of removing the rocks is the construction of a railroad. The doctrine of intuitive ideas is often damaged by its advocates. It is asserted that consciousness testifies to much not referable to this source. The direct matter of consciousnesss, sensations, thoughts, feelings, volitions, are undeniable. There is no ground for dispute when any fact is a product of consciousness, that is, belongs to men- tal states. It is quite a different question, What is involved in the data of consciousness ? The validity of our judg- ments, the ideas under which they proceed are to be arrived at by analysis, by reasoning, and are not directly vouched for by consciousness. It is sometimes said we are conscious of force, and therefore of a cause in putting forth voluntary effort. The true statement would rather seem to be, we are conscious of volition, and of the subsequent sensations which accom- pany action, but not at all of the hidden link of power which unites them. Indeed, it is not always possible for us to tell whether the intended muscular result will follow the volition. Some paralysis may have intervened, arrest- ing the flow of energy, and the interior connection lies so wholly beyond consciousness, that we can only determine the presence or absence of suitable muscular conditions by a tentative effort at movement. If we were conscious of force, force itself would be phenomenal, and lose its sub- phenomenal character. It would cease to be a causal idea and would become a sensible fact or effect. The simplest statement of causation is, Every effect must have a cause. In this is involved the expectation of the perpetuity of nature, since every change in the effect, as itself an effect, would demand a new, specific cause. With no apparent change in causes, we anticipate previous results, since these must follow from the unchanged forms SPONTANEITY. 221 and conditions of action. A prolonged duration of the present physical system is expected by us, unless we see, or think we see, reasons for change in the government of God, or grounds of change in the system itself — an intro- duction at some point of new forces. § 9. Causation is the law of connection between phys- ical facts ; spontaneity, the idea we have next to offer, is the ground of connection between purely mental facts. These two sets of facts are as distinct in their dependencies as in the fields in which they occur. Confusion may easily arise from blending physical and mental phenomena in the nervous system which lies as a medium of interaction be- tween them. All physical facts and facts immediately dependent on them come under the law of causation. Causation involves the presence of forces, definite in quality and quantity, which, in transition from form to form, give rise to events. These forces have always some mode of existence which exactly expresses them. A power of mind can not so be conceived. It is not a realized patent or latent force of a given order, waiting on a proper occasion to pass through another form termed thought. The law of thought, that of truth, can not be maintained on such a supposition. Truth would become a simple effect, and lose its very nature. The mind must, as a power, seek, discern, and embrace the truth, otherwise it is not truth. This search implies spontaneity on the part of the mind, a self-guided power that moves toward its own ends. The laws of thought, of feeling, of volition do not abide as forces in this power, but are the product of the rational insight involved in the power itself. Language constantly regards this distinction, but not; with the firmness it might. Over against forces should stand powers; against effects, actions ; against causes, incentives and motives ; against instruments, agents. 222 REASON. While spontaneity is the cardinal fact of mind, so much so as to be essential to any apprehension of mind as mind, it has more frequently been discussed as liberty. Proof of the actual possession of liberty by man as a voluntary agent, and a precise statement of what is involved there- in, will be presented later. Liberty is to be distinguished on the one side from those necessary connections which are causal in character, and on the other from chance, the denial of all dependence on antecedents. Indeed, strictly construed, there can be no chance events. The positive notions of causation and liberty, which cover the entire phenomenal field, do not permit them. It is only under a qualified form, as events with unknown or incalculable causes, that chance ever appears in the field of facts. Liberty allows the influence of motives, but not their measured, definite, irresistible influence. We admit and deny in the same instant the application of the word in- fluence, admit the word in its substance, deny it in the form which its connection with causal events has given it. Herein is the peculiar and primitive character of the con- ception, that of a connection which is not necessary, of persuasion which is not imperative in either branch of the alternative, of influence which does not push with a fixed, determinative force towards a given volition. The will is neither capricious, nor mathematically calculable in its action. It is free, and submits freely, so far as it submits to the motives before it. There is no great difficulty in this conception so long as we let it alone. It is when we begin to compare it with other conceptions, that its pecu- liarity appears, and this we are liable to mistake for intrin- sic absurdity, falsity. This idea of liberty — the motives lying before the will, not back of it ; persuading, not impelling it — is primitive, and brought by the mind to the explanation of a class SPONTANEITY. 223 of facts that require it, those of choice and responsibility. The sense of obligation, with the subsequent feelings of virtue and guilt, of approval and condemnation ; the facts of government, of reward and punishment, the mind can not understand or fully accept without the interpretation of the idea of liberty ; without making the connection be- tween choice and motives, between personal action and the circumstances under which it takes place, one of freedom. Hence springs the notion of liberty, and the obstinate de- fense and maintenance of it by so many, in spite of faulty definitions, in spite of this inability to render any explana- tion of it satisfactory to the purely scientific mind. We are not conscious of liberty. If we were, there would be no room for discussion. We no more know the exact nature of the connection between the motives and the will from experience simply, than we do the connection between the volition and subsequent muscular action. In view of the accepted fact of accountability, and the ab- sence of all sensible constraint in motives, the mind predi- cates of the connection liberty — itself supplying the idea, and applying it to the phenomena; exactly as to another class of facts, it, in the same independent way, brings the idea of causal interdependence. In each case the mind pro- ceeds to meet and expound the facts with its own independ- ent notion, seen by itself to be applicable to the conditions of the problem. The movement is exactly that which takes place in the explanation of other experiences under the no- tion of space ; and of still others under that of time. The snpersensual nature of the idea of liberty must be admitted by all, certainly not less by those who deny its intelligibil- ity, and ridicule the assertion of its existence, than by those who accept both. It seems quite evident, that if freedom does exist, it is the regulative idea presiding over the facts of choice ; the form under which, the connection of the 224 REASON. will with tlie causal forces about it is to be conceived. Indeed, philosophers of the empirical school usually deny the existence and notion of liberty, at least as insisted on by Intuitive Philosophy. No one can reach, or has striven to reach, the notion of liberty through outside experience. It has, when accepted, been referred directly to conscious- ness, or to an intuitive power. § 10. A ninth intuitive idea is that of truth. Truth is the law of thought. It is the agreement of the judgments of the mind with the facts and principles to which they pertain. It is thus the goal of all intellectual inquiry. Truth may seem to be a special application of the notion of resemblance. More careful thought will distinguish the two. Between the intellectual statement and that to which it pertains there is no resemblance, as between one color and another, one form and another, one feeling and another ; nor is the correspondence of truth that of identity, the van- quishing point of resemblance. The two terms, the one an immediate product of the mind, the other a fact of some order in the Universe about us, are in every way distinct from each other in the formal elements of being ; they coa- lesce only in that single relation which we express by the word truth. The power to discern the relation is a pe- culiar power of mind ; and that it is an intuitive power is seen in the fact that at least one of the two terms brought together is an intellectual product. The senses therefore can not pronounce on truth, while the judgment simply affirms that true which is seen to be so. § 11. We have now reached an idea, whose nature and origin have been the occasion of much diversity of opinion. The conclusion we arrive at as to the nature of right, will profoundly affect our intellectual and practical life. The phenomena that call forth the discussion, though often narrowed by the theory adopted for their explanation, are, RIGHT. 225 in a general way, accepted and agreed upon. They are these. Certain forms of action are known by us as right, others as wrong; a sense of obligation accompanies the former when urged upon us, and of satisfaction and ap- proval when performed by us. The latter, on the other hand, when distinctly contemplated as wrong, deter the mind from acceptance by a minatory sense of duty, and punish the commission by a clear feeling of guilt. Of the presence and operation of these facts, history and language are full. Neither the speech nor the actions, the laws nor the religion of men, are intelligible without them. The testimony of individual experience is repeated in that of communities and nations. From the beginning men have been dealing with virtuous and vicious acts, with right and wrong courses of conduct, with innocence and guilt, respon- sibility and irresponsibility, honor and shame, praise and censure, rewards and punishments. These ethical ideas grow in the race as it advances. Our legislation, our social institutions, our daily actions, our religious beliefs are full of them ; and new labors of reform are constantly putting them into more pithy and pungent shape. Ethical science commands a large share of attention, and takes under its survey more and more broadly the actions of men. The shades of feeling involved vary from remorse and despair to the slightest uneasiness, from the triumphant self -justifi- cation of the martyr to a transient thrill of delight. Sin, wickedness, guilt, duty, right, righteousness, integrity, jus- tice, holiness, are a few of the weighty words under which these grave thoughts take their way. The facts involved being thus comparatively bold and salient, in a measure admitted by all, what is that theory of intellectual powers which best covers and expounds them ? The perception of right and the feeling of obligation are inseparable ; they are the intellectual and emotional sides 226 REASON. of one mental state. An obligation can not be felt without some direction or line of action to which it attaches. An obligation must be of a specific, definite character. An obligation without attachment to any act, is unintelligible, is no obligation. The quality right, seen in an act, is that which at once calls forth the feeling of duty, and directs it into a particular channel. E"o more can the intuition be separated from the feeling than the feeling from the intui- tion. Indeed, it is chiefly through the strong sentiment that accompanies it, that we discover the distinct character of the intuitive act. Language abundantly recognizes this double bearing of ethical insight. We have the word right as expressive of the intellectual recognition of moral law ; and the words ought, obligation, duty, presenting the emo- tional element. The theories which do not accept the original, simple, inseparable character of the idea right, explain the intel- lectual element by the generalized notion of utility. This is done with very different degrees of success by earlier and later writers ; but the empirical school agree in making utility the intellectual ground of ethics. We have appe- tites, sensibilities, tastes, affections to be gratified. Any thing or action which affords pleasure to any one of these is useful. This common power, which belongs to so many objects and relations, of furnishing some form of enjoy- ment, or some condition of it, is abstracted under the word utility. The inquiry which guides the conscience, it is said, is this inquiry into pleasure, into immediate and future enjoyment ; and that, if fairly and thoroughly pushed and made to cover all gratifications high and low, it is an ; exhaustive statement of all that takes place in ethical re- search. While this is an inadequate theory of the intellec- tual grounds of duty, it is difficult to disprove it. What is ; affirmed by it does take place, and is a most apparent and a J BIGHT. 227 most necessary part of the process by which we arrive at a practical conclusion as to a line of action, whether it be right or wrong. The usefulness of an action, in a broad and deep sense of the word, is a correct criterion of its moral character; it becomes, therefore, very difficult to show, that it does not cover the entire ethical element. The quality right, like the quality beauty, is seen in an intellection, that is, in. an act whose relations and bearings backward and forward have been inquired into and settled. What are the results which flow from it ? What are the feelings it expresses? How will it work forward in the world of facts % How does it work backward on the emo- tions ? These are the inquiries which disclose to us the in- tellectual bearings of the action, and prepare us to pro- nounce wisely on its character; they are also those which determine its utility. So far the ground is common to the two theories, sensualistic and intuitive. At this point they diverge. Says the one philosopher, these facts exhaust the grounds of intellectual action; says the other, .they prepare the conditions of a final, intuitive act overlooked by you, pronouncing the action not useful or otherwise, but right or wrong. The last words are not, and can not be measured by the first. In the intellection which we have reached in part at least as you have reached it, we discover a farther transcendent relation to ourselves, which we term right, and from which springs all our ethical action. In this we affirm we have the testimony of language with us, which by no means confounds, or allows us to confound, these two notions of the right and the useful. Nay, it separates them in clean and clear division from each other, reserving an emphasis for the one which it never thinks of bestowing on the other. It is, however, when the emotional element is consid- ered, that the utilitarian theory is seen to be most obviously 228 REASON. inadmissible. It does not satisfactorily meet the question, Whence arises the sense of obligation which is the salient feature of the right % It strives to make answer by affirm- ing that the feeling of duty is conventionally imposed by the community in satisfaction of its own sentiments, and in view of what is advantageous to itself. The obligation of ethical action is thus referred to education, to social and civil institutions, in their own behalf laying the pressure of duty on their subjects. Says Bain, " Authority or punish- ment is the commencement of the state of mind recognized under the various names — Conscience, the Moral Sense, the Sentiment of Obligation. The major part of every commu- nity adopt certain rules of condiict necessary for the com- mon preservation, or ministering to the common well-being. .... Every one, not of himself disposed to follow the rules prescribed by the community, is subjected to some in- fliction of pain to supply the absence of other motives: the infliction increasing in severity until obedience is at- tained. It is the familiarity with this regime of compul- sion, and of suffering, constantly increasing until resistance is overborne, that plants in the infant and youthful mind the first germ of the sense of obligation." — The Emotions and the WiU, p. 481. His definition of Conscience is, " An ideal resemblance of public authority, growing up in the individual mind, and working to the same end." The community grounds the law of action partly on utility, and partly on the transient sentiments which pos- sess it, and so, with a variety of sanctions, trains the child to obedience. " A certain dread and awful impression is thus connected with forbidden actions, which is the conscience in its earliest germ or manifestation." This theory derives a force which does not belong to it from the very fact that social law, appealing, as it often does, to our moral nature, acquires thereby a prescriptive RIGHT. ' 229 power which would not otherwise be attainable. If there were no foundation for custom and law in our moral con- stitution, the results of social instruction and discipline would be much less than they now are. With this grave advantage afforded by the frequent coincidence of our moral constitution and social customs, the theory still plainly fails to cover the facts. It should be observed, moreover, that man's habitual disobedience to moral law weakens its authority, obscures its phenomena, and thus greatly aids the effort to confound it with conventional rule. Notwithstand- ing these causes of obscuration, we believe a better theory still remains visible in the facts. (1) We have repeated examples of what general agree- ment and enforcement can accomplish, and the results are of another kind from those arising under true moral force. Take again, from another point of view, the illustration afforded by fashion. A kind of censure to which the masses of men are exceedingly sensitive, is constantly and unsparingly inflicted on those who disregard fashion. Yet the most infatuated devotee of the fickle goddess would hardly venture to regard scrupulous obedience as a virtue. Such an one is quite content if she escapes positive censure in her fashionable follies. How very different, also, the feeling arising from a violated fashion, from wearing a pro- scribed coat or hat, from that which affects the sensitive soul under the sense of wrong action. Allow each violator to be equally appreciative of the law whose precepts have been infringed, and we have, in the one case, mortification, and in the other guilt. The most scrupulous observance of the details of fashion, of fashion enforced by two thirds of the community, can not, does not, bestow the sense of virtue ; nor disobedience the feeling of vice. Take again the standard of honor enforced among cer- tain classes, as among soldiers, or gamblers, or on the Stock 230 * REASON. Exchange. The penalties here inflicted on disobedience, are as unsparing as the parties can make them. Yet such a custom as dueling is broken down by a purely moral senti- ment, based on the individual conscience, struggling with and at length conquering the general consent of the com- munity. It may be answered : Yes, but the sense of utility is with those who favor reform. Granted, but it is not, under the theory as presented in its present form, the notion of utility that imposes obligation, but the concurrent, edu- cational force of the community, and this is fully pledged to a custom which nevertheless calls forth on the part of a few a staunch condemnation, finding at length such response in the consciences of all as to lead to the abandonment of the censured act. Now, if the question were one merely of wisdom, there would be no mystery in the formation of a new opinion, and hence in a change of action. The diffi- culty under the theory lies in explaining how moral obliga- tion, which rests on an educational basis, which arises from the enforced sentiment of the many, which is the volume of sound made by a multitude of voices, can be brought to bear against an overwhelming majority, to the breaking down of those very beliefs whence it springs. How can one, two, three, outshout the crowd ? How can there arise a counter-sense of duty, when this sense is simply the concurrent opinions of men, sustaining as sacred the cen- sured institution. Duty would thus be like respectability, like popularity. They do go, and must go with the domi- nant party, and can not be used as an incipient force against themselves. (2) The thief, the gambler, the speculator, rest their laws on an educated sense of honor peculiar to themselves, and while they do secure obedience, sometimes more self- sacrificing and implicit than much of that which arises under moral law, it is notoriously with little or no reference BIGHT. 231 to such a law. They do not mistake their precepts for morality ; they are scrupulous, not conscientious, in their obedience to them. Occasionally to throw a slight coloring of morality over their actions, is the most they aim at. In a community in which slavery for many generations has been the law of the land, we find, nevertheless, au inde- pendent moral element getting a foothold. Conscience is appealed to, and a vigorous moral warfare springs up in the teeth of uniform custom. Nor do those who justify slavery do it on the ground of uniform practice, except so far as this is regarded as an expression of opinion on the part of those who have thus held their fellows in bondage. Other grounds than the mere fact of custom are sought, grounds which, so far as they exist, have a true justificatory element in them : the good condition of the slave ; his inferiority ; the general social order ; the exigencies of the case. I may almost say, that never is the appeal directly made between intelligent parties in an ethical discussion to naked custom and its penalties, for the defence of a line of conduct. This is a fact very damaging to the explanations offered. Men are never reverting to the bare fact of enforced law, as the ground and justification of law ; yet this after all is made the source of the sense of law. Moreover, in the very face of such enforcement, there does spring up, in sin- gle minds, a moral sentiment, which with pure moral power breaks down institutions hitherto unanimously sustained. We thus see what prescriptive force can do ; that it is by no means identical with morality, and that it constantly comes in conflict with the power this manifests, and yields to it. (3) Again, this theory fails most signally in cases in which the moral phenomena are most distinct. In the explanation of mixed conduct, of actions assuming an ethi- cal form, disguising themselves under moral sentiments, it prospers somewhat ; but when the moral element is promi- 232 REASON. nent and pure, it comes short. A conscientious man be- comes a martyr to his convictions of duty. He stands against the community, and confronts its authority, its al- leged line of duty, with his own independent convictions, his own sense of what is right. All the explanation of these most startling and pregnant facts in the world's moral history, facts that above all others catch the rational eye, and disclose the new force that is flaming up in them, is that of the " Self-originating or Idiosyncratic Conscience." It is an instance of " the transfer of the sentiment of prohib- ition from a recognized case, to one not recognized." That is to say, with no notion of obligation but the enforced one of education, the individual may, nevertheless, transfer it so strictly to his own independent, unsustained speculation as to oppose these serenely and unhesitatingly to the ut- most stretch of the authority of the community over him. This is a transfer indeed, a transfer that is a transformation, that discloses a sentiment in kind and quality totally un- like that with which it commenced. It went into the co- coon a worm, it comes out a butterfly. This is no explan- ation ; it is a confession of defeat. Better would it have been to have left the phenomena unexplained. (4) Further, we do not day by day impose duties on others in the manner that would be indicated by the above theory. Scarcely anything could be more adverse to the methods of those who are constantly using moral force, who are addressing and stimulating the conscience, than an appeal to the common sentiments, that is popular senti- ments, of those approached. Indeed, to such persons it would seem unworthy, sometimes even absolutely immoral to urge action on others primarily on the ground of the cus- toms and censures of general society. Nor could these cen- sures often be made to subserve the purposes of morality. The apostle of moral truth expects more frequently than RIGHT. otherwise to confront this public sentiment, and his appeal is not to what has been or is, but to the individual idea of what ought to be. The practice therefore which would flow logically from this theory of enforced morals, is not at all the practice of the actual, ethical world ; it is rather that of those classes who are feared and warred against, as always careless of the law of right, and often disobedient to it. Kindred expositions, insufficient to cover the facts to which they are applied, are found everywhere in the works of philosophers who advocate this theory of morals. " By remorse, we understand the strongest form of self-reproach arising from a deep downfall of self-respect and esteem." The Emotions and the Witt, page 106. This definition ap- plies to a conspicuous act of misjudgment, and most plainly does not reach the fact of remorse. Again, love is said to be " as purely self-seeking as any other pleasure, and to make no inquiry as to the feelings of the beloved personal- ity." This assertion leaves out the entire moral element which belongs to love as an affection, and is true of it only as a passion. The peculiar effect of " signal generosity " is referred to the " shock " given to the " mind totally unpre- pared" to see kind offices rendered to an enemy. Mill makes our sympathies with others in their injuries the basis of our sentiments of justice, a condition of feeling, certainly, which as often perverts justice as secures it. These and kindred solutions show the weakness of utilitarianism in handling striking moral facts, and how greatly it abridges and mars the facts themselves by a forced, belittling esti- mate of them. Eor is the sense of obligation any more satisfactorily ac- counted for under this theory by referring it directly to the idea of utility. At times, Mr. Mill seems ready to do this. As the useful, in the concrete, is the pleasurable, this refer- 234 REASON. ence would involve the assertion, that pleasure, as pleasure, is felt in human experience to be obligatory. This would farther include the statement, the stronger the pleasure the greater the sense of duty ; and, as our own enjoyments are more distinctly conceived than those of others, that these are pre-eminently enforced in practical morals ; and farther, as present gratification yields more intense feeling than anticipated indulgence, that the pleasures of the hour are especially watched over by conscience. Each and all of these conclusions are in exact contradiction of the facts. If there is anything in reference to which we feel ourselves left to our own unrestrained choices, it is our pleasures. The moral nature has not laid upon it the superfluous task of enforcing these ; but rather that of restraining them. By playing cunningly between the two, public sentiment on the one hand, and utility on the other, some embarrass- ments may be evaded by the theorist ; yet neither nor both can be successfully made the source of the sense of duty. When we are brought face to face with any, the wisest statement of the utilitarian law, as for instance that we are to seek the greatest good of the greatest number, how is that statement to gain any authority with us ? Evidently by our rational penetration into its inmost quality. Enforced in any other way on us it loses power to bless us. Its in- trinsic fitness does not save it, if we do not see that fit- ness. Enforced obedience is slavery for all the parties in- volved in it. Insight is the only refuge of manhood. Ultimate, honorable guidance must come to us through our own powers. "While these failures of explanation rob utilitarianism of all claims to acceptance, is there not in it a yet deeper difficulty in supposing that a simple notion, like that of obligation, can be other than primitive and independent of the action of society ? What would be thought of a philos- BIGHT. 235 ophy that should refer compassion, love, hope, as induced feelings, to the influence of others over the mind. Evi- dently all extraneous action is of no avail to awaken a feel- ing not given in the emotional constitution itself. A sense of duty, of obligation, is as simple as any emotion can be and if we acknowledge its presence, we must look on it as primitive in our constitution. But a sense of obligation is not intelligible as a general unattached feeling, indicating no definite line of conduct, haunting the mind as a vague premonitory fear, ready to be seized on by the first foreign force, to be applied as an alien impulse, having no necess- ary existence in the individual nor office for him. The im- posed opinion of others can not create a feeling ; the feeling of duty, like every feeling, must have a deeper basis than this. A general notion of obligation, with no intellectual element, no specific direction given to it by the mind whose it is, is as incomprehensible as would be a general impres- sion of truth, or delight in truth, with nothing presentiug itself as truth ; or a vague satisfaction in beauty, with no object regarded by us as beautiful. What can be found in our constitution, allied to such an unattached, unelicited emotion ? The vague feelings of fear sometimes present to the mind nevertheless disclose to more careful inquiry some occasion and ground of attachment in past experience and existing circumstances. There are but two open, plausible theories of our moral constitution ; the one which recognizes it as an original, in- dependent part of our constitution ; and the one which, through generalization, explains its manifestations by the facts of our physical and social position, making utility and public sentiment the germs of its intellectual and emotional elements. The last, in its pure, naked form, produces a far off semblance of the facts, replacing love and duty with fear and interest, and mistaking the forces at work in 236 REASON. a selfish, immoral world, for the true constitutional links of a higher, a holier, state. There are, however, theories which strive to combine these two, and while, in the last analysis, they are utilitarian in their principles, they keep aloof from the avowal, and include elements which only logically belong to an intuitive philosophy. Utilitarianism relies on the happiness afforded by correct action as the sole motive to it, and falls short of ethics in not being able to impose any line of action with authority, or to enforce one form of enjoyment in prefer- ence to another. Indeed, it has no sufficient standard by which to decide between pleasures, and to prefer one class above another. The question of the actual satisfaction ex- perienced by different persons m different lines of action must, like that of physical tastes, be left with the individ- ual, and if he prefer physical, to intellectual and social en- joyments, one cannot, under a mere law of highest grati- fication, impose on him the opinions of others. I do not need to inquire of a philosopher as to which apple is sweet and which sour, which agreeable and which disagreeable ; nor shall I much respect his view if it differs from my own. Thus, in all questions of pure pleasure, each man has his bias, and is not likely to yield it to a speculation that runs counter to his own experience, the final interpreter to him of the nature and quality of enjoyments. An effort to obviate this difficulty has been made by affirming the superior character of moral pleasures, and from this supreme quality reflecting back on the actions which secure it a sense of obligation. Herein is found the stolen element of a better theory. If we rely on the good which diverse lines of conduct produce to define and enforce our action, then we are entitled to these several kinds and degrees of satisfaction to direct and establish conduct, and to no more. Let all the sources of pleasure, making the RIGHT. 237 catalogue as discriminating and exhaustive as you please, be represented by the letters A, B, C, D, E. Let each one choose between them as he, under the guidance of his own tastes and capabilities, is able, in kind, degree, duration, dif- ficulty of attainment; and thus mark out for himself the path of prudence. He can not now go farther, and add to the motives urging any one proposed line of conduct a pe- culiar blessedness which is to crown it as right above all others. This is to establish again in our constitution a moral law, to restore to it intrinsic obligation, and thus secure the unspeakable satisfaction of obedience. All that our quiet, careful reasoner, overlooking the various sources of pleasure, and choosing between them, is entitled to, is, if he select wisely, the satisfaction of sagacity. He is always right when he is prudent, and the rewards of right sink to those of prudence. The self -congratulation of shrewdness, takes the place of the blessedness of a law implicitly obeyed, clung to in darkness and in light. Eo peculiar happiness can follow obedience to right, till we have recognized it as an antecedent, supreme, self-enforced law. As long as it re- mains a line of conduct resting for support on its pleasura- ble results, it must look to these exclusively, adding noth- ing to them save the satisfaction of sagacity. The right must come before the satisfaction which springs from obey- ing it. Herein is revealed a difficulty which more or less em- barrasses every presentation of the utilitarian theory. We grant, that what is right is always ultimately in a broad sense useful, but the moral nature, itself an independent means of gratification, a pre-eminent source of good, is often the necessary condition of its being so. The martyr sets this one pleasure over against all other pleasures, and wisely; yet he never would have done this, if he had started with the idea that the right action is only the sagacious 238 REASON. choice between enjoyments other than those which belong to the moral constitution. We are not in onr theories to have, and not to have, at the same time, the law and the rewards of the moral intuition. We are not to make ethi- cal pleasures to arise simply from the successful pursuit of other pleasures, and yet allow them themselves to be fur- tively included among these pleasures between which we are deciding. Many lines of action are obviously useful when accompanied with the gratification of our moral sen- sibilities, which are not so, when these as independent sources of good are left out of the calculation, as they must be in any honest evolution of a utilitarian theory. Spencer, in his Data of Ethics, falls headlong into this error. He affirms that life is for the sake of pleasure, and therefore that " that conduct is good which subserves life." If we suppose perfection of character, he argues, to lead only to pain, then that perfection itself disappears. He herein overlooks two obvious relations. Virtue can not be separated from the satisfaction that virtue occasions. This is to cut into parts a living thing ; this is to destroy the very notion, and of course if the notion becomes some- thing other than what it is, the results will be correspond- ingly different. Nor does it follow that because satisfac- tion necessarily attaches to virtue, that that satisfaction, as a new pleasure, is the motive of virtue. A disinterested act as disinterested is peculiarly pleasurable. If one " is blessed in performing an act of mercy," he is blessed because he did it as an act of mercy in oversight of per- sonal interest. Confusion at this point ought not to be any longer possible. The word pleasure is also plainly of a very generic order. There are the most marked differences in pleasures in kind as well as in degree. Yet if the utili- tarian grants this, all his weights and measures' are at once broken. RIGHT. 239 Nor is the intuitive philosophy, rightly presented, at all open to the repeated taunts of Bentham, that each individ- ual by a blind irrational power may thus pretend to decide what is right, and capriciously lay down a law absolute for himself and for others. All the investigation that Ben- tham or any other philosopher may bring to the practical effects of action, to its immediate and ultimate results, finds a place in our moral judgments. It is in full intellections made up by exhaustive inquiry, that the reason sees the right. We might as well say, because the judge authorita- tively decides a case, it is of no avail for the lawyers thor- oughly to present it, as to say, that because conscience ad- judicates between right and wrong, it is of no moment that the action to which the discussion pertains should be fully understood. It is the intellectual conception of this action which is declared right, and if this conception is incom- plete, then a verdict intrinsically correct is practically false, as pronounced on a hypothetical case and not a real one. The last decision, that of conscience, we believe to be cor- rect ; the presentation of the case, that on which this de- cision is made, to be often incorrect. Here enter the full fruits of investigation and protracted experience, an oppor- tunity for a broad, honest, faithful survey of the facts of the exact case to be made up and presented at the judg- ment-seat of the ethical sense. This merely gives the weight of law to what the other faculties have pronounced prudent and wise. There is no more opportunity for caprice and individual assumption here than in any debate concerning the qualities and bearings of actions. We designate in common language as conscience that action of the reason which discovers the right, and this is the ground or centre of our entire moral nature. Any theory which regards obligation as simple and ultimate, therein accepts the intuitive and independent nature of the 240 REASON. right, in the meaning in which we have employed it. Ob- ligation must arise in view of something, and in view of it in a moral relation. The two are as indivisible as the fla- vor and savor of a peach, the perception of the one and the enjoyment of the other. The system of ethics to be evolved from the above view is briefly this. All moral emotions, the entire moral nature is conditioned on a moral intuition, which we term that of right. This relation of rational acts, arrived at by a simple stroke of the eye of reason, in grounds previously unfolded, and which uniformly relate to the actions of free, intelligent sensitive beings, involves as an inseparable element the feel- ing of obligation. Here is the final authority of morals in the moral intuition. A reason can be given for the decis- ions of conscience in this sense, that the character and bear- ings of the acts pronounced right can be given ; not in this sense, that the intellectually discerned relations of these ac- tions are, aside from a distinct action of the moral faculty upon them, a ground of obligation. ]STo " good," as a good can give a law, can give a moral basis of action, since to do this it must go beyond its own appetitive range, and reach into the moral field of authority. It is to account for au- thority that we invoke the moral nature. § 12. The third idea regulative of our intellectual life is that of beauty. Concerning the existence at this point of peculiar phenomena that require explanation, there is no discussion. Yet results of analysis are quite different ; some reaching a simple, original idea ; others resolving beauty into utility, or unity and variety, or making it the product of association. That beauty is intimately connected with utility, that it is always accompanied by unity and variety, that taste is strongly influenced by association, and, in some cases overshadowed by it, are undeniable; yet that these explanations, in conflict among themselves, fail each of BEAUTY. 241 them to cover the entire facts, seems equally plain. Beauty is not proportioned to utility, is not always attendant upon it and exists sometimes with little or no utility, save that which the gratification of taste itself affords. Unity and variety are frequently present with no corresponding beauty, belong to structures which do not pertain to the fine arts, and thus show an independent existence and range. As- sociation explains many of the judgments of those who give little attention to intrinsic beauty, who under the influence of others yield their opinions to be swayed by the prevalent sentiment ; yet just in proportion as the presence of taste is manifest, as the perception of beauty is developed, as the phenomena to be accounted for are obvious and declared, this explanation fails. The leaders in fine art have no higher association from which to derive their estimates of excellence, while the different, external, accidental pleas- ures, that may for them incidentally find connection with works of art, are no sufficient ground for their uniform es- timates, singling these forth in all generations as objects of peculiar power and value. But this theory of association, of character transferred to objects of beauty from the relations in which we find them, is met by the fact that we have a pertinent example of what association can do in this same direction in affecting our es- timates of things ; and that it wholly fails to sustain the explanation here offered of the facts of taste. The admira- tion the general public express for a new fashion is almost wholly due to association, and what are its characteristics ? This esteem is fickle, contradictory, and wholly destitute of standards of judgment. Though in the present, unanimity may be complete, successive periods differ greatly in the forms rejected and accepted. Fortuity and the most extrav- agant fancies reign, and are equally imperious in their con- tradictory commands. The whole realm of fashion is one o£ 242 BEASON. unreasoning association, and it stands in conspicuous con- trast with that of taste, refuting the explanation offered of its stable phenomena. The uniform admiration bestowed by different nations and generations on objects of beauty ; the first high estimates which give direction to public opin- ion ; the word beauty, accepting in careful speech no syn- onym ; the fine arts, a distinctly bounded territory, eliciting the most skillful and prolonged attention ; and the well-es- tablished principles of this department, show that the fickle fanciful connections of association furnish no sufficient the- ory of taste. That the quality, beauty, accepted as unresolvable into any other, is of intuitive origin is seen in the fact, that it is not directly a quality of things, but of intellections. An intellection is the product of the mind. The qualities, forms, and relations of an object — its expression — are by studious observation brought before the mind. This esti- mate which the intellect makes of all that unfolds the char- acter, the emotional power of an object, is an intellection, and in the object thus conceived, thus unfolded in the thoughts, beauty is seen to inhere. As beauty thus does not belong to a flower, a tree, a landscape, a bird, a man, merely as a sensible object, but to them as products of an arranging, vitalizing, perfecting power ; as it is seen not in the thing simply, but in it as conceived by the mind, it must be the object of an interior, intuitive faculty, which can take into its contemplation the appropriate intellection. An act of exposition more or less complete, has followed perception, and thus the object has been taken from the senses into the mind, and there awaits the insight of the reason. The qualities one and all which make up the expression pronounced beautiful are not the very beauty which we attribute to the cathedral, the painting, or the statue. The skill, proportion, height of the towering edifice THE INFINITE. 243 may be discerned separately from that final effect, that joint and supersensual power, that more than analytic pleas- ure, which we term beauty. This is not the craft of the workman, the single nor the combined excellence s of the work, but an overshadowing power through which these have their chief value, by which the seal of a fine art is put upon them. The intellectual relations of an object, capable only of an inner presentation, are that in which, as sub- stance, the reason sees beauty to inhere. Beauty is not these simply, though it comes and goes with their varying combinations. § 13. It only remains to speak of the infinite, the last of the intuitive ideas, and one that has recently given rise to much discussion. It finds application in several direc- tions, and perhaps, in the development of the mind, as early to space as to any other form of thought. The notion of space cannot be dwelt on without soon suggesting this idea of the infinite. The mind soon sees the inapplicability of any measures, limits, finite relations to space, and that, in the very moment of establishment, they are swept away by the on-going movement. Space lies without, as much as within, any line we choose to run, and the nearer has no advantage over the farther side. The mind under this new necessity laid upon it, with this new occasion given to it, grasps the idea of the infinite, of unmeasured and im- measurable extension. This conception is not the result of mere weariness, is not the affirmation of an inability to pro- ceed farther, does not spring from repeated and reiterated failure ; it is rather the force and insight of the mind that disclose it. It is seen that there can be no advantage in pressing the imagination to its utmost flight, that the con- ditions which are now present at this point of space, must recur everywhere, no matter what the position attained by us ; that one point and one position here or there, that each 244 REASON. bound longer or shorter, are facsimiles of every other, and therefore contain the solution of the problem as perfectly as if it had been raced after with the most wearisome efforts. The mind does not then distress itself in search of a limit, and fail ; it discovers that there can be no limit ; it penetrates the conditions of the problem, and brings for- ward the notion of a true infinite, which it sets over against the finite, and is at rest, as it knows that nothing other or more is to be found elsewhere. Thus the mind hits upon the true infinite, not by expe- rience, not by exhaustive effort, but by its own penetration of relations ; and through this idea it understands another of the conditions of its experience, and declines exertion which it sees to be necessarily futile. Standing, not mov- ing ; by insight, not by baffled effort, it grasps and hence- forth uses this notion, so super-sensual in character, so necessary for the exposition of the being we possess, the universe we inhabit. Space, as infinite, admits of no divi- sion. ~No plane can cleave it, no line pierce it. In strict language, it is without parts, at least so far as these imply remainders. The true infinite is subject to no addition, subtraction, multiplication or divison. These are processes which find play in the finite alone. A second point at which this notion arises early is in the contemplation of time. Here, too, the mind discovers that the conditions of conception are not in the least varied by movement, and that the years which beheld the laying of the foundations of the world were no less central than those which now are, or those which shall behold its overthrow. Geologic seons lie lapped in eternity, with no more power of measurement than the point which defines pure position on the board before me. Here again there is no opportunity to take aught from, or add aught to the infinite, to eternity. Indeed we may not strike it into two infinite halves by this THE INFINITE. 2^5 fleeting moment the present, as if it were a node jointing the past to the future. A hemisphere is not a sphere because it meets on one side the conditions of the defini- tion. A true infinite must be immeasureable in all the directions in which measurement can be applied. A for- ward or a backward stretch, leaving a definite, finite period in the opposite direction, constitutes no true infinite ; the lines which pass out from any given point are not infinite, they lack an essential feature of the infinite, interminable- ness. They are limited in one direction. We are always to distinguish between the indefinitely great and the in- finite. Mathematics deals with the one, metaphysics with the other. A series of figures increased as you please, can never express an infinite amount, and therefore no infinite can be twice or thrice as great as another infinite. This borne constantly in mind, and we shall easily dispose of a portion of the perplexities Sir William Hamilton has thrown around the subject. " A quantity, say a foot, has an infinity of parts. Any part of this quantity, say an inch, has also an infinity. But one infinity is not larger than another. Therefore an inch is equal to a foot." Neither an inch, nor a foot, nor any other definite quantity, has an infinity of parts — parts that are parts, that have any size, will exhaust any dimensions short of the infinite, and the quotient still remain finite. " A wheel turned with quickest motion ; if a spoke be pro- longed, it will therefore be moved with a motion quicker than the quickest." This example and similar examples, are mere riddles arising under a play of words. There is no absolutely quickest motion, and no motion that is infinitely rapid. The perplexity in these cases does not all spring from the notion of the infinite, but from the effort of the imagina- tion to transcend its own conditions in a false search by a 246 REASON. false method after the infinite, or the infinitesimal. The imagination must have finite, phenomenal quantities to deal with. These, therefore, are always capable both of mul- tiplication and division. The fancy may carry on the proc- ess till it gets weary ; confounded with the results, it may mistake its own embarrassments for those of the entire mind. It does this only by overlooking and denying the true na- ture of the infinite, and the source whence alone it can be rationally looked for. It should not distress the mind, be- cause the end of a circle can not be found by chasing round and round it. No more should it, because that which has not dimensions cannot be reached by cutting down, and at the same time saving, that which has. This is striving in the same instant and act to hold on to the finite, and to take it away, to keep it and to get beyond it to the in- finitisimal. It is no more a startling and discouraging fact, that the imagination can make nothing out of nothing, nor give limits to that which is without limits, than it is that the body can not be suspended by a spider's thread. Re- move the support beyond a certain amount in either case, and there must be a downfall. A third direction in which this notion is applicable, is to the attributes of God. God is infinite in power, in wis- dom, in goodness ; that is there are no limits to these attri- butes within their own nature. All that power can do, the power of God is able to do. The infinite in space presents itself under other forms from the infinite in time, and both of these in a way yet different from the infinite in power. The nature of power is not altered by the affirmation of its infinite extent. This merely removes its limits. It can no more do now than before what is not pertinent to its nature, what must be the product of wisdom or of grace. The no- tion, in its application to God, comes to assume those per- sonal relations, that independent perfection of existence, THE INFINITE. 247 which we designate by the Infinite the Absolute. God is thus lifted above the reflex action of causes, as well as above their antecedent action. Not only is nothing back of Him, there is nothing before Him, giving condition and law db extra to His nature. God as infinite is not limited in action ; as absolute, he is not straitened by any reaction. The material with which man works, limits the work ; no limitations reach the labors of God save those contemplated and established in his creative thought. The infinite in this form, in these its various applications, we must defend as a positive, intuitive idea— indeed, if it be an idea at all, it must be an intuitive idea. The first objections against the positive, valid character of this notion which we shall consider, are those of Sir William Hamilton, presented under what he terms, The Law of the Conditioned. It is claimed, that this impres- sion, like that of causality, arises from the powerlessness of the mind, not from its insight. The line of argument is much the same as in the case of causation. The following with omissions, is his presentation of the subject. It is found in the Lectures on Metaphysics, p. 527. " We are altogether unable to conceive space as bounded, as finite ; that is as a whole beyond which there is no far- ther space. * * * On the other hand, we are equally power- less to realize in thought the possibility of the opposite con- tradictory ; we cannot conceive space as infinite, as without limits. You may launch out in thought beyond the solar walk, you may transcend in fancy even the universe of mat ter, and rise from sphere to sphere in the region of empty space until imagination sinks exhausted ; with all this, what have you done % You have never gone beyond the finite. • • Now then, both contradictions are equally inconceiv- able, both are equally incomprehensible ; and yet, though unable to view either as possible, we are forced by a higher 248 REA80N. law — that of excluded middle — to admit that one and but one only is necessary." He then treats in the same way the minimum of space, the maximum and minimum of time, and proceeds, " The sum therefore of what I have now stated, is : that the conditioned is that which is alone conceivable, or cogitable. The unconditioned that which is inconceivable or incogitable. The conditioned or the think- able lies between two extremes or poles." Later he says, " These poles are the absolute and the infinite ; the term absolute expressing that which is finished or complete ; the term infinite that which can not be terminated or con- cluded." The doctrine of the law of the conditioned is the most characteristic feature of the philosophy of Hamilton, and is open to obvious and fatal objections. It does not explain (1) why the mind is thus embarrassed in its conception of the maximum and minimum of space and time ; nor (2) why it is ever led to vex and torment itself with these im- possibilities, forsaking the conditioned where traveling is comfortable and profitable, to scale cloud-heights which never give foothold to the foolhardy assailant ; nor yet, most strange omission, (3) why of two impossible concep- tions equally perplexing, we are called on to accept the one, that of infinite space, infinite time, in place of the other, that of bounded space and time. A better theory is able to offer an explanation of these difficulties. The mind is baffled in a conception of a maximum and minimum of space, because a faculty is set to the task which deals ex- clusively with the phenomenal, aud it is no more curious or surprising that the imagination can not attain to the infin- ite, than that these limbs of ours can not mount a sunbeam, and so reach the heavens ; or, more aptly, than that we can not see, hear, smell the infinite ; since the senses are the analogues of the fancy, both covering in a different way the THE INFINITE. 249 same field. We have given the imagination a work to it impossible and preposterous. Why are there these excur- sions of fancy into impracticable regions % Because over- looking the direct, intuitive grasp of the mind, and still haunted by the notion of the infinite, we put spurs to the steeds of the imagination to see if we may not in this way overtake it. The sober, plodding judgment turns aside from the thinkable to the unthinkable, in hunt of a ghostly conception which is real enough to bewilder the eye with strange appearances, but too unsubstantial to be grasped and handled in physical fashion. To pursue spirits or flee from spirits on horseback is of little avail, though, with man's belief in the spiritual world, the nature of the pursuit and its philosophy are sufficiently plain to the quiet looker-on. A most fatal failure of this theory is to explain why we uniformly and certainly accept infinite space which has no advantage to the mind over the supposition of finite space. This embarrassment at once disappears, if we sup- pose the notion a positive one, provided by the mind to be placed in explanation and comprehension over against the finite. The theory of Hamilton succeeds in eliciting the perplexities of the subject, but brings to them no solution. But it will be said, the intuitive theory has its own and yet more fatal difficulties. How can the infinite be a posi- tive idea ? Yery easily, if we assign it to the right faculty, and make it simple and ultimate ; as easily and intelligibly as red is red, or sweet, sweet. In neither case can we go beyond the ultimate fact, and we have fortunately learned in the more familiar instance to give up the effort. This objection may come in the form of a second theory of the infinite, to wit, that the notion is a negative not a positive one, involving a denial merely and not an affirmation. That the word is negative in form is a fact of no signifi- cance ; so are inhuman and indecent. If the word infinite 250 REASON. simply set forth the fact of non-existence, we should at once lay aside the article, and no longer speak of the in- finite, any more than of the nothing. It is because it stands over against the finite, embracing the sum of possibilities and powers not expressed or measured therein, that we call it the infinite. If its negative form contained the true secret of the word, it would occasion no more perplexity, would contain no more profound depths, than does the finite. Nothing is as intelligible as something, the termina- tion as the extension of physical objects, and if the mind did accept the word as a mere denial of anything more, it would accept it contentedly, without this endless bother and perplexity, this groping on for something not yet reached. It is said, in proof of this negative character of the notion, that it is inconceivable. This we grant, and have given the reason why it is inconceivable. It is not an ob- ject for the imagination. No more is the notion of causa- tion, nor of liberty, nor of right, nor of beauty, Nothing which is not phenomenal, nor under the immediate form which phenomena are assuming, is a subject for the imag- ination. It is further said, The infinite is not thinkable. " To think is to condition," is to throw into finite relations, is to destroy the notion of the infinite. The same answer as that already made is still open. The list of our faculties is not exhausted when we have marked off the imagination and the judgment. It is possible that the reason was given to us for this very end, to reach ideas not otherwise present to the mind. We hardly see why it should be present, or thought to be present, to furnish thinkable and conceivable objects, that is, objects arrived at by other faculties. In what sense, however, is it true that the infinite is not thinkable ? It is true in this sense only, that it can not be approached by explanations grounded on resemblances, that THE INFINITE. 251 it can not be made the subject of judgments, at least, of those which limit it under finite analogies. And why should we expect it to be ? Do we not antecedently see and say, that this process must be destructive to the very nature of the notion ? Why then proceed to allege the fact against it? We can do this rationally only by involving the assertion, that the judgment and imagination are our exclusive faculties of knowledge ; and this begs the ques- tion at issue. To reject the reason because it does not do the very superfluous work of giving an idea capable, by like- ness and relation, of falling into the list of previous ideas, is to misunderstand the object of the faculty, or to assume that its existence is impossible. We might as well object to the validity of our knowledge of an odor, because it is not thinkable, or, forsooth, conceivable under color or sound. In this sense, then, we admit the infinite is not thinkable ; but all thinking is not under limitations and conditions. Sometimes it is quite the reverse. To say that God is infi- nite is to deny conditions of Him. To say that The Infinite is, that He is free, that He is holy, is not to condition, to limit God, rather the reverse. The fact that we cannot go farther, and conceive the acts in and by which His liberty and holiness express themselves, except under a finite form, does not destroy the meaning or significance of the antece- dent assertion, it merely presents another case of a familiar difficulty, that of getting from one province of knowledge to another. Different tracts of cognition do not lie together, like the provinces of one empire, the transition one of move- ment only. Here springs up another modification of this theory, that of Herbert Spencer. He regards the notion of the in- finite as of an illusory character, shown by the very fact that every effort to give proportion and definiteness to it, baffles us, and results in driving it into more remote re- 252 reason: gions. "We admit the perplexity which a portion of our faculties, whose action we are most familiar with, and from which we are accustomed to receive most of our conclusions, experience in handling, or rather in striving to handle, the infinite. This fact presents to us no difficulty ; we see the reason why these faculties are not adequate to the labor laid upon them. Indeed, our belief in the infinite would be overthrown by a successful presentation of it, either by the imagination or by the judgment under its own forms, and is established by this very failure on their part. The objec- tion of our adversary is proof with us. On the other hand, the opposite view, that the notion is wholly illusory, is involved in difficulties that it cannot evade. How can Spencer insist that any presentation of the infinte is not adequate, when he has no notion of what the infinite is ? How can a notion be shown to be illusory, except by a growing intuition ? How can Hamilton require us to accept by faith that which is unintelligible, absolutely and completely so. Here are real contradictions. There can be no general denial of the applicability of any and all conceptions of the infinite, without postulating thereby some notion of the infinite with which these are compared, and, as falling short, are pronounced wanting. One notion of an utterly unknown thing is as good and as adequate as another. Neither can faith make that an object of belief which is utterly unknown to the mind. The faith of Ham- ilton, and the vanishing conception of Spencer, are both self-contradictory, as being alone able to arise under the furtive, but real light of an idea present and ruling in the mind. No false conception of the Deity can be set aside except by one which is better, or is deemed better ; no faith can be expressed except toward a Being thought to be. These perplexities find no removal. To escape, therefore, difficulties whose reason is forthcoming by difficulties that THE INFINITE. 253 find no solution, is to forsake the light for darkness, is to employ exposition with a loss of expository power. Nor are formulae of thought which are inadequate, in a limited sense false, unserviceable, if their deficiencies are clearly seen by the mind that uses them. The expres- sions, infinite power, infinite wisdom, infinite goodness, con- tain as statements two things: the qualities indicated by the nouns, power, wisdom, goodness / and their unlimited degree, pointed out by the adjective, infinite. Our ideas of the first may gain in precision and clearness without affecting the applicability of the adjective which sweeps away their limits. We may inquire experimentally into the nature and forms of power, and yet well understand that these precise manifestations are swallowed up in, included under, infinite power. We thus use in mathematics the first term of an infinite series to define and represent the remain- der ; or we make the rule for the area of an inscribed poly- gon that of the enclosing circle, on the ground of the con- stant approximation of the one surface to the other with each increase of the number of sides. Yet the one never ab- solutely conforms to the other. The intellectual formula for the infinite is, This and more. The noun gives that which is to be expanded, the adjective, the law of its expansion. The this of the formula gives room for inquiry and growth, the more cuts us off from regarding a part as the whole. This is a movement of thought practically simple and safe ; no more inexplicable, no more dangerous than the use of suppositions in mathematics which reach toward the exact truth without finally covering it, which put one thing for another on the ground of constant approximation. Con- ceptions are habitually employed in mathematics which are inconceivable. We regard circles as perfect, yet the de- scription of a perfect circle is to the imagination an impos- sible task, for the same reason that it is to the senses. Start 254 REASON. with a describing point. If it move for the least interval in a straight line, so far the line is not curved ; if it begins to bend before it has traced the least portion of a line, it has nothing from which to bend or curve. It must begin to move and curve at once, and an image of this the fancy can not form. Even a point, in order that progress may be made by passing through it, must have some breadth, and this breadth, if it is to give an initiatory direction from which the cur ye is to depart, must be straight. In analytic conception we resolve the descriptive process into motion and departure, or bending from that motion; we can not conceive these two to be absolutely and constantly synchro- nous, yet without this the circle is imperfect. The imag- ination follows after the hand and eye, and as these are not exact, neither is it. § 14. Having presented the twelve intuitive ideas which constitute the mind's intellectual furniture, and also the grounds of proof in each case, we propose further to draw attention to some considerations which belong to all of them, establishing their character, and separating them from generalizations. Necessity and universality have been fixed on as the criteria of these notions. The two tests are liable to be mistaken for one, and are so under a certain rendering of them. To distinguish these from each other, we should understand by necessity, that certainty of con- viction which attaches in all minds to truths purely depend- ent on intuitive ideas. Thus there are in the definitions and axioms of Geometry many secondary intuitions, refer- able to the primary intuition, space. From these there spring convictions, in the certainty with which the mind receives them, wholly unlike those dependent on experi- ence. That two straight lines, lying in the same plane, and for a space equally distant, will remain so through their entire length, is an assertion which the mind accepts at CRITERIA. 255 once, as a necessary truth. Nothing, probably, but the exi- gencies of a theory, would ever lead one, with Mill, to strive to trace a conviction like this to experience. Certain it is, that no mathematician ever thought of establishing it by induction. Experimental truth never imparts such im- mediate and perfect belief. Of a like nature is the instant and unavoidable assurance that the changes taking place before us have a cause. Whenever a statement is solely dependent on a regulative idea, it becomes a necessary or demonstrative truth. Universality, remaining a separate criterion, must refer to the constant presence of one or other of these ideas in every judgment ; to the fact of the impossibility of dis- tinct, declared thought in any mind without them. These universal antecedents of thought cannot be furnished by thought itself. Thought cannot supply its own conditions. The universality of their presence in each act of mind and in all minds becomes thus a proof of their supersensual nature. It seems to us, however, that it is a careful analy- sis of the processes and growth of thought, that is to estab- lish each idea by itself ; to lay open its transcendental char acter, as in the case of the infinite and liberty ; or its neces- sary, antecedent presence to a certain class of judgments, as right to ethical judgments, consciousness to the appre- hension of mental facts. The three criteria, the necessity of the involved truths, the universal presence of one or more of those notions in all judgments, the transcendental nature of the conceptions themselves, are not applicable all of them with equal clearness to each of the twelve ideas, and must be applied and sustained by a distinct discussion of the mental phenomena involved. A more analytical application of criteria would involve five tests of intuitive ideas, (1) immediateness, (2) necessity, (3) universality, (4) identity, and (5) transcendental charac- 256 REASON. ter. These are not distinct tests, but different applications of the same test, that of a simple idea, disclosed as such in intellectual analysis. A simple idea must come, when it comes, abruptly, must carry with it absolute conviction, must be the only solvent of the facts it expounds, must ever be identical, and transcendental. The absolute identity of space, of time, of number in all their applications is note- worthy. Resemblance, though it may lie between different things, is yet ever one and the same relation ; right marks the one fact of a rational law of action. § 15. Another vital point in this discussion is whether these ideas are to be regarded as purely subjective, as mere mental forms brought to the object-matter of thought, or whether they pertain as external, necessary forms to that matter itself, thus possessing an independent being. The first belief, as advanced by Kant, that intuitive ideas are the mere moulds of thought, becomes the initiatory term of idealism. It is as contradictory to the universal opinions of men as any philosophy well can be. We do not say, that it is contradictory to consciousness, for it is not, but that it sets aside as wholly invalid and without foundation the uni- versal convictions of men, thereby casting great improba- bility on its own conclusions. Cause and effect constitute the notion by which we more especially establish the existence of the external world. Not to accept as just and safe the inference to which this idea of the mind lead us, is (1) to deny the integrity of our faculties, and to introduce a fatal scepti- cism to which no after limits can be set. It is a fundamen- tal principle of sound philosophy, that the integrity of no faculty can be denied, nor its guarded, normal action be set aside. If, therefore, we recognize the universal presence of the notion of cause and effect, we have no more right to treat it as illusory, than we have thus to regard vision REALITY OF IDEAS. 257 or memory. Spencer justly says, " That Space and Time are 'forms of sensibilities,' or 'subjective conditions of thought,' that have no objective basis, is a belief as repug- nant to common sense as any proposition that can be found." This conclusion is reached in philosophy by rejecting without reason an action of mind universally present. We do not, indeed, know the objective world in perception, since consciousness is a condition of mental phenomena only, and these are not identical with the physical phe- nomena which they represent or accompany ; but we do know it inferentially under causation. The action of the mind herein, as clear and constant and universal as any action, implies a power or faculty whose office is to make these disclosures. It may be said, that this view is as open as the oppos- ite to the criticism of disregarding the general conviction, since this is not merely that we know, but that we act- ually see and feel, the outside world. The cases presented by the two theories are very diverse. The one rejects en- tirely conclusions universally accepted ; the other, in careful analysis of a complex operation, refers them to an obscure element, easily and frequently overlooked. The popular mind regards sight, touch, as simple operations, and so ascribes to them our knowledge of the external world. It is deficient in analysis, not erroneous in its reference. Phi- losophy resolves sensation into distinct operations, and as- signs to one of these, that of causal inference, the imme- diate proof of outside existence. It is to be claimed that the general action of the common mind should be regarded as normal ; it is not to be claimed that analysis may not go farther than ordinary concrete judgments. The mind soon learns (2) to distinguish between sensa- tions and thoughts, between phenomena which come and go at its own bidding, and those which are entirely independ- 258 REASON. ent of its will. It necessarily assigns the one a different source from the other. (3) As sensations in different or- gans are found to be connected with the same object, this fact, in an additional and confirmatory way, establishes for the mind its external and independent existence. Touch and sight aid each other in fixing and locating the source of the impressions in each sense. The sensations and percep- tions are found to come and go together, and are therefore inferred to spring from a common cause, external alike to each organ. The location of the senses themselves, the gradual apprehension of the objects, distances, and relations of the external world, are processes of which we shall have occasion to speak more fully. It is sufficient for our pres- ent purpose to observe, that under the notion of cause and effect sensations and perceptions are distinguished from other facts of mind as having an independent origin ; that these external causes are slowly confirmed by repeated ex- periences, entering through a variety of organs, and that as a result of this normal movement of mind, men do every- where arrive at, and believe in an external world, the same to them all. The cause is as real as the effect, and to ac- cept a sensation as actual, is virtually to accept for it an in- dependent cause, and, under the instruction of protracted experience, an external cause, external not merely to the mind, but usually to the body also. This movement as spontaneous and universal cannot be invalidated without an overthrow of the credibility of a portion of our faculties. (4) Observe also that the result is the same for all; men move in one external world. One set of objects, one relation of objects belong to them all, and they harmonize their actions by the validity of this their common experi- ence. Make the world subjective to each individual, and you virtually deny for each the existence of all others. The preposterous conclusions of pure idealism could only OTHER IDEAS. 259 be made to rest on the most undeniable proof ; nor on that, for the effect even then would be one of general confusion ; of speculations wholly at war with practical conclusions ; of the discord of knowledge rather than of sound, settled, consistent belief. A similar line of proof has been carefully applied by Dr. Hickok to the notions of time and space. The reality of space has been shown to be the only condition on which the phenomena of the physical world can be the same for us all, included in " one whole of all space," open to com^ mon knowledge and common use. (5) Moreover we dis- tinguish imaginary space — space which the fancy furnishes as a setting for its pictures — from real space. The space of the senses and that of the imagination are entirely different, showing that space as a form of thought is at once distin- guishable by us from space, an external form of real being. We may advantageously distinguish between formal en- tities and real entities. The substance of a pillar is a real entity, the appearance under which it presents itself is a formal entity. An event is a real entity, the time and the space in which it occurs are formal entities. Formal enti- ties share the fortunes of real entities. The mere image of a pillar in a mirror has neither, in actual existence, the sub- stance nor the form of the pillar. If a thought exists so does the consciousness in which it arises. If the one is imaginary, the other is imaginary also. Other ideas have been offered as intuitions, and great patience and skill of analysis are requisite in settling these first forms of thought. Relation is one of these notions easily presenting itself as intuitive. It is obscure general- izations that are especially open to such a reference. Re- lation does not express any one specific connection, any one form of dependence, but many and most diverse forms. It is one of the most vague and broad of generalizations. 260 REASON. Resemblance is a specific form of connection, not so rela- tion. If relation expressed any intuition, it would express a large bundle of them. Moreover, other regulative ideas involve in different forms this idea of relation, and can riot maintain their integrity without it. Resemblance, causa- tion, liberty are specific relations ; number, time, space, in- clude many relations. Hence we must regard relation as a generalization, whose various concrete, forms are found in other regulative ideas, and the combinations of phenomena under them. An intuition always involves the essential unity or simplicity of the idea, as that of time ; a generali- zation involves the variety of the quality or relation, like that of sweetness or of usefulness. § 16. An effort is made by empirical philosophy with increasing distinctness to strengthen by inheritance the processes by which it supposes the general mind to have reached among its convictions those expressed as intuitive truths. Especially is it thought that the necessity that is attributed to these ideas is to be explained in this way. Fundamental convictions once reached by the mind are passed over, confirmed and enlarged by descent, till they assume an instinctive, intuitive character. Spencer, in his Data of Ethics, enforces this view in explanation of the right. A sense of obligation is due to "preferences and aversions rendered organic by inheritance." " The intui- tions of a moral faculty are the slowly organized results of experience received by the race while living in presence of the conditions of the highest life." To this opinion there are plain and decisive objections. (1) Knowledge proper, clear mental conviction, does not pass by inheritance. The father does not transmit to his son his skill even, much less his mental acquisitions. Knowledge which is of the nature of training, which has a large physical element, and is closely associated with in- INHERITANCE AND IDEAS.' 261 stinct, may in a measure pass by descent ; as a modified form of nest-building with a species of birds, or bunting qualities in a dog. The more, however, that acquisition partakes of the nature of knowledge, the less is it trans- missible. That our highest, most penetrative insights, like those of morals or mathematics, should owe anything di- rectly to physical inheritance, is quite opposed to the laws of heredity. (2) Such descent would also reverse the general order of progress. Instincts precede knowledge, rather than knowledge instincts. Intelligence bases itself upon the organic, instinctive life ; rather than itself issues in these foundations. Habit or skill, though somewhat beyond thought, is yet thoroughly permeated with light, is some- thing quite above instinct. The field which instinct prop- erly covers, it tends to occupy to the exclusion of knowl- edge. The region of instinctive activity is a dark, opaque one. It is the deficiencies of instinct which knowledge is called on to supplement ; while instinct anticipates knowl- edge, and, in its own direction, renders it unnecessary. The true order of growth is overlooked by this theory. Organic forces and instinct, not knowledge, initiate life ; while knowledge holds its own in possession more and more perfectly. An insight that should lapse into an instinct would tend to deaden the intellect; but nowhere is our knowledge so luminous, so complete, as in our intuitions. This knowledge is at the very farthest remove from a blind, half- organic under-current. Its certainty it wins not from instinct or inheritance, but from insight. The constructive order of the world is this: purely physical forces support and minister to organic forces ; organic forces expand into and nourish instinctive forces ; these in turn make way for associative processes ; while associative processes prepare the ground for and find their 262 REASON. interpretation in rational activity. By instinct we mean snbtile constitutional connections, .through which actions, having the form of intelligence, are automatically accom- plished. It is simply an extension, as in the spinning of the spider, of organic stimuli. By association we mean, the union of facts of experience in quasi judgments by memory.* Each higher stage in this series will react on and modify that below it, but the fundamental dependence is the one indicated. Any other relation would make the higher endowment the preparation for the lower one, and its condition in development. Instances may easily be given in which the later gift modifies, the earlier one; but this is quite a subordinate fact that must itself find explana- tion in the previous relation here presented. Intelligence will work its way in a limited degree into instinct, and secure transmission by descent ; but this will take place only in lower forms of life, and is a wholly in- sufficient theory of interpretation when applied to man's highest powers, powers that are not instinctive but intui- tive. These must be accepted in their supreme quality, and their action upward and downward sought out. Intelligence, in its higher forms, holds beneath it a large constitutional automatic region which it can penetrate as voluntary power, and ultimately possess and control as habit. Instinct and habit are allied to each other in form, but quite distinct in origin and office. Instinct is an ex- pansion from below of automatic action ; habit is the higher life finding its way into the lower. (3) Moreover, the first knowledge, on which inheritance is made to rest, remains to be explained, and adequately explained leaves no room for inheritance to add anything. Inheritance can do nothing unless there is something to in- herit. Whence comes that incipient perception of right, or * For a farther discussion see Growth and Grades of Intelligence. GROUPING OF IDEAS. 263 those first convictions of necessary truths, which have been confirmed by descent % We are to remember that descent obscures rather than makes clear the rational element, these earlier insights, therefore, must have been as bright as the morning light, and are still to be explained. (4) Farther, these intuitions are now with us as insights, not as instincts. They express mental powers, not organic processes. The truths of morals, the axioms of mathemat- ics, indicate the clearest and most decided action of the in- tellect. It is strange that these latest convictions should be likened to the organic impulse with which a bee builds its comb. (5) These higher truths also, as those of mathematics, have not been with the race to be confirmed by descent. They have come late and suddenly, and have come with full force. (6) Nor does uniform, empirical truth gather by inheri- tance this sense of necessity. The world would still be perfectly willing to find a white crow. § 17. The twelve ideas now presented are capable of being grouped in various ways. Space and consciousness are the two diverse and complementary fields in which all phenomena occur. The higher plane has, as ideas peculiar to itself, consciousness, spontaneity, truth, right. The only ideas peculiar to physical events are space, and cause and effect. The infinite, on the other hand, is an idea that keeps aloof from the phenomenal, comes in only to compre- hend the finite, and, in its personal form, to give the inves- tigations of the mind a final goal, — one from which they may start, and to which they may return. Existence finds its chief significance in its resemblances ; space, in its numerical relations; time, in the causal se- quence of events. The first couplet gives us the facts of being, and their character; the second, the most abstract 264 REASON. relations of things in co-existence ; the third, their relations in sequence. Cause expresses the law of evolution in the physical world; spontaneity, that of the spiritual world. Forces are installed in distinct measure and form in space, their initiation is an act of mind. Liberty in actual choice forecloses liberty, and henceforward realized force moves with a necessary impulse to its goal. These ideas admit of a collective grouping in reference to the method in which they cover all phenomena. Existence. Number. Resemblance. Space. Causation. Time. Beauty. The Infinite. Consciousness. Spontaneity. . Truth. [Right. We start with existence as the fundamental affirmation in connection with all facts. This is followed by a recogni- tion of their plurality, and this — we are speaking of logical order — by their variety. Plurality and variety, number and resemblance, tend at once to emphasize each other, though the latter, in its discrimination, is an advance on the for- mer. These three ideas are common to all phenomena, whether of matter or of mind. At this point the stream of events divides, though its facts flow forward in each branch alike under the idea of time. Physical events, arising in space, and impelled by forces, take up the notion of causa- tion. Mental phenomena show themselves in conscious- ness, and are linked, so far as they are purely mental facts, by spontaneity. The law of pure mental action is truth. This connection constitutes the logical coherence of thought. But rational conduct includes liberty, and liberty as choice ALL POSSIBLE IDEAS. 265 necessarily involves an alternative, a law from without, from above, to be obeyed or disobeyed. This law is given in the intuition of right. To this intuition the mind adds that of beauty, lying in somewhat the same direction, and expressing a certain supreme recognition of, and pleasure in, the thing that is aptly done. Beauty embraces two terms, form and spirit, and is found in the perfect union of the two. It thus arises at the junction of physical and intellectual facts in perfect expression. From the flow of either series of events there comes the suggestion of the infinite, as the all-encompassing thought. As causation changes not, neither waxes nor wanes, while liberty alone can make a beginning and shape events to a purpose, the idea of the infinite united with that of personality gives to us the Infinite, the completely comprehensive Being, from whom all things flow and to whom they return. Efficient causation backward brings events to rest in God ; final causation forward does the same thing, and so the force and thought of God interlace all things. § 18. We have no such complete oversight of the con- ditions of human knowledge as to be able to say, that the formative ideas now offered exhaust all its relations. We can only say, that no others seem to be needed, save these and those involved in them ; and that the general corres- pondence of these twelve ideas with the directions of human thought is very plain. A complete system of categories outlines all forms of human knowledge, and gives the sub- phenomenal links of thought which bind these forms to- gether. In testing the sufficiency of such a system, we must re- member that these ideas, in their application, give rise to many subordinate ones ; as time includes past, present and future ; and space, here, there ; above, below, in short all 2Q6 REASON. the words of relation under this pregnant notion. Under existence are shaped some of the primary axioms of logic ; A is A, A is not not- A, A is or is not. In the develop- ment of resmblance, we meet with agreement, disagreement, antithesis ; and under number with unity, plurality, totality equality. In connection with causation and time, we have succession, continuity, identity, action and reaction. Iden- tity here means the oneness of a substance or a form with itself. Identity, as the complete correspondence of a thought with a thought, arises under resemblance and spontaneity. Under space we have quantity, and under causation and space we have the subordinate -notions of substance, actual- ity, necessity, fate ; and under consciousness and spontan- eity, possibility, liberty, chance — as an event not ordered in reference to an end. Under truth, beauty and right, ap- pear the secondary ideas of the false, the imperfect, the wrong ; and on the other side, the perfect. With the in- finite arises its correlative, the finite. We do not strive to give all these subordinate forms, but simply to indicate a few of them and their relation to primary ideas. We subjoin the categories of Kant : ( Unity, Quantity. •< Plurality, ( Totality. { Affirmation — agreement. Quality. •< Negation — disagreement. ( Limitation — partial agreement. ( Substance — internal relation. Relation. •< Causality — dependence. ( Reciprocity — external relation. ( Possibility, Modality. < Actuality, ( Necessity. These are chiefly subordinate forms of primary ideas, ENUMEBA TION OF IDEAS. 267 and by no means exhaust the list. They neglect the pri- mary categories, and even include, as in quality, a partial generalization under the senses. In relation and modality, we have generalizations of dependencies involved in pri- mary ideas. These categories have therefore but little claim to attention from a psychological point of view, being constructed to express the logical forms of judgments rather than the ultimate ideas under which they proceed. CHAPTER Y. The Dynamics of the Intellect. § 1. We are to speak in this chapter of the growth and interaction of the intellectual powers, of the dynamic states of the mind. The Intuitive Philosophy has been censured, not without reason, by the Sensualistic School for contem- plating the mind only in its maturity, with no sufficient al- lowance for the results of previous conditions upon it, — for the effects of growth. This criticism we so far respect as to find a conspicuous place for truths which have been chiefly urged by such men as Spencer and Bain, always shap- ing them, however, to a new position and purpose. We are not prepared to admit any hereditary influences which vary the fundamental conditions of the problem of our in- tellectual nature. . The varieties of character, the growth of national and race distinctions, find explanation here ; but no sufficient proof has yet been given to establish, or even to render probable, the transformation of species by the accu- mulated changes of descent, with no increments of power. The past is not equivalent to the present. We must still regard each normal individual as the type of the race in its essential features ; nor are we ready to look upon any one of these faculties as the product of simply external condi- tions, the sum of growing, hereditary tendencies. When, on the other hand, we contrast the infant w T ith the mature man, it must, we think, be admitted that the complete activity of the latter, is very different from the tentative, partial movement of the former. It is to this de- GEO WTH OF MIND. 269 velopment of intellectual power that we first direct atten- tion. (1) The earliest distinct mental phenomena are doubt- less those of sensation, are physical feelings. These should be conceived as perfectly pure, that is as simple states or activities of mind — for our present purposes there is no dif- ference between a state and an activity of mind ; both are activities. These first sensations may be of one kind or of another, but are more likely to enter through the general sensational system than through a specific sense, to be sen- sations of pain, local or pervasive, demanding relief, and rising with acute, jagged certainty into the light of con- sciousness. It matters not what are the first sensations, since it is a changing series of sensations that invites attention. These are each simple, single, mental states known in the very fact of their existence as sensibilities. Separately, they are capable of no analysis, no division whatever. A pain, a taste are as individual as any objects of contemplation can be. To suppose these to reveal directly an external object, would be to suppose that the phenomena of matter become the phenomena of mind, and are known directly as such. We can only allow, then, that sensations disclose them- selves directly in consciousness, all beyond this is inferential. At this stage of growth, possessed of sensations merely, the infant is as ignorant of his own physical organs as of the world about him. He absolutely knows nothing save the varying pains and pleasures that flit through that unlocated region called consciousness, itself more often hidden under the cloud of dreams than open to the new light of waking perceptions. A tongue, a hand, an eye, a foot, are wholly beyond the scope of his knowledge ; nothing physical, ex- ternal to consciousness, is as yet recognized. In adult years we so instantly locate each sensation, that it seems to us that it itself declares its position. "We are doubtless to conceive of the mind as using the entire body, as making it directly 270 DYNAMICS OF THE INTELLECT. and immediately instrumental in reaching and influencing the external world. The brain is the chief seat of power, but is no more the mind, is no more a condition of its ac- tivity, than the nervous system generally, spreading through and through the body, and perfectly possessing it. But this instrument of the mind is not directly known to it. The mind uses the body and controls it unconsciously, in the dark, not in the light. Its shape, form, and members even, are all to be learned by experience. We may hesitate at first to admit this, but a little thought will compel the concession. If the mind in sensation itself knows and locates the in- struments of those sensations, then ought the mind to know its internal organs as well as its external ones. These are often independent sources of pain, and in the nervous sys- tem are as indispensable means to perception as the special senses ; yet the existence of the stomach, the brain, the liver, the interior formation of the eye, the ear, the ner- vous fibres and their ramifications, have all to be learned, must all be made objects of examination, and declare noth- ing to us directly of their own existence. These do not differ as regards our original knowledge of them from the tongue, the finger-ends, except in the fact that we neces- sarily learn the existence and form of the one set of organs much earlier than we do of the other. We return to the consideration of our first intellectual states — the flow of simple, subjective, unlocalized sensa- tions. (2) General sensations would be quickly accom- panied by more special sensations, arising from appetites and from special senses. That special sensations follow general sensations is evident from the fact, that those senses which are peculiarly full in their primitive data, as the eye and the ear, are called latest into action ; and also from the fact, that these senses in the earlier forms of animal life GEO WTH OF MIND. 271 are fitted to give only general distinctions. Appetite, taste, touch concur in the first wakeful experiences of the infant, and these are organically united as instinctive stimuli in the act of suckling. The coming reason finds its first con- scious experiences in this border ground of appetite and in- stinct — in sensations still obscure and mainly automatic in their action. It has been thought that the elephant owes much of his intelligence to the delicacy of touch in his trunk, and to its extended mobility. These facts certainly give him an unusual command of the conditions of in- quiry. (a) General and special sensations would immediately give rise to voluntary action in their gratification. The first conditions of life are met by automatic connections, but the rational consciousness begins at once to build its own constructions on these foundations. The limited num- ber of sensations are at first distinguished as pleasurable and painful, and each class is accompanied by more or less of spontaneous muscular effort, gradually changing into vol- untary effort, fitted to retain the enjoyment or escape the pain. The pleasures of touch and taste are especially con- centrated on the tongue, and the infant first spontaneously and then more consciously seeks the breast in gratification of its sensibilities. Later, the feeling awakens in the hands, and the child is not at ease till these are laid on the mother. In these earliest, tangible sources of pleasure, secured and maintained by muscular effort, the infant rests ; wanting these it worries, and moves inquiringly till they are re- gained. Later, other forms of sensation succeed ; the hand grasps more definitely, and seeks a greater variety of ob- jects ; the ear is cheered by the voice of the parent ; the eye is delighted with the brightness of the lamp-light, or with the sun-light. In these last cases, it is evidently more as sensations than as perceptions, more as organic impres- 272 DYNAMICS OF THE INTELLECT. sions, than as distinct cognitions, that the new objects find admission and confer pleasure. Slowly the eye learns to separate objects just at hand, and distinctly discern them, though possessed of no peculiar brilliancy. It recognizes the face of the mother, and at length follows, even into the distance, her retreating form. Still, its range, for a consid- erable period, seems limited, scarcely passing the verge of the cradle. (4) The mind is now ready to harmonize in action its different senses. Taste and touch are united, and each ob- ject that can be handled is subjected to the double test. The hand also follows vision, and strives to lay hold of each new thing. The ear learns to direct the eye, and the dis- tant voice wins the attention of both organs. The process of acquisition goes on till a definite mastery of each mem- ber is secured, its peculiar impressions discriminated, and the visible world unfolded and rolled out in its marvellous complexity of forms and relations. This movement, from the beginning, takes place under an objective form. The sensation is not enjoyed subject- ively, dreamily ; but objectively, really. The pleasures are attached at once to an object and a state ; thus also the pains. The spontaneous, muscular effort with which they are connected facilitates this external form of experience, by attaching enjoyment to objects independent of the senses themselves, to things momentarily lost and momentarily re- gained. Distinct, muscular exertion aids in distinguishing different states, in marking their attainment, maintenance, and loss. The objective character of early experience is also heightened by the degree in which it is composed of sen- sations — often the definite rugged sensations of pain — as opposed to perceptions, and later, of external, as contrasted with reflective pleasures. Language presents the mind as GEO WTH OF MIND. 273 especially passive and receptive in feeling ; and attributes the efficiency to the exterior occasion of the emotions. This we observe also in uncultivated, immature persons. Their attention is particularly directed to the objects and sources of pleasure. Their appetites and passions lead them inevitably to this objective life, to this hanging upon the external conditions of pleasure, this clinging to the bosom of nature. The notion of cause and effect — its own momentary enjoyments the effect — attaches the mind, as yet little more than a bundle of sensations, strongly and at once to the external world. Slowly it unfolds the facts of this world, the avenues and dependencies of its own pleas- ures, its senses and the things which minister to them. The internal rather than the external is overlooked. The senses are separated from the objects which affect them, but the attention of the mind is much later referred to itself, as truly subjective to them al\ If we were to neglect the objective character of exper- ience from the outset ; if we should suppose the mind for a time floating from sensation to sensation on the inner, tidal movement of its own phenomena, we should find in- creasing difficulty in making the transition, and in justify- ing it when we had made it. We are rather to regard the mind as at once borne outward toward the sources of its enjoyments, and as realizing these in and by their causes. We should likewise observe the great aid which muscular effort gives in interpreting and locating sensations. By this means the child at first automatically, later voluntarily, renews and discontinues its physical impressions, till the mind has matured its knowledge of them, their diversities and conditions. The relations of space are especially de- pendent on movement for their determination. The eye and the hand work with each other in exploring surround- ing bodies and intervening spaces, while a series of/ sensar 274 DYNAMICS OF THE INTELLECT. tions record the motions of the arms and fingers. By move- ment we repeat at pleasure the problems offered by exten- sion, and secure varying conditions for their solution. In this growth of the mind into the possession and handling of its instruments, into the rudiments of experi- mental knowledge, the appropriate regulative ideas are present doing their work, though of course they are unrec- ognized by the mind, as is the fact of sensation itself in the first feelings, or the fact of judgment in the early per- ceptions of likeness. It is the substance of experience, not its forms, the facts of experience, not its conditions, that occupy the attention. Experience is not for this reason destitute of form, or without conditions. The first when and where, though as yet unanalyzed, involve time and space, as certainly as the last. (5) One of the great labors of childhood is ready to follow the steps now taken, the conversion of sensations into perceptions by virtue of associative judgments. This change takes place chiefly in seeing and hearing, though the movement reaches the other senses somewhat, and, in the absence of the higher senses, may thoroughly transform them. This remarkable conversion takes place so early in life and so rapidly that we may readily overlook its extent. The ear and the eye, though more frequently indicating in- stantly the directions and relations of objects, obscuring the judgments of the mind by their rapidity, are sometimes so slow and uncertain in their decisions as to make the presence of their reflective processes conspicuous. We frequently have occasion to listen attentively in order to judge of the character and distance and nature of an unfamiliar sound. An object seen across the water deceives us, is farther off than we think it to be. Our estimates of the height of a cloud are very uncertain ; or of the size of unfamiliar objects, especially when our ordinary standards of measure- GEO WTH OF MIND. 275 raents are taken from us, and the proportions, as of a cathe- dral, are grander than those to which we are accustomed. The incomplete state in which the work still remains here reveals the fact, that size, form, direction, are to the eye solely matters of judgment. The eye may also be deceived. The fans of a wind- mill may seem to revolve in a direction opposite to the real one. We explain this as an error of the accompanying judgments, induced by an unfavorable position. The same form of error occasionally occurs in touch. The fingers being crossed, and the hand placed behind its possessor, he is often not able to decide which one has been touched. The ordinary accuracy of judgment is lost on account of the unusual conditions under which it is exercised. The vast majority of our localizing power being manifestly of an acquired and experimental character, we are inductively led to the conclusion, that all of it is of this nature ; and the more so when we find that the most steadfast and stub- born conclusions are occasionally at fault, when formed un- der changed conditions of judgment. The patient whose limb was to be removed, returning to a state of conscious- ness, can only determine by observation, whether it has been amputated. Indeed his sensations may lead him, through the accustomed reference of pain to the accustomed quarters, to feel the limb in its place, and this though weeks may have elapsed since it was lopped from the body. Touch is the sense whose localizing power is regarded as the most immediate, while its acquisition of this facility is most concealed from us by remoteness of time. The de- pendence of this sense, in common with others, on experi- ence for its localizing power, is seen in the fact, that on the finger-ends, where it exists most perfectly and in most con- stant use, we distinguish much more completely and ac- curately than on other parts of the body. A considera- 276 DYNAMICS OF THE INTELLECT. ble space must intervene between two points applied simul- taneously to the person elsewhere, before w r e can discern them as two ; they may approach very closely, and yet be separated in sensation by the fingers. This sense can, by special cultivation, when other senses are wanting, be made so much more perfect than it now is, be so filled and rounded out with instantaneous judgments, as to have but a slight resemblance to its former self. The raised letters of the blind are distinguished by most persons slowly and. with the utmost difficulty, while the trained touch glides rapidly along them, almost as the visual nerve moves over the printed page. The blind in some instances acquire a power and precision of touch inexplicable to us, and are enabled to carry on employments, like engineering and war- fare, from which we should regard them as entirely ex- cluded. Ziska was among the more distinguished of gen- erals. When the entire mind is directed to this avenue of communication with the external w T orld, it brings it by in- cluded judgments to an unthought of perfection, and widens it into a wonderful inlet of information. The most remarkable development of this lower sense is offered in the case of Laura Bridgman. " In 183 7 a deli- cate, light-haired girl, nearly eight years old, who at the age of twenty-six months had lost sight, hearing, and to a great extent the senses of smell and taste, from an attack of scarlet fever, was brought from her home in New Hamp- shire to the Perkins Institution for the Blind, in Boston."* A mind of great vigor was here cut off almost wholly from the external world, by walls of separation unpierced by the ordinary senses. In spite of these difficulties, her pow T ers were awakened to complete activity, and put in compara- tively full communication with the outside world by means of the one remaining sense of touch. Nothing could better * See an article on Mind, No. 14. GEO WTR OF MIND. 277 disclose the relative independence of the mind of any par- ticular sense, its own large contributions to every sense, and the very secondary weight in intellectual development of simply organic and hereditary forces, than the very won- derful triumph, on the one side, of instruction, and, on the other, of native power. The identification of physical and intellectual unfolding incident to normal growth quite dis- appears, and we find the mind reaching its goal by means peculiar to itself. (6) In connection with this development of perception, — a work which is never quite finished— those acquirements are made which combine in harmonious exertion mental and muscular action. Here belong graceful and dexterous movement, all forms of skill and art, and above all the learning of language. The perfect mastery in articulation of any speech is so great an attainment, that it is very rarely successfully accomplished but once in life. The thorough acquisition of a foreign tongue in middle and later life is deemed impossible. The following of the let- ters on the printed page with the eye, and the easy, instant and exact utterance of them with the rapid and complex play of the muscles involved, is, indeed, the acme of accom- plishment in a combined mastery of body and mind. This work, which the child does unconsciously, the laborious efforts of later life fail to repeat. Most busy and fruitful are these early years of childhood. Scarcely again do we learn so many and so perfect lessons in so brief a period. What the painter by slow analysis is able to reverse, pre- senting spaces, directions, distances, forms, on a plain sur- face of varying colors ; rendering the landscape, with an area of many, square miles, on a canvass of scarcely more square inches ; the child of a few years has learned to do with far more perfection, opening up and out the simple vignette of the retina, till it fills in every part the magnifi- 278 DYNAMICS OF THE INTELLECT. cent stretches before and abont us. Nor does the growing mind stop with understanding the things about it. It finds, in a marvelous way, those harmonized powers by which it moves among these objects, modifies them, and reproduces them in articulate speech. (7) With these acquisitions partially made, the mind is ready for the ordinary work of what we formally recog- nize as study — the acquisition of knowledge. This proc- ess will receive farther attention. (8) The last step in this rational movement is the clear recognition of regulative ideas, allowing each its due posi- tion and force. This makes the rational process rational to itself. Regulative ideas are not first present as objects of dis- tinct recognition, but as unthought-of principles which guide our consideration and apprehension of the phenomena before us. They may sooner or later, or not at all, be an- alyzed out as distinct elements of thought, though as un- conscious ingredients they are, in some one or other of their forms, present from the very beginning. It is not till the class of phenomena to which it pertains are brought forward, enter into the experience, and call forth the attention and judgment, that any one of these ideas, as that of beauty, of liberty, or of right, will find development, application and, latest of all, comprehension. § 2. The mind, once in possession and use of its facul- ties ; its perceptions and sensations made complete and in- stant in their action by the absorption of the needed judg- ments ; the intuitive notions all present, aiding to expand, locate, relate, and expound the several objects and events of experience, and give form and rational ■ coherence to thought, is ready for the acquisition of what is more com- monly known as knowledge. This mastery of conditions is so early, so spontaneous, so inevitable, that we more fre- GROWTH OF MIND. 27i> quently overlook it, and regard the complex result as imme- diate and direct. For the same reason we hardly expend a thought on the ways in which spoken language is secured by the child, and look upon education as commencing with the learning of the letters — the written alphabet. Yet the first acquisition, though imitative aud spontaneous, involves a more fundamental training and penetrates deeper into the physical powers by far than the second. The intellect, once in possession of itself, finds chief oc- casion to expand its knowledge under the notion of resem- blance. It is through this that it traces and interprets the lines of force ; and by these that it gains power. Yet we cannot accept the statement, that all judgments can be analyzed into resemblance, into agreement and disagree- ment ; and yet more do we not assent to the assertion, that these resemblances are sought for their own sake. Each regulative idea furnishes the ground of a distinct predica- tion, not to be resolved in its very essence by the most sub- tle analysis into any other. Moreover, resemblances are of value, and only of value, as they are the indices of agree- ing forces, as they are the surface marks which disclose the concealed lines of connection between objects and events. Power is the fundamental element of knowledge, that which makes its search pleasant, and its acquisition profita- ble. The desire for knowledge which gives no power is like avarice, the morbid play of a just impulse. To know the exact number of leaves on a tree, their position and form, the precise way in which some ancient but insignifi- cant event happened, the very words in which some second- rate poet expressed himself, is to know to no good purpose, is to have the semblance, not the substance of wisdom, the shell, not the kernel of truth. Resemblances which are accidental, which betray no relationship, as the size and form of a boy's marble when compared with the pebbles on 280 DYNAMICS OF THE INTELLECT. the beach, or the agreement of sounds and signs in unre- lated languages, have no interest and subserve none of the purposes of knowledge. A resemblance which is a mere resemblance, which casts no light on the past, and gives no clue to the future, which discloses none of the forces at work in the world, is unfruitful, and the knowledge of it of no value. That which makes the search after agreements so unremitting are the axioms of causation: that Like causes are followed by like effects, and that Like effects in- dicate like causes. These transform a knowledge of real central agreements into power, put us in connection with the plan of the world, enable us to bring new forces into it, and take new and coveted effects from it. Resemblance is simply the key to the storehouses of forces. Uncultivated minds, so far as they pursue knowledge at all, do it under this form ; an observation of resemblances with reference to an ulterior possession and control of causes. The savage distinguishes between the different kinds of tim- ber, because he expects the same external indications to remain the accompaniments and marks of certain interior qualities of strength, weight, elasticity. A bow of the same material he believes will exhibit the same good points with which he is familiar ; a spear of like wood possess like pliancy and toughness. Language comes in to mark and hold together for the mind these agreeing things, by which the implements of man, and his successive wants are to be supplied. Science, the advanced and complete movement of thought, is but a more rigid separation of like with like, a more careful selection of central qualities, a complete and interdependent classification of objects, both that the re- sources of the globe, in all its ministration to human life, may be laid open, and also, that the concealed chart of laws, according to which the events of the present come pouring GROWTH OF KNOWLEDGE. 281 down from the past, and go forth to occupy the future, may be disclosed. While our experience, then, finds its first efforts directed to resemblances, these lead to pro- founder inquiries into causes, those links of force which lengthwise and laterally bind together the physical events of the world. At length the purely objective character of knowledge passes somewhat away. The mind gives heed to the agent as well as to the instrument. Having acquired power, it learns to value itself, the possessor of that power. With more pure reflection and subjective attention, it inquires into its own faculties, and the laws of their control. Now come forward new intuitive ideas, liberty, right, beauty, disjoining philosophy from science, and setting the first over against the second as independent of it, and comple- mentary to it. This change and jar of transition consti- tute the great danger attendant on the acquisition of this form of knowledge. The forces and notions of the one field are intruded into the other, and those who suppose themselves the most patient of inductive philosophers are really visionary theorists, adopting a disguised a priori method; since they bring to a new department methods and conceptions alien to it, and refuse, vacating the mind of prejudice, to examine and classify these fresh phenomena according to their inherent characteristics directly observed. There is thus more or less of vibration between the two fields. Now the philosophical, now the scientific concep- tions rule the inquiring mind. The present, passionate, physical researches and methods of thought are sure to be followed by a recoil against forms of inquiry so one- sided, so unscrupulous in their application. The deductive method was never more arbitrarily applied to science, with less correction from experience, than is the inductive method now to philosophy, bringing with it the forms and 282 DYNAMICS OF THE INTELLECT. forces of physics. Induction transferred from one field to another, without fresh starting points and new limitations, :s really disguised and unsafe deduction. Knowledge stalks on with alternate strides, and, in the rhythm of progress, the swing of one limb makes way for that of the other. The movement is in truth somewhat more complex than here indicated. In the earlier periods of thought, the notion of spontaneity crowds upon that of causation ; phys- ical events are displaced by spiritual ones, or are referred to supersensual causes. Hence come the endless throng of superstitions. These at length are driven back by science, and the reverse error commences. Causes take the place of powers within the domain of mind itself, and an unrea- sonable skepticism displaces unreasonable belief. Later the mind learns to balance its convictions, and make them proportionate within themselves. The mind measures all things by the scope of its own powers too much to rest on the naked facts of the world. The forces which they disclose and the plan which they re- veal, with the wisdom of its conception and the kindness of its execution, push the thoughts farther back to the source of these truly intellectual and moral elements. The progress, also, which is discovered, together with that irresistible claim wmich the mind institutes for completion, for ends reached and fruits acheived, push it forward in thought, and lead it willingly to gather up the issues of existence into the hand of Him who gave it. That this movement may be final, that a true compass and circuit, source and conclusion of the actual, the finite and the necessary, may be found ; that the mind may rest in one last stroke of comprehension, it brings forward the highest of its intellectual solvents — the Infinite, the Absolute. A free and holy personality is to be made to the mind and heart, the cause and compass of the universe. This movement becomes complete and assured in connection GROWTH OF KNOWLEDGE. 283 with revelation, an outer voice, which takes hold on inner powers, and gives steadfastness and certainty to their con- clusion. The mind is not left alone to travel these outly- ing highways of thought. In this growth of knowledge through science, philoso- phy and theology, deduction and induction play interming- led and inseparable parts. Deduction from necessary ideas, from definitions and axioms intuitively conceived under them, gives us first principles, the ever present instruments of inquiry. Induction is nothing without a theory, a con- ception of some sort, running side by side with its classifica- tions, guiding and interpreting them, and ready deductively to furnish shining strokes of exposition. The theory with its derived , conclusions is most impotent and misleading, save as induction presides at the birth and growth of it. The wise mind is always laying up the facts of nature, like stones in a building ; but laying them up under a plan which it has caught by penetrating beneath the surface ; by interpreting signs and relations unintelligible to the merely physical eye. Here, then, in the growth of general knowl- edge, we have the counterpart of that which we find in the individual mind. The elaborative faculty, the understand- ing, is ever playing between the sensations and the intu- itions, weaving them into a rational experience. In like manner the philosophy and the science of the world are bringing downward, deductively, the theories of the mind ; are bringing upward, inductively, the phenomena of nature and mind, and slowly uniting them into one compact web of knowledge ; the exposition running as light through the facts, and the facts embodying and presenting the expos- ition. The one process is as necessary as the other, the woof as the warp. § 3. We wish to mark briefly the means by which the mind advances in acquisition — the instruments of iutellec- 284: DYNAMICS OF THE INTELLECT. tual growth at its service. Sensations, perceptions, the data of consciousness enlarge for it the material of thought, and are themselves simple, ultimate forms of knowing. Nothing can replace them. Colors, sounds, odors, flavors are appre- hended exclusively through the organs by which they enter. Further, they give us inexhaustible material for inquiry, facts to which the mind may bring its explanatory processes, and which it may work up into knowledge. Intuitions without these, as mere intuitions, would remain empty for- mulae, intellectual solvents with no mysteries to resolve. Next come judgments. These are the steps in the ra- tionalizing, comprehending process. To be able to form a judgment, is to be able to put forth true intellectual effort ; is to turn the key in locks that guard all knowledge. It implies completeness of mental furniture, the entire mate- rial of growth. Simple judgments are the staples of knowl- edge ; while they may be formed under any idea, that of resemblance assumes chief significance. All classification proceeds through this, and is a first and last step in progress. It is by a comparison of qualities that our knowledge of the objects about us becomes serviceable. Much, perhaps the larger share of our progress, is made by simple judg- ments, related indeed to one another, but not interlocked in reasoning. By a series of inquiries, we place objects in their appropriate classes, and furnish them ready both for our intellectual and physical uses. Seasoning, or interlocked judgments, follows simple judgments as a means of progress. There is considerable disagreement as to the forms and character of reasoning, arising largely, we think, from a different use of words. One form of procedure is covered by the words reasoning and logic as used by Hamilton, and another as used by Mill ; while others combine, with more or less confusion, the two uses. Hamilton, by a definition, confines the province of FORMS OF REASONING. 285 logic to the necessary laws of thought, or practically to the demonstrative evolution of conclusions from premises that are given. He does not inquire into the manner of ob- taining the premises, but only into the certainty and safety of that purely intellectual process by which, as verbal prop- ositions, they are found to hold those other verbal proposi- tions known as conclusions. The whole movement is thus detached from facts as facts, and, according to the general use of words, is, when reasoning at all, deductive reasoning. That is, the conclusions are wholly contained in, and de- monstratively taken from, the premises. Hamilton gives a technical and peculiar application to the words inductive and deductive, regards both forms of reasoning as equally demonstrative, and leaves wholly out of his logic that true induction, usually so-called, to the elucidation of which Mill has given his entire strength. Induction in its com- monly accepted meaning, the establishment of a general principle through a limited number of specific examples, is all the reasoning which the sensualistic school can consist- ently recognize. What others regard as deductive reason- ing, they are compelled to look upon in ultimate analysis as inductive. Deduction can be nothing more with them than the re-statement of a specific case already included in the establishment of the general principle, or major pre- mise, from which it is now taken. ~No conclusion is strictly demonstrative, since it is in advance of the premises on which it rests. The degrees of evidence for new state- ments, statements not confirmed by direct observation, vary with the amount and character of experience on which they rest. The entire system of logic, therefore, as presented by Hamilton, has for them comparatively little value. It is a cunning play upon words, rather than an estimate of facts. They are interested in the growth of laws, principles, out 286 DYNAMICS OF THE INTELLECT. of those separate instances which are only to be gathered and interpreted by patient, careful, and often doubtful in- duction. Each party thus neglects a valuable field which the other exclusively cultivates. All that Mill regards as reasoning, Hamilton scornfully rejects from the province of logic as invalid, as not presenting with certainty the con- clusions in the premises from which they are taken. Mill, on the other hand, can only look on the complicated syllo- gisms of Hamilton as a cumbersome restatement of work already done, of knowledge already gained. Much is undoubtedly included by Hamilton, in the for- mal expansion of his terminology, as reasoning, which would generally be regarded as simple statement, as the fruit of single judgments, as the results of perception. This desk contains this drawer : This drawer contains this paper : Therefore this desk contains this paper. These propos- itions form a syllogism under one of the forms into which he divides deductive reasoning. Most would regard them as in no proper sense reasoning, but rather as a formal, un- servicable statement of a fact, learned by observation. So also his inductive reasoning is made up of cumbersome formulse of classification. " Gold is a metal, yellow, duc- tile, fusible, and so on : These qualities constitute this body (are all of its parts) : Therefore this body is gold." Here is no argument properly so called, but the rendering of the results of the experimental test of a bit of metal, with the accompanying act of classification. There would seem to be room in a logic, covering all the forms of reasoning, and those of reasoning only, both for deduction and induc- tion, using the words in their more general and generally accepted meanings. An important branch of logic finds representation in Hamilton and Mill respectively. What is reasoning % It is the reaching of a new con- clusion, certain or probable, by means of two or more inter- FORMS OF REASONING. 287 locked judgments. We would lay stress on the word new, and on the words certain or probable. Our necessary ideas and our theories suffer expansion by a purely deductive process. Geometry is a deductive science, derived from intuitions, definitions, and axioms. Astronomy and me- chanics are full of pure deductions, resting on conceptions of force confirmed by experience. How much is involved in certain, simple statements, we often only learn by a series of related judgments. This is one of the earliest forms of scientific reasoning, and presents in mathematics, pure and mixed, its most extended and serviceable results. The cer- tainty, and, when fitting data are found, the celerity of its conclusions, abundantly explain its fascination, and the po- sition it has held in investigation. The introduction of a mathematical unit, and application of number to a subject, have usually been the signal for a rapid advance. This deductive reasoning rests on intuitive steps, and will readily fall into a syllogistic form. The syllogism is perfect ; for the premises as premises, in their very state- ment, are seen to contain the conclusion. No outside cir- cumstances affect their relation. Inductive reasoning, on the other hand, deals only with probabilities, because it pertains to things imperfectly known and to facts whose conditions are ever changing. It rests at bottom on the intuition of causation, the simplest state- ment of which is, Every event must have a cause. Its cor- ollaries are, that every effect measures its cause, that the two are exactly commensurate, and, that sameness in one is proof of sameness in the other. These spring from the original, independent conception of causation. Proof un- der this notion, would be as certain as under the ideas of space and time, were we always dealing with perfectly fixed and perfectly known premises. We do not by observation so penetrate the nature of objects, and the character of com- 288 DYNAMICS OF THE INTELLECT. plex phenomena, as to be sure of the elements present, and sure, therefore, of the effects that may be expected. We do not know exactly how far one wood differs from another, one metal from another, one element, so called, from itself at a former period. Much less do we know all the circum- stances which affect the complex problems of life, which in- fluence the growth of a tree, which are concerned in the health of a man, in the welfare of a community. We here, therefore, advance from one case to another along uncertain links of likeness, not knowing positively whether the agree- ment covers the essential points of the two cases or not. The various degrees of likeness are identity, sameness, re- semblance, analogy. In induction, by which from several examples we infer a general principle, we are proceeding on a resemblance more or less obscure, hence more or less uncertain. Differ- ent cases stand on their own independent merits, and the probability in each is in proportion to the certainty with which the agreement in the example covers the force or forces involved in the causation under consideration. That all magnets attract iron, is a conclusion on which we rest with entire conviction, having by such uniform observation traced this* result to this cause. Yet it is not an impossi- bility, that some new substance or combination of sub- stances should exhibit the other properties of a magnet with the omission of this. We cannot say, how new con- ditions of action may modify the force termed magnetism. Now, by far the larger part of the reasoning of natural science and of e very-day-life is of this character, creeping from resemblance to resemblance, and unable to affirm of its best conclusions, that they are demonstrative. To this reasoning, the syllogism is not applicable, since the premises as premises are partial, and do not contain the law in its full breadth which is to be evolved from them. This phil- FORMS OF REASONING. 289 osophy of experience, therefore, can lay no great stress on the syllogism. The only service it can assign it, is that of a convenient re-statement of conclusions already arrived at, and this, not in the exact line in which the first, the real ar- gument lay. This was on the road upward to the principle, whereas the syllogism lies in the way downward to a specific example included under it. When inductive matter re- ceives syllogistic statement, either the statement is defec- tive, or the general principle is assumed, and then the case in hand is taken from it. The argument by which we mount to a general law, does not suffer a syllogism ; the seeming argument by which we descend to a particular fact is but a restatement of previous knowledge, and yields a syllogism deductive in form. Of the defective, inductive syllogism, the following is an example ; The metals A, B, C, represent (not are) all metals ; A, B, C, expand under heat ; therefore all metals expand under heat. This result is proximately not absolutely true. If the law had been established by sufficient observation, that all metals expand when heated, the following would be the deductive syllogistic statement of a single fact covered by it. All metals expand by heat : A is a metal ; therefore A expands under heat. The confu- sion which has arisen in the various estimates of the value of the syllogism seems to find its sources in the language employed, in two restricted definitions, and, more than all, in failing to estimate the influence of different philosophical: systems on the respective methods of logic. The two forms of reasoning differ (1) in their data.. Deduction starts either with intuitive data, or with state- ments of facts regarded as complete ; induction rests upon facts and statements obviously partial. (2) The connections between the premises in the one form of reasoning involves axioms only ; in the other it includes one or other of all the shades of resemblance contained between perfect, identity 290 DYNAMICS OF THE INTELLECT. and mere analogy. Between these two there is a contin- uous series of weakening agreements, which we have di- vided into parts by the words, identity, sameness, resem- blance, analogy. Identity is a secure basis of reasoning, but we cannot absolutely affirm it ; sameness, a close agreement in qualities, admits considerable doubt ; resemblance, a less close agreement, renders the argument insecure ; while an- alogy, an agreement in relations, may only call out a pre- sumption. (3) The two forms of reasoning differ, therefore, broadly in the force of the conclusion ; in the one it is de- monstrative, in the other it has every shade of certainty less than the highest. (4) They differ also in form. The syllogism belongs to deduction only ; if applied to induc- tion, it is either incomplete or assumes the thing to be proved. Life is thus permeated with both elements, with certainty and with doubt. Wisdom and strength are found in handling the two conjointly. We have followed the ordinary division of reasoning, though we doubt its completeness. We prefer to regard rea- soning of three orders, according as it deals with pure intui- tive data, with empirical data, or with these data assumed as absolute. Mathematics is the chief example of the first order. Pure mental conceptions are unfolded through long trains of reasoning into appropriate formulae and proposi- tions. The connection under which this is done is that of equality or equivalence. The reasoning proceeds by affirm- ing — or denying — an exact equality. The conclusion is identical with the premises as clearly understood. This rea- soning may be termed production, the leading onward of data from one form of expression to another. In dealing with things, we are constantly embarrassed by the want of perfect insight. We hesitatingly reach a general conclu- sion by combining under it few or many examples. This is induction. Such a conclusion being reached, and ex- ATTENTION. 291 pressed as a class, we may, either in extending our knowl- edge or in reviving it, reason from the class to any one of its subdivisions. This is done under the Aristotelian prin- ciple, that what is affirmed of a class is affirmed of each member in that class. This is deduction, and its logical formula is A is A. § 4. The intellect being thus furnished with faculties, and stored with their fruits, we inquire into its control over them, its directing influence. We speak first of the government of the thoughts. The mind can direct the eye, the ear, to any object it chooses, and command their pro- longed attention. (1) It can make any object the subject of protracted contemplation, and confine the analytic and reasoning processes to it. It can intensify and guide its mental activities in degrees varying with the power which previous practice has given it. This voluntary direction and handling of faculties is attention, and is referable to that personal force from which all the faculties as separate directions of action spring. The number of objects which can at once be made the subjects of attention has been a question vigorously de- bated. The mind seems to be single in what may be termed its line of movement, its chain of connections ; but to be able to unite in this movement many diverse things. Our thoughts braid into one experience, link in one argu- ment, diverse subjects; they proceed by junction and in- clusion, evening and strengthening the thread with mate- rial drawn from the right hand and the left. The reason why it has been doubted whether the mind can attend to more than one object is in part found in the fact, that the very effort of the mind to decide the question serves to occasion that fixed, full, complete attention which is concentrated on a single object, and leads to the partial exclusion of other objects. Of course we cannot give the 292 DYNAMICS OF THE INTELLECT. entire attention to two objects as two, struggling in the same instant to contemplate them with distinctness separ- ately. Failing in this, we have hastily concluded that the mind can attend to but one thing at a time. Let the thoughts move freely, and it seems obvious that we do con- sider several objects at once, some of us more, some less. The shepherd counts his flock as they pass before him or stand around him. He will more likely do it by threes or fives, grouping the numbers by a stroke of the eye. One practiced in dividing paper into quarter quires will instan- taneously, on the ruffled edge, select the number six, and with astonishing rapidity run through the pile. This ten- dency in enumeration to divide objects into greater and smaller groups, according to the degree of skill, plainly re- veals the power of the mind to contemplate at once sev- eral objects. Indeed, were the mind limited to absolute singleness of attention and direction, its states would suc- ceed each other in a disconnected and independent form. Every judgment involves two terms and a relation between them. A more important question arises, (2) as to the power which the mind possesses in introducing to itself the ob- jects which it may afterward consider. So far as these are external objects it may open for them the avenues of percep- tion, and then select among them those which it will more carefully observe. It may also seek the locality of remem- bered or described objects, and thus prolong their consider- ation. In this direction, the mind is limited, first to things that are ; second to those among these known to it and ac- cessible to it. A large share of the government we have over our thoughts is found in our mastery of the external conditions of life, of situations and circumstances. A deeper inquiry lies in the questions, (3) How far does the mind control the order of ideas that are passing through it ? POWER OF MIND OVER ITS IDEAS. 293 and How far is the flow established and maintained by in- dependent connections ? The doctrine of association has been the occasion of ascribing a dependence and passivity to intellectual con- nections which we deem wholly false. The association of ideas has been accepted as an ultimate fact, and itself with- out explanation, been proffered as an explanation of every other fact. This solution has proceeded under physical an- alogies, especially those of habit in the body. A form of activity, often returning to the muscles, so interlocks the nerves and muscles, so passes over their connection from the voluntary to the automatic region, that the mere fact of repetition becomes a reason for many movements not di- rectly intended. Under the suggestion of this fact, ideas are spoken of as associated, and this association seems to be often thought of as involving some direct, almost mechani- cal connection of one idea with another ; as if the first evoked and drew on the second by an immediate force. Thus we have such expressions as the " cohesiveness of ideas," "the principle of cohesiveness," "the property of plastic adhesiveness," " the tenacity of association." Those who thus use the law of association, refer the order of ideas in the mind to it, and give the mind itself but little con- trol over them, beyond that of hastening or checking their movement. It is the living power of the mind, rather than an in- trinsic coherence of ideas that combines them into thought, and locates them in revery. The memory proceeds along the connections of place, time, resemblance and causation, because these are the forms under which objects are prin- cipally presented to it ; and the groups of memory princi- pally determine the connections and dependence of concep- tions, when they return to the mind. One object tends to restore in memory, more or less distinctly, the entire group 294 DYNAMICS OF THE INTELLECT. of which it forms a part, and its earlier and later relation- ships are renewed, because the memory is by it directed to that portion of experience in which it has played a part. Ideas are thus interlocked in memory, and return to the mind surrounded more or less completely with their ad- juncts, their companions in previous knowledge. A second ground of association is that of the deductive dependence of ideas. The logical power of the mind on the presence of a part expounds it by reference to the whole ; or on the presence of a whole unfolds it in its parts. These two forms of association correspond to the two methods of acquiring knowledge. Observation, induction present objects as physical wholes, and the memory so retains them ; analysis, deduction unfold ideas, and the memory and the logical faculties combine to repeat, on fit occasions? these processes. The cement of ideas is the living forces that use them, not a dead adhesiveness belonging to them as ideas, or dependent on the nervous conditions of their pres- ence. It is not a reverberation of tissues, but of thoughts, to which attention should be directed. We see at once, then, that the power of the mind over its trains of ideas is greater than many are willing to admit. Take any one moment, with the tendencies aud memories of the past fixed, the circumstances of the present estab- lished, the current of the desires strong and declared, and thoughts and conceptions may seem rather to sweep inde- pendently through the mind, a deep, uncontrollable cur- rent, than to be called forth and used by it. Take, how- ever, a longer period, let the mind desire to assume con- trol, and this appearance of helplessness will pass away, and our impressions will be reversed. Times are set apart to definite inquiries. The passing hours bring each its suggestion of its part of the plan. The memory is more and more stored with material suitable to the investigation POWER OF MIND OVER ITS IDEAS. 295 or to the effort in hand. External circumstances favorable to the inquiry, are secured. The desires, quickened by exer- cise, lend their aid in constraining and spurring on the thoughts. The purpose, kept in view, evokes from the memory on each new exigency, every fitting idea in the increasing circle of its information, while the logical con- nection of ideas guides the pursuit along the right trail. Under these conditions, we shortly behold an intellectual power which works as intensely, as directly, as uninterrupt- edly toward its end, as the engine whose valves and pistons and wheels are driven by a mechanical agency. A way- ward pleasure of vagrant connections may turn the thoughts for a moment aside, but not more frequently nor more un- fortunately than the flower or the fruit the wayside traveler steadily pursuing his journey. States and powers of mind are not indeed instantly determined, immediately gained ; but tendencies are established, and control acquired as cer- tainly here as in any form of effort. The person himself may determine within the limits which the surrounding world presents him, what shall be his resources of thought, and what the motives calling them " into act and use." The whole movement is a living one, under a living in- telligent power, and is no more to .be expounded as a dead process, an adhesion of one thought to another, than is the life of the plant or of the animal to be traced to simply chemical forces. The very secret of life is to combine ma- terial into living organs ; the very knack of mind is consec- utive, coherent, self-supporting thought and self-directed action. If we look upon the phenomena of mind as passing on under its observation, its immediate power over them is found in its ability to select any one, and to intensify and prolong its consideration. This is attention. Its second more comprehensive power over the flow of experience lies 296 DYNAMICS OF THE INTELLECT. in the introduction of new terms by memory, by following out logical relations, or by altering external conditions. But the full sweep of its government, embracing long periods is found in shaping distinct lines of inquiry and of action, and in adjusting circumstances to their accomplishment. The whole stream of experience may thus be turned into new channels. § 5. The last point on which we have occasion to speak under the dynamics of. the intellect is the difference in mental endowments between the brute and man. We are necessarily somewhat theoretical in handling a subject so much beyond direct knowledge ; but trust our theory will commend itself as the simplest explanation of the facts, with the -least assumption and the fewest forces. There seems to be no proof, that any animal, the most sagacious, possesses any intuitive ideas, and consequently that it forms any judgments properly so-called. There is no conscious estimate of the value and bearings of sensations, no classi- fication of them inductively, no conclusion deductively drawn from the premises as such. Sensation, perception, memory and imagination evidently belong to the higher animals, and by these faculties, we believe, all the intellec- tual phenomena they present can be readily explained, while the ascription of fuller powers than these to them brings difficulties which cannot be easily met. To those who doubt the possibility of presenting the appearance of reason- ing processes with these limited and elementary powers, we would commend the works of Bain, and kindred philoso- phers, who, with patient and adroit analysis, think them- selves successful in resolving the phenomena of mind, in their most exalted forms, into the play of sensations and perceptions. They at least render this service to true phi- losophy, of enabling us to explain brute life, without ele- vating it in gifts to a rational platform. Those who do not ANIMAL POWERS. 297 believe that the races of men could have sprung from one pair, may be referred to Darwin ; those who cannot explain the sagacity of the dog, his apparent sense of shame and ap- proval, without endowing him with the entire circle of human powers, moral and intellectual, may well find profit and conviction in the works of the sensualistic school. The truth is, memory and perception can present, with close agreement through quite a wide range of conduct, an image of rational and moral behavior. Memory can unite impressions and their appropriate accompanying actions in permanent associations, exhibiting results as safe and saga- cious as if the union had taken place by judgment. We constantly interpret the conduct of animals under the analo- gies of our own experience ; an act more unphilosophical even than for the accomplished and sensitive man to infer the exact counterpart of his own feelings in the clown from an agreement of external actions. The aspen trembles with- out fear : the dog skulks and crouches in apparent shame without a sense of guilt. The severe tones of voice, the sharp eye, punishment associated in experience with like ac- tion, are a sufficient explanation of conduct which we often hastily regard as showing the germs of a moral nature. Indeed, this inferring the same sweep of thought and feeling from coincident actions in man and in animals leads constantly to the most insecure and unfortunate conclusions ; unfortunate when they are made the grounds of cruel ex- actions, and the tyrannous handling of domestic animals. Says Professor Whitney, in his treatise on Language and the Study of Language : "A dog, for instance, as surely apprehends the general idea of a tree, a man, a piece of meat, cold and heat, light and darkness, pleasure and pain, kindness, threatening, barking, running, and so on, through the whole range, limited as compared with ours, of matter within his ken, as if he had a word for each. He can as 298 DYNAMICS OF THE INTELLECT. clearly form, the intention, ' I mean to steal that bone, if its owner turns his back and gives me a fair chance,' as if he said it to himself in good English. He can draw a complex syllogism, when applying to exigencies the results of past ex- perience, and can determine ■ that smoking water must be hot, and I shall take good care not to put my foot into it ; ' that is to say, ' water that smokes is hot : hot water hurts : this water is hot : ergo it will hurt my foot,' "—page 414. While making no objection to the spirit of the pas- sage, we regard its philosophical implications as all wrong. Keen perception and quick association by an active, reten- tive memory offer a complete explanation of the facts in- volved, and of kindred ones, without supposing the pres- ence of a single act of judgment, of one thoughtful junc- tion of premises and conclusions ; nor the recognition of any general idea or general principle. The fear of the master is present, and the desire of the bone ; withdraw the first, and the last comes into unobstructed operation. The sight of steam and a delicate sense of heat, associated with pain under exposure, apply as direct a restraint to action as the shutting of a valve to the ingress of water. The difference between the two cases lies in the fact, that in one instance, the restraining power appears in, and works through, con- sciousness, and in the other it does not. That association is sufficient to explain the apparently thoughtful action of brutes, is seen, in the first place, in the way in which their sagacious tricks are acquired. A cow learns to open a gate ; but how ? First, by acciden- tally or impatiently rubbing her head and horns against it, and thus loosening the latch. This process, repeated once or twice, establishes a connection between the act and its results, and later, when she wishes to be free, she worries the gate open. A change of fastening relieves the diffi- culty, not because the new method of reaching the latch is FACULTIES OF ANIMALS. 299 necessarily impossible to her, but because it is not accom- plished by the same blind movement which removed the previous one from the catch. The horse learns to untie himself ; vary the knot, and his skill disappears. That the protracted experience of the brute must yield to it not very unfrequently a repeated concurrence of the same cause and effect, 'and thus enable it to reach the one through the other, in those cases in which appetite impresses on the memory the connection, is obvious. Indeed, that this happens so rarely, is quite as much a matter of surprise as are the few cases of apparent skill. We know the cunning, vicious tricks which a street animal acquires ; but we also know that in a keen appetite on the one side, and much persecu- tion on the other, it has under the law of association the most unwearying and vigilant instructors. The restive horse, scorning the restraint of fences, has compounded his education of short and easy attainments. The spiteful nag, grazing on a village common, has learned the ins and outs of advantage, the safeties and dangers of provender, by many a sharp thrust and sturdy thwack, and it is not surprising that it has quite a store of ideas pricked into its tough, retentive hide. The same truth is seen in the methods of training — as of the dog and the horse. The first effort is to establish a definite association of reward with the action to be done, and one of suffering with the action to be avoided. Says a skillful handler of horses, " The difficult point is to secure the right action in the first instance. Every approach to it should be at once recognized and encouraged. The animal should be petted and rewarded at each repetition, till the thing required becomes habitual" — that is, till the right association is established. On the other hand, the wrong action is painfully and peremptorily checked, till the ten- dency to it is corrected. In the meantime, the fitting words 300 DYNAMICS OF THE INTELLECT. of command uniformly accompany the discipline, and it re- quires no intelligent apprehension of language to lead the horse to stop at the word whoa, when it has been repeatedly accompanied with a severe jerk of the bit. A change of masters always interferes with the training of animals, as for instance of a yoke of oxen, because there is a breaking up of associations, a diversity, and hence a confusion of methods. Passion and hasty punishment like- wise retard the education of an animal. The reason is obvi- ous under the princple of association. If the brute were in a measure rational, he might interpret aright the flogging, and profit by it ; but, acting under association, his con- sciousness is simply flooded with suffering and fears, and henceforth, on the like provocation, he becomes restive and excited in anticipation of a similiar, painful experience. Punishment that is not proportionate to the wrong, or does not immediately follow it, and spring as it were out of it, is of no avail. The association is lost, and no reasoning process is present to take its place. All the facts of skillful and successful discipline in animals come in to corroborate the assertion, that action, with them, follows the appropriate perception under fixed associations. But it may be asked, why is not the opposite sup- position of reasoning faculties an admissible one? We answer it involves at once the entire circle of regulative ideas, postulates more powers than are needed to explain the phenomena, and it is not consistent with the fact that brutes exhibit no such growth as should, in some instances at least, follow the rudimentary possession of such high en- dowments. If the animal reflects, there is no reason why he should not occasionally express by language, at least by signs, the results of that reflection. One rational thought is not possible without the possibility of two, of three, of many thoughts. One syllogism carries with it the entire FACULTIES OF ANIMALS. 301 logic, and such powers would quickly command expres- sion. This utterance of judgments we should the more anticipate, as the most sagacious brutes are in constant company with man, and might learn from him, in some instances, vocal language, in others, sign-language. The fundamental difference in mental action between the brute and man, incident to the absence of intuitive ideas, is the fact, that man alone deals with abstractions, generalizations, conceptions. The animal has to do directly with things and their images. All analysis proceeds under an intuitive idea, and no sooner reaches an abstraction than it calls for a sign, a word to express and hold fast the prod- uct. The animal can not be taught language, because it has no occasion for language, lacking abstractions either of qualities or of relations ; and the animal never is taught language, no matter how many words it is made to repeat, or how many sounds it associates directly with concrete feelings and actions. Without the demand occasioned for language by an incipient act of abstraction , of some sort, language is impossible, and with this demand it is unavoid- able. Contemplate things solely as present and in the con- crete, and the senses quite suffice.* The only way in which a brute does show intelligence is in action, and this may as well spring from associa- tion as from reflection. The utmost efforts of instruction expended by man on animals, even when it has reached to the mechanical repetition of words, has only secured re- sults in conduct readily referable to slow, established, and patiently confirmed associations, the varying perceptions of the animal putting it, in connection with accompany] no- pains and pleasures, on the clue of the behavior designed for it. * For a fuller discussion see Comparative Psychology, or Growth and Grades of Intelligence, Chap. VII. 302 DYNAMICS OF THE INTELLECT. Moreover, if rational ideas are conceded to the brute, they must be granted in a more powerful and perfect form, rather than in a less perfect form than to man. The chicken, the young of many animals, almost immediately begin to successfully estimate all the relations of objects in space. They evince more mastery over them at the end of a few hours or days than does the child at the close of as many years. If, therefore, any judgments intervene in this process ; if the perceptions do not directly, by an immed- iate transfer of stimulus, secure and guide the motion ; if there is not the same spontaneous completeness in the action of the mind that there is in that of the body, what a mar- velous, unaccountable rapidity of development should we have here. We must exalt in accuracy, ease and celerity the reflective processes of the animal far above those of man. This seems, to us at least, a reductio ad absurdum*. But, if the sport of the lamb, its leaping and running ; if the flight of the bird, and the ease with which it hits and rests on the spray, indicate no conscious recognition of space, the presumption is that other less astonishing powers have no basis in reasoning or in intuition. We object, also — though this consideration may have little weight with some minds — to the character which this idea of reflection ascribes to the consciousness of the brute. A thoughtful animal would be one of the most unfortunate of beings, the incubus of its physical structure weighing down its destiny. Rational hope and fear to a being like this would be an unnecessary and cruel source of suffering; nor do animals often show apprehension and alarm except in the immediate presence of danger. But it will be said, there are examples of sagacity on the part of animals which candor forbids us to refer to association, to anything short of reflection. To this we answer, these examples require more searching inquiry as FACULTIES OF ANIMAL8. 303 to their exact form and value than they have received, as the shades of action that distinguish association and reflec- tion are unobtrusive and delicate ; and few are aware of the extent of results easily within the scope of association alone. Farther, we are not considering what would be re- ferred in man to reflection, but how much is possible to quick perceptions, strong appetites, and a ready memory, when they are left to act alone, and are not therefore su- perseded or embarrassed by reasoning. Says the writer last quoted, " It has often been remarked, that the crow has a capacity to count up to a certain number. If two hunters enter a hut, and only one comes out, he will not be allured near the place by any bait, however tempting ; the same will be the case, if three enter and two come out, or if four enter and three come out, and so on till a number is reached which is beyond his arithmetic." How far are we to give credit to these current statements is very uncertain, but granting their accuracy, they do not require for their explanation a distinct recognition on the part of the crow of numbers, a conscious subtraction and the acceptance of a definite remainder. Concede these, and the sagacious bird would quickly find in the objective teaching of the rowed cornfields before him, an express provision for a grander arithmetical procedure. Within narrow limits, groups of two, or three, or four, or five objects are directly and readily distinguishable in perception aside from numera- tion ; beyond these they do not so vary the impression as to make the difference easily observable. Groups of twenty and of twenty-one persons will hardly be distinguished by a stroke of the eye. Certain separable sensations, there- fore, may be associated in the experience of the crow with danger, while others inseparable have made no such impres- sion. Let, however, one of the twenty men always remain, and doubtless the crow would soon attach danger to this 304 DYNAMICS OF THE INTELLECT. number also, and the philosophers find in the new fact proof of a growing power of calculation. The crow learns by experience to fear man, that is to connect danger with certain perceptions. In rare cases, under protracted expe- rience and varied discipline, he might carry this association two steps farther, to three definite, closely united impres- sions ; a hut, the entrance of three, the departure of two. This experience, provoking alarm in him, would extend by admonition to others, and would at once receive the inter- pretation above given. We find it very difficult not to at- tribute to such actions the same degree of thought and in- telligence which would be indicated by them in us. Yet this tendency should be easily overcome, when we remem- ber that we are compelled to cover up by the word instinct actions which in man would show the most wonderful knowledge and skill. It is certainly no very strange thing, that three perceptions should, in the ready memory of a crow, alert and watchful, by life-long instinct and habit directing its attention to like facts, find at length a fixed association with danger. It is narrated, that a raven hit upon this method of de- frauding a dog of a portion of his dinner. The raven would approach so near and so annoy ingly as to provoke pursuit. This pursuit would draw the dog from the dish, and the raven, quick of wing, would immediately rise and pounce down on the unguarded meal. Observe how easily such a series of associations would be formed, the acts constituting it finding union and undesigned repetition in experience till they became a habit apparently shaped on a rational pur- pose. Impelled by hunger, the raven would naturally ap- proach the dog as near as he dare venture ; the dog as na- turally would resent the intrusion. The raven, pressed by pursuit, and rising on the wing, would see the unprotected dish, and at once pluck a portion of the coveted food. This FACULTIES OF ANIMALS, 305 process would repeat itself a second and a third time, till, connected with the desired result, it would become direct and constant. What shall be said of the reasoning of the dog who repeatedly suffered from such a form of depreda- tion ? It matters little whether the above instances are true, others like them are true, and admit of similar explanation. The fear and caution of a dog when he has committed an offence, the cunning and skill of a fox, the pliancy of a horse, are not surprising, when we consider their quick senses, sharp appetites, retentive memories, protracted, varied, and severely enforced experience, and inherited ten- dencies. Knowledge, moreover, is communicable between animals by inheritance and by transfer. The obedience, docility and training of the horse are readily imparted to his yoke-fellow, and the fear and sagacity of a fox help to awaken like qualities in his companion. The practical value of the above conclusions is very great, in teaching us how to handle, and how to estimate, brute life ; and still more in establishing a fitting barrier between it and rational life. If this difference exists be- tween them, then is man unapproachable by the animal. He stands on another platform of being. It is not an acci- dent of physical structure, the absence of language, less fortunate or less protracted development, that divide the two ; but entirely new endowments, bringing with them a new and exalted sphere of being. Man shares conscious- ness, a perception and retention of external events, with the animal ; but not his intuitions of the invisible, nor his rational apprehension and government of action, nor his moral and spiritual endowments. BOOK II. § 1. We have now reached the second class of mental phenomena, that of the feelings. These have received less attention than the intellectual faculties. They are far more numerous and complicated, and have been more recently regarded as a distinct division. The three classes, recog- nized by Kant, have since his day been generally accepted. Knowing, feeling, and willing, are each forms of action so simple, that it is easier to perceive, than to state their dif- ferences. Indeed, expository definition of each is impossi- ble in other than synonymous terms. Each is known and only fully known by experience. There are, however, cer- tain diverse relations of these several acts, or states of mind that may be pointed out. Though the feelings were late in receiving attention as a distinct portion of our mental endowments, popular lan- guage has so far severed them from our thoughts, as to re- fer them to a separate part of our nature. It is a method of expression still somewhat unusual to common speech, to talk of the emotions of the mind ; we more frequently hear the words, the sentiments, the emotions, the feelings of the heart. A first distinction to be marked between knowing and feeling is that one proceeds under a double, the other under a single form. The thought and the object of the thought lie in the mind, distinguishable parts of one process ; while the feeling is a simple mental state. This has been ex- pressed by saying, that the processes of thought are more objective, those of feeling more subjective. This language, DISTINCTION BETWEEN FEELING AND THOUGHT. 307 however, seems not quite explicit. In one point of view, the feeling is more objective than the thought. To be sure the thought attaches itself necessarily and distinctly to an object, but that object itself becomes a subjective one, some- thing grasped and held by the mind as an object of contem- plation, so that the entire movement maintains a subjective character. On the other hand, a feeling is often occa- sioned by an action or an object external to the mind, un- der whose influence the emotion is suffered. This object, in connection with our stronger and more well-defined feelings, evokes especial consideration, is seduously sought after or avoided, and thus imparts a peculiarly objective turn or tendency to emotion. Take such passions as love or hatred, such sentiments as admiration and contempt; con- sider the appetites and the desires, how objective are they in the frame of mind and cast of action they produce. In- deed, the first condition of contemplation, a quiet, subjec- tive handling of a topic is, that the feelings be hushed, that these restless children of the household be put to sleep, and the thoughts be left to move uninterruptedly within their own circle. On account of this ambiguity of the word subjective, and the marked external tendency given by feeling to action, we prefer to speak of thought as bipartite and feeling as simple. Neither method of presentation holds equally well in all forms of the phenomena concerned. Sensation most distinctly separates itself from perception by its more definite and local action in the organ involved. In speaking of a feeling as subjective, reference must be had to the emotion itself, and not to the contemplation of the object which may call it out. A second diversity in thinking and feeling is found in their dependence on volition ; the former is more, the lat- ter less immediately the result of voluntary effort. The thoughts are more directly reached and guided by the will 308 THE FEELINGS. than are the feelings. Indeed the most of these are so occasioned by the immediate and unavoidable presence of external conditions, that it is only indirectly and with con- siderable delay that volition can reach and modify them. Our thoughts, our subjects of reflection are the primary objects of volition, while the feelings are slowly changed with a change in their physical and intellectual conditions. While the thoughts are more directly subject to the will than the emotions, the emotions more immediately influ- ence the will than do the thoughts. Here is found a third difference of relation. The state of feeling is the direct ground and occasion of choice, while our opinions govern the will only as they first govern the heart. The only opportunity of confounding knowing and feel- ing seems to arise from their common relation to conscious- ness. We express the fact that our feelings, as our own, are present to the mind by the language, I know that I feel, I know that I am angry, I know that I have sympa- thy with the suffering. We thus seem to underlay feeling with knowing as if the one were but a peculiar form of the other. The same reasoning, however, would apply to vo- lition, and the difficulty springs only from the defect of language. We express the simple and single fact of a feel- ing under the form of a double act, one branch of which is an emotion, and the other a cognition. A better analysis has enabled us to see that the expression, I know that I feel, no more implies a double act than the kindred asser- tion, I know that I know. An act of knowing is distin- guished from one of feeling or of volition in involving a disclosure of something beyond itself. The three, in in- volving consciousness, stand on common ground. Herein is the M-partite character of knowing apparent. A percep- tion that encloses no judgments is undistinguishable from a sensation. DIVISIONS. 309 § 2. The feelings may be divided by their intrinsic character, or by the objects or conditions which draw them forth. The first would seem the more just ground of dis- tinction, yet the second finds easier application, and closely allies itself to the first, since different grounds give differ- ent emotions. Our first division into physical, intellectual and spiritual feelings proceeds on the conditions or occa- sions on which they are respectively called forth. The physical feelings are located in the body, have a physical source, and pertain to the states of physical organs. The intellectual feelings arise in connection with the judgments of the mind. It is the perceived relations in which we stand to objects about us, and especially to other men, which call forth these emotions. Their ground then is an intellectual one; since if we were destitute of thought, forethought, if we could form no conclusions concerning the effect of things, their approach or their possession, the effect of the actions and character of others upon ourselves, we should be left destitute of these feelings, and only sub- ject to the immediate play of physical forces upon us. The third class of feelings is the spiritual. The word spiritual is not so definite as the other two. We employ it to designate the highest portion of our nature, that by which we have a rational and responsible life as opposed to a merely intellectual one. Now it is our intuitions, more particularly a limited portion of them, which confer these higher powers, and put us in these higher relations. The sentiments elicited by these more profound revelations, this deeper insight into the rational world, the truly spiritual world, are the spiritual feelings. More concisely, the spir- itual feelings are those immediately conditioned on the in- tuitions. Of these several classes, the first may belong in feeble form to the lowest animal life, and in full form to the 310 THE FEELINGS. highest. The second belongs chiefly to man, though in a few of the nobler animals, it finds partial presentation in connection with the tacit anticipations, the informal con- clusions of association. The dog does, through the educa- tion of a retentive memory, permanently interlock what, for want of another word, we must call conceptions, and is, therefore, ready for the feeling of joy or fear in view of anticipated results. Yet, in fullness and variety, these emo- tions do not compare in the most sagacious brute with the corresponding class of feelings in man. Indeed, much that we regard of this character in the animals below us, is but the false, the flattering interpretation which we bring from consciousness for the explanation of acts, in their external form alone, like ours. The dog licks the hand of his mas- ter, and that master conceives it, not as the act of a blind, instinctive fellowship, worth intellectually no more than the good- will of the cow that cards with her rough tongue the hide of her gratified companion, but as a distinct expression of a clearly defined attachment. The third class, from the nature of the case, belongs exclusively to man, and, in its full forms, to the cultivated, the developed man, — one who has been ripened out of physical sensations, out of the half- way ground of the simple connections of thought into the habitual and active play of his intuitive powers. The words by which we designate the emotions are, for the most part, very loose in their application. Of these the word feeling is the most general. It ranges through the three classes. The pains and pleasures of the body are feelings ; equally so are the fears and hopes of the prudent, the delights of the artist, and the satisfaction of one obe- dient to moral truth. The word emotion, is applicable to the feelings of the two higher classes, hardly to those of the lower; while the word sentiment finds at least its fullest meaning in the third class only. "We designate as -DIVISIONS. 311 sensations, physical feelings exclusively ; as passions, intel- lectual feelings exclusively, — though only a part of them reach the intensity indicated by the term ; — and as affec- tions, the higher, the moral emotions exclusively. That, however, which is especially confusing in the language of the emotions, is the different states included under one word, for instance, love. We love the food that pleases us, we love the wealth that gratifies desire, the scenery that delights the taste, the person whose character meets the ap- proval of our moral sense. We have occasion, therefore, to put feelings covered by the same word, into entirely dis- tinct classes, and to regard love as an appetite, a passion, or an affection, according to its several objects. The word now shows a decided tendency to withdraw from the lower field, and take up its abode in the higher one. CHAFTEE I. The Physical Feelings. § 1. The physical feelings are distinguished from others by arising directly from the body. They have a physical source and locality somewhere in the body, or, like nervous debility, are diffused through it. They are divisible as re- gards general quality, into pleasurable, indifferent and painful feelings. By indifferent feelings we do not mean complex states of mingled pain and pleasure, but states de- clared to consciousness, but neither as yet agreeable or dis- agreeable. The three divisions, if we look at them in ref- erence to action, may be termed the stimulative, the indic- ative and the repressive feelings. The condition of cer- tain organs indicates a preparation, or want of preparation for activity. Thus an appetite gently aroused prepares the way for indulgence. Simply as an appetitive movement, as yet neither balked nor gratified, it is hardly an occasion of pain or pleasure, but merely gives suggestion of a line of gratification. As we begin to indulge the appetite, a sensible, declared pleasure sets in, stimulating farther indulgence, and this continues till the present power of the sensibility is ex- pended. Then a second indifferent or indicative feeling succeeds, dissuading, without pain, from further indulgence. If this limit, however, be over-passed, positive discomfort follows, decidedly repressing activity. These three states may be regarded as a series of alternating cycles through which the physical feelings tend to move, and in one or other of which, when active, they remain for the time being. GENERAL SENSATIONS. 313 There is a farther connection between the three states in the fact that they arise successively in one organ or set of organs. § 2. The earliest of these physical feelings are general sensations. These may arise either from conditions which affect the body extendedly, as those which occasion lassitude and unusual vigor, or the sense of pressure, or of heat and of cold, or of numbness ; or they may indicate the condition of some one organ or set of organs, as nausea, tooth-ache, or irritation in the eye. This class of feelings it is not easy to enumerate. Some of them approach in character very closely the special senses, while others appear but rarely, and subserve a very limited purpose. There is, perhaps, no organ, or portion of the human body, which may not be- come the seat of a peculiar feeling, more especially a pain- ful feeling, indicating difficulty and demanding relief. As a class, the sensations which disclose states have more fre- quent reference to some repression or modification of action than to its excitation; and present themselves under the form of suffering instead of enjoyment. The reverse is, however, many times true. Buoyant life declares itself in physical impulses, at first obscure, but leading when fully developed to the intense pleasure of sportive action. Re- dundant power tends to explosive efforts and renders such exertion very enjoyable. Feelings which indicate states of the body or of its special organs are for the most part present only as they tend directly to affect action, and through the will to se- cure either exertion, repression, or changed conditions. The stimuli that regulate involuntary action do not usually come into consciousness. Respiration, in its safe and meas- ured movement, is secured by nerves and muscles that act and react on each other automatically, with no direct cogni- tion of the mind. Let, however, some unusual state arise ; 314 THE PHYSICAL FEELINGS. let the air be restricted, or become very impure, and dis- tinct sensations follow, provoking in extreme cases the most violent exertion. The larger portion, then, of those sensa- tions which spring from some unusual condition of our phys- ical organs, are present to indicate a line of action ; at least to compel inquiry, and set the reflective powers to the work of guidance and correction. Thus are the nature and limits of the physical, physiological laws under which we live de- clared to us ; the times of activity and repose, the forms and bounds of indulgence, and the necessity of remedial meas- ures. As most diseases find their true remedy in some form of rest or of restraint, we see that the pains which indicate them are not only directly repressive of effort, but indirectly also through the increased advantage which arises from an appetite denied, from labor laid aside. On the other hand, the power to do begets corresponding effort, and is rewarded with a pleasure which in turn stimulates the body through the mind, and tends to make the exertion nutritive of the faculties to which it belongs. We cannot go to the extent of the view presented by Bain, which makes pleasure and pain automatic, the one stimulating, the other arresting action, much like the opening and closing dampers of a steam-engine. Such direct effects they frequently have, but more often they incite or correct action through the inter- vention of thought and volition. Indeed pain may momen- tarily quicken action, and pleasure may ultimately exhaust the strength, and so slacken effort. The sensations stand in too living, too complex a relation to our vital, intellectual and voluntary powers to submit easily all their relations to a single statement. Pleasure and pain alike exhaust power, but the one with, the other without compensation. An half-hour of intense suffering takes away not simply the strength, — play would have done this in part — but leaves the nutritive powers depressed. The exertion of enjoyment, on SPECIAL SENSATIONS. 315 the other hand, while expending the present store of power, re-acts favorably on the vital forces. Intense pleasure at its consummation trembles on the verge of pain, and in- tense pain, when not utterly exhaustive, passes back at its expiration into intense pleasure, occasioned partly by con- trast, and partly by the flowing in again of vital power to its normal channels. Spencer's assertion that "Every pleasure increases vital- ity, and every pain decreases vitality " * is also too sweep- ing. Both pains and pleasures may tax vitality ; both may be remedial ; and both, may be unfortunate. This relation to vitality does not explain the peculiar character either of pains or pleasures, nor exhaust their intellectual offices. § 3. The second source of distinct physical feelings are the special senses, the organs of sensation. The chief of these, at once recognizable, are touch, taste and smell. Sen- sations and perceptions should be distinguished, and these classed with cognitions, and those with feelings. Percep- tions have with some clearness a bi-partite character ; the object and the action directed towards it at once appear. The seeing, and the object seen, are necessary complements to each other; whereas by taste and smell we only indi- rectly and inferentially reach the source, given under an- other sense, that of sight. Evidently in sensation, we are engaged with the feeling ; in perception, with the source of the impression. Perceptions also differ from sensations in having so little of a declared local character, that, though, physical in their sources, they no more reveal their physical connections than does pure thought. Sensations, on the contrary, disclose themselves as a certain peculiar state of a given organ, and are therefore to be ranked as feelings. Of all the senses, touch occupies the most intermediate ground ; while its phenomena ordinarily present the phases of: feeL * Data of Ethics, p. 87. 316 THE PHYSICAL FEELINGS. ing, it may in the absence of the higher senses of sight, of hearing, become so far intellectual as scarcely to direct at- tention to the sensation. It thus becomes the unobserved medium of knowledge, the matter revealed being the only object obviously before the mind. Any sensation may be the occasion of a judgment, bearing the mind outward to a particular object ; the peculiarity of touch is, that often by habitual use for this end, the sensational element is lost sight of, sinks from observation, and the perceptive ele- ment rises in its place, making this ordinarily over-shad- owed sense a not inefficient substitute for sight. These special senses, all of them, stand closely connected with the intellect, and have thus been more frequently united with the organs of perception, and fallen into the first class of mental powers. The distinction now made seems, however, fore-shadowed in the physical fact, that the senses of sight and hearing are so immediately con- nected with the cerebrum, the seat and instrument of thought, that a removal of this destroys them, though leav- ing the other senses unimpared. * Touch, taste and smell, however, while primarily feelings, are used constantly as means of discrimination and guides to action. They very frequently draw after them conclusions, set in motion the judgment, and thus return on the will through the media- tion of the mind. This is the ordinary action of a pure, well-defined, special sensation. Taste may be so pungent or so nauseating as to produce a direct, involuntary action of ejection ; but odors and flavors are usually, in their ef- fects on action, simply grounds of discrimination by which we are guided in accepting or rejecting the object before us, in assigning it a definite position among the things used by us. Our sensations thus start from the central, the per- ceptive, the indicative point, and then become either stimu- lative or repressive, according to their nature. APPETITES. 317 Sensations are also three-fold in their relations to enjoy- ments. From the midway ground of indifference, they pass into pain and pleasure. Their double office is here again very obvious. They are means of independent gratifica- tion as well as of guidance. They are sources of abundant physical pleasure, and find a primary purpose in this their direct character as feelings. In this connection, they act more immediately on the executive powers, stimulating the effort necessary for their gratification, and checking any movement that gives rise to pain. Sensations then, are in a double sense stimulative by their direct character as feelings, by their indicative character, revealing to the intel- lect the nature of the objects about it. It is, however, in the first aspect alone, that they can be divided, as feelings, into the three classes, stimulative, indicative and repressive. Those sensations are chiefly indicative which, in reference to pleasure or pain, are indifferent. Things inimical, de- termined chiefly by the eye and ear, are recognized in part by touch, and sometimes by taste and odor. This discrim- inative use of the senses is an acquired one, and lies apart from the purpose which they subserve with all as avenues of enjoyment. Thus their perceptive and sensitive uses show a tendency to separation and mutual exclusion. § 4. A third distinct class of sensations are the appe- tites. These are closely united to those indicative feelings which declare the condition of an organ. They differ from these only in being more special, returning with regularity, and performing a constant and fixed service in the animal economy. The appetites are specialized and regularly re- turning physical feelings demanding a specific act of gratifi- cation. Both in the special senses and appetites, there are a definiteness and constancy of purpose, not found in general action, as well as a source of ever returning pleasure, almost independent of effort. While the senses are specialized to 318 THE PHYSICAL FEELINGS. indicate external relations, the appetites are specialized to disclose internal states. Indeed, the appetite for food, as a means of enjoyment, so closely unites itself with taste and odor, as to yield with them a compound gratification inca- pable of practical analysis. The return in most of the ap- petites is at measured intervals; in others the spaces are more irregular. According to this definition, the desire for sleep is an appetite. Hunger and thirst are impulses recur- ring more fixedly ; sexual appetite, one that is renewed less certainly. An appetite in its first action, as yet neither gratified nor denied, is indicative ; and indifferent as regards pleas- ure and pain. It is, indeed, the condition of the pleasure which is to arise from indulgence, but is itself hardly either a distinct enjoyment or a declared annoyance. One or other of these, however, it quickly becomes, according as its intimations are accepted or withstood. Different appetites may be suppressed and modified with very different degrees of success, according to the purpose they subserve in our physical constitution. One is as im- perative as the wants it indicates ; another is, in the posi- tion it holds, very much the product of intellectual and moral forces. The appetites are physical indications and guides of action, and, in their healthy indulgence, uniformly give pleasure ; in their denial, or excessive indulgence, as uniformly inflict pain. The pleasures and pains which accompany them are, carefully watched and collated, safe guides of action. They are, nevertheless, far from being sufficient, automatic forces, securing the results of physical well-being. While they are at first direct stimulants and immediate restraints, they are chiefly, in the human consti- tution, operative through a wise election and pursuit of pleasure, a sagacious avoidance of evil. The brute and the rational constitution seem to show an important distinction OFFICES OF SENSATIONS. 319 at this point ; the one is wholly automatic in the restraint and control of appetite ; the other leaves the checks chiefly to reason. The purposes served by our sensations are various, fre- quently co-existent, and always concurrent. Of this, the special senses, the appetites and the feelings which accom- pany the active powers, are examples. A large circle of enjoyments are through them added to our physical organ- ism, and a pleasurable life provided for. Immediately con- nected with this is a second purpose. A direct, physical stimulus is, through these feelings, administered to that nu- tritive and muscular action on which the well-being of the body depends. Pain abates, pleasure promotes effort. The one exhausts, the other stimulates, and, within certain limits helps to renew the strength by which it is fed. A third purpose of our sensations is found in the knowl- edge, otherwise unattainable, which they impart of the states of the body, the conditions and demands of its several or- gans. They thus become the basis of that reasoning by which we adjust action, food and remedial agents to our real wants; make an intelligent provision, and lay down wise precepts, for our immediate and future well-being. A fourth and somewhat more remote ministration of our sen- sations is to general knowledge. Through them, we come in contact in a new way with surrounding objects, take cog- nizance of a different set of qualities, and thus make more complete and perfect our classifications. There is a tend- ency, in thus making our sensations means of intellectual discrimination, somewhat to abate their force and character as feelings. Of this, we have sufficiently spoken. The relation of the physical feelings to health and activ- ity is easily seen. Unimpeded activity is pleasurable, but the seat, the source of pleasure, is found in an original conformation of the physical man ; as much so, we appre- 320 THE PHYSICAL FEELINGS. hend, as the enjoyment of a fragrant rose in the peculiar power of the special sense of smell. "We are not to sup- pose that we have explained either pleasure or pain by re- ferring them respectively to unrestrained, and to impeded activity. We are able to give some of the conditions, and some of the consequences of physical sensations, but their immediate causes in the organs themselves, and in the mind, we cannot give. The last arid exhaustive analysis we can not make. A feeling as a feeling is ultimately and, shall we not say, sufficiently known in itself. Before passing to the intellectual feelings, we mark some border facts which prepare the way for the transition. What are termed natural affections, are examples of transition facts. We suppose these words strictly employed to designate feelings aroused by physical facts, physical ties ; not intellectually considered, but sensationally exper- ienced. There seems to be a small remainder of such af- fections in man, but they are so lost in the higher feelings, stirred by the same facts intellectually considered, that it is difficult to separate them. The animal is, for a time, pas- sionately attached to its young. These affections seem to follow in a direct, physical way from the sensations pres- ent. The helplessness of the young apparently forms no ground of the emotion. The young of another animal may become the object of immediate and bitter attack. The substitution of another offspring for its own is successful only when the perceptive instincts of the parent are baffled and misled. Something of this direct attachment seems to appear in the human parent, though it is so overlain and modified by feelings of a purely intellectual character as to play no very important part in our constitution. Doubtless the tenderness of the mother does owe something of its quick, yearning, responsive action under the claims of the infant t the purely physical conditions of the relationship. TRANSITIONAL FACTS. 321 Irritability, which is often a physical state, and may al- ways be more or less due to physical conditions, neverthe- less does much to determine the character of the conceptions present to the mind. There are inseparably mixed with their intellectual causes, immediate physical conditions, which often make them in degree, if not in kind, what they would not otherwise be. The force of the emotions, and the fact of their presence are often determined by physical states. CHAPTEE II. The Intellectual Feelings. § 1. The intellectual are distinguished from the physi- cal feelings by the fact of their dependence on objects and relations presented to the mind, and thus, in a secondary way, influencing the emotions. The sharp thrust of a weapon brings instantaneous pain ; the abuse of an enemy arouses anger only as it is understood and mentally con- templated. These feelings may also be divided, as regards emotional character, into pleasurable, indifferent and pain- ful; and as regards their relation to action, into stimula- tive, indicative and repressive. These two divisions — and this is especially true of the second — do not so much ex- press intrinsic characteristics as passing relations. Essen- tially the same feeling which in one relation or in one de- gree is pleasant, may in another relation or in another degree become painful. So also the same feeling, as fear, may at one time quicken and at another restrain action. The intellectual feelings are divisible into primary and secondary feelings. The secondary feelings arise from the relations which things assume in consequence of the primary feelings ; while the primary feelings rest, at one less remove, on constitutional endowments. The primary feelings are of two orders, according as the relations which call them out are simply intellectual, or are also those of interest. The root of the first order is in the passing phases of the intel- lect, of the second order is in the intellect only as it ministers to deeper constitutional impulses. The feelings of the first order are those called out in connection with habit, by PRIMAB Y FEELINGS. 323 separation and reunion, by things or events new or unex- pected, by wit and by humor; and those* of the second order are desires. There are obscure feelings of comfort which attend on the habitual in action, and of discomfort which arise from the interruption of habits. There is a large mass of less obscure feelings of satisfaction and discomfort which are connected with the presence or absence of friends. Wonder is a more purely intellectual feeling, and is awak- ened by things new in our experience. Surprise is a yet stronger feeling elicited by that which contradicts our ex- pectations. Both indicate the wakefulness of the intellect to a change of data. Wit is the joining of ideas apt to our purpose by an unexpected relation ; humor is the join- ing of things or images apt to our purpose by an unex- pected relation. Things are in this definition opposed to ideas in the former definition, but include persons. The unexpected relation of the first definition refers to some secondary or remote connection as opposed to the philo- sophical connections of thought ; in the second defin- ition, it indicates an unaccustomed, and incongruous link. Though the feelings that accompany wit and humor are of an intellectual order, they may be easily united, as in ridi- cule, to personal feelings. All of this order of feelings turn on the familiar and unfamiliar. In habit both mind and body are involved ; in associations, the emotions and the thoughts ; while in wonder, surprise, wit and humor, the thoughts are deeply concerned. The second more intense order is composed of the de- sires. These may be termed the appetites of the mind, as they express its appetences, its longings, its objects of pur- suit. They have been usually spoken of as directly native feelings. Herein there seems to be some confusion of ideas. If they were direct, unreasoning impulses, they could not fall into the second general class of feelings, to wit : those 324: THE INTELLECTUAL FEELINGS. which have an intellectual basis. That they are not spon- taneous, immediate impulses, a little thought will be suf ficient to show. As universally stated they are directed toward abstract ideas, not toward concrete objects; they are desires of wealth, of power, of knowledge, not for wam- pum, for the ability to bend a bow, or to calculate an eclipse. Now a desire directed in the outset to a generalization, to an abstract quality, is an absurdity, since no such quality can be present to the mind except as the result of much comparison and many judgments. Neither should we avoid the difficulty by saying, that these desires fasten themselves with native, original force on specific objects under each of the categories of desire. There are no specific objects which draw forth universal desire, and which can stand as concrete types, or representations of the notions of power, wealth, honor. Specific powers become points of interest and desire according as they are able to gratify certain native appetites or tastes. Possession is a matter of inter- est to the child only as the thing claimed stands in some relation to its sports, by which it is capable of promoting its enjoyment. Possession, without some connection with our pleasures, has no significance, either in early or later life. A square mile of territory on the frozen continent of the Antarctic Zone, has no power to awaken desire in any man. Now this discerning of the relation of things to our appetites, our ac- tive powers, our tastes, which makes them valuable, is an intellectual activity, receiving constant expansion as we grow older, and leading us to attach importance to the ownership of an increasing variety of things. The igno- rant man cares not for a book, except as he can sell it ; be- cause the mental conditions which make possession impor- tant to him have not been met. Our desires, then, are secondary feelings uniformly DESIRES. S2; evoked by the perceived relations of objects to our primary native feelings ; our appetites below, and our tastes above. Without either the lower region of animal tendencies, or the higher region of spiritual impulses, desires would not exist ; because those objects now included under the term wealth, or those possessions known as knowledge, would have no value, having no power to minister to our pleasure. The statement has been made only on the positive side ; of course we include the corresponding negative considerations. Objects may excite desire, because they enable us to escape pain. *An action, however, which stands in no relation to either pain or pleas nre, must be one to which we are wholly indifferent. We have, then, no occasion to suppose, indeed no intel- ligible grounds for supposing, the presence of native de- sires in our constitution for certain abstract qualities, or for abstract qualities under a concrete form ; because, first, the relation of wealth, power, knowledge, to our happiness is a sufficient explanation of our desires for them ; because, second, these desires come and go with this relation — the miser even not being able to prize that which can not, un- der any conditions, be sold ; and third, because there is a difficulty in supposing generalizations, arrived at by much reflection and constantly expanding, the direct object of a simple, primitive feeling. The very notion and definition of a primitive feeling is rather the immediate action of some object or intuition on the emotional constitution. The secondary relations to our well-being, which things disclose through the intellect, are grounds of our intellectual feelings. These are all in one sense secondary, though they are so in two degrees, and may be subdivided among themselves as primary and sec- ondary. In classifying the desires, we are then classifying the ob- 326 THE INTELLECTUAL FEELINGS. jects which draw them forth. Desire is an emotion essen- tially of the same character, whatever that be to which it attaches. The mind does not remain indifferent to those things and states which it sees to concern its enjoyments. This fact is expressed in a feeling toward them which we term desire. A desire is the inclination of the mind toward things which it sees to be the direct or indirect sources of pleasure. It rests back as a secondary feeling on those primary sensibilities to which the world directly ministers. Now the variety of objects which gratify man, and the va- riety of their separate ministrations are so great, that it is not easy to give an exhaustive classification of them. Those general words which divide, yet include the most of the things pursued by men, are wealth, power, honor ; truth, beauty, and virtue. It is easy to fall into confusion in speaking of the de- sires. The difficulty arises chiefly from not keeping the words by which we express them at the same grade of gen- eralization. A great variety of means, near and remote, general and special, separates our actions from the ultimate gratifications at which they aim. These gratifications, lying in the outer circle and due to constitutional appetites, pas- sions, tastes, are the grounds of the desires, which attach to any or all of these intervening means. It is these means that are grouped under very abstract general expressions, and classified as the desires. If the classification is to be of any worth, it must take place at one grade of generaliza- tion, must lie in and cover one circle, to the exclusion on the one side of still more general terms, and on the other of more specific ones. Thus wealth, power, honor, — by honor are meant positions and circumstances of honor — beauty, truth, virtue unite to cover the two halves of one circle of the generalized objects of human effort. Knowl- edge is not added, because it is divided between power and DESIRES. 6XL truth. As a simple means to an end it is a form of power, as in itself productive of pleasures it is truth. Having so classified the desires, we must guard against a tendency to recognize as distinct desires any of the more remote objects to which wealth, power, honor, beauty, truth, virtue are the means. It is accurate, if not fitting language to say, I desire revenge. The heart also yearns for objects of affection, and that it itself should be made an object of love. When suffering pain, we desire its removal ; when fearing punishment, we desire escape. These are not new desires under our list, but a few of the many grati- fications to be reached by wealth, power, honor. The desire of happiness is sometimes added to the list. The objection to this is, that this desire is a still broader generalization, including all the others. This desire em- braces all our desires, is the utmost stretch of analysis and abstraction. Admit this, and there is no opportunity for further division — all impulses are grouped under one gen- eral impulse common to each. The desire for existence is also a desire dependent for its force on those other desires which make life pleasurable. It, too, is a common condi- tion of them all. We regard desire, as a feeling, as indifferent ; neither pleasurable nor painful, at least in its earlier forms. When nourished into full strength, it may assume a more positive character. A desire for wealth that is, as yet, neither grati- fied nor balked, while it becomes an immediate ground of pleasurable " activity, while it gives direction and concert to the feelings, can hardly of itself be called distinctly painful or pleasurable. This is seen in the ease with which desire passes into pain or pleasure with any increase or decrease of the obstacles to its gratification. In the ordinary, fa- miliar balance of effort and reward, desire guides rather than vexes or excites us. When it produces pleasure, it is 328 THE INTELLECTUAL FEELINGS. rather by the activity it inspires, the hopes it enkindles, than by its own natnre as an impulse : when it provokes suffering, it does so by the unusual obstacles it encounters, by the disappointment of fruitless effort. A pure desire seems to be as simply indicative as any feeling can well be, to make way for the current of emotions that is sure to rush along in its channel. The desires have different degrees of strength according to the minds in which they arise, and the objects toward which they are directed. The desire for wealth passes with a few into a passion, and becomes the most exacting of ircu pulses, while, with others, it is so gentle an incentive as to control but few of their actions. Herein, again, is seen its secondary character. The mind that habitually forecasts the future, that brings coming enjoyments into clear con- trast with immediate pleasures, is one in which the desires show their full strength. The conditions of their activity are fully met, and they soon come to rule with undisputed sway. One, however, in whom the primary appetites are exacting, and the reflective powers feeble, renders but way- ward and intermittent obedience to the desires, and leaves the events of life to be fashioned by the objects in most immediate connection with the sensibilities. The strength of desires also depends on the nature of the objects sought — a farther result of their secondary character. The pursuit of wealth, of power, of honor, may, in rare instances, settle down into an exorbitant passion in minds in which the lower circle of vigorous, primitive sen- sibilities is united with moderate reflective faculties, fur- nishing a clear, yet nevertheless limited horizon of effort. In many cases these desires are relaxed by the disappoint- ments which attend upon them, or the unsatisfactory na- ture of the results when realized. The desire for wealth is likely, under the force of habit, under the momentum of SECOND AR T FEELINGS. 329 the mind, either to pass into the blind passion of avarice, or to suffer abatement from the limited character of the good wealth can confer. The desires for truth, for vir- tue, on the other hand, grow under success with a nor- mal, rational growth. Each acquisition is a stimulus to further acquisition, and the satisfaction of possession in- creases every moment with possession. The mind more and more justifies its choice to itself, and congratulates itself on that which it has accomplished. The desire for wealth is like a stream that at length finds a precipice so high that in its leap it is lost in air, dissolved again in mist, and never resumes a peaceful flow ; while the love of truth and virtue, more tranquil currents, swell in volume, and roll on increasing waters to the ocean. § 2. On either hand, the desires give rise to a large class of feelings dependent upon them. We will speak first of those pleasurable ones which accompany success, and thus stimulate effort. They fall into four classes, those which arise from success as being achieved / those which arise from success as achieved by ourselves ; those which arise from success as achieved by the aid of others ; and those which arise from success as achieved by others. Im- mediately consequent on a state of desire, are the feelings of hope and joy in view of the prospect of obtaining the object sought. Indeed, hope is resolved in analysis into the feeling, desire, and the intellectual condition, expecta- tion. We would rather regard these as the occasion of the emotion than the very emotion, hope. Joy accompanies success, and passes through various stages lying between tranquil satisfaction and triumphant exaltation. These feelings spring immediately from a free flow of the activi- ties called forth by a successful desire, and, in turn, greatly quicken their action. The emotional state thus becomes instantly complex, consisting of the immediate effect of an- 330 THE INTELLECTUAL FEELINGS. ticipated pleasures, and the realized pleasure of fully em- ployed powers. This has always been regarded as one of the most unalloyed forms of enjoyment — that evoked by the grasp of coming good by the mind as a certainty, to- gether with the high exercise of its own faculties in secur- ing it. The stimulated powers and feelings not only yield the delight of successful action, but the imagination makes the most of the pleasure promised, and overlooks the vexa- tions and disappointments which too frequently embitter the actual enjoyment of it. This concurrence of the prac- tical and imaginative faculties leads to an exalted state of feeling, especially when neither experience has sobered, nor age made" sluggish, the emotions. The second class of pleasurable feelings comes from the connection of effort and success with our own action. They are pride, vanity. These emotions are most influential over action, and constitute a large part of its reward. Yanity, the pleasure which the mind receives from the admiration, the favorable notice of others, exists with various conditions and under very different degrees of intensity. In its mod- erate forms, it is a quiet incentive, and only becomes ill- grounded and foolish when it leads to a neglect of real ex- cellence and solid attainments, in favor of popular powers and showy acquisitions. Within its legitimate sphere, it closely unites itself with that desire for the good opinion of others which the good man may well cherish. There are few feelings which sustain the inferior desires as those for wealth and position, as constantly and effectively as this of vanity. Wealth owes its attractions, with many, to its abil- ity to captivate and dazzle the public eye, to open gaping mouths, and bewilder feeble wits. Pride arises from the same good opinion of one's self and one's possessions that characterizes vanity. It is, how- ever, accompanied with more independence of character, and SECOND A B T FEELINGS. 331 does not, therefore, find its gratification so much in the ad- miration of others as in its own admiration. Vanity loves parade, delights in the flow of popular sentiment, floats its gay shallop on the good opinion of others, and is stranded when public favor, like a shallow stream, is lost on some sand-bar. Pride, in its high opinion of itself, despises others, receives indifferently or contemptuously their ad- miration, and, like an ocean vessel, rides solitary on the heaving tide of its own conceit. Like vanity, it has a legit- imate form. As just self-esteem, it furnishes strength and independence to character. It accompanies all grades of desire. The food which the accomplishment of our desires affords to our own good opinion of ourselves, and our love of the admiration of others, are the most constant and certain, most secret and sweet, of the pleasures of success. In a modified form, these feelings enter into our highest moral sentiments. The various words by which we desig- nate these feelings, derive their meanings in part from the different degrees of the same emotions, and in part from the supposed justice, or fitness, with which the feeling is entertained. Conceit, self-conceit, assumption, self-compla- cence, indicate a vanity or pride in advance of the grounds for it in our power or possessions. Indeed, the words vanity and pride are also more commonly used to mark these ex- cesses of feeling, than its restrained and praiseworthy forms. Self-confidence, self-respect, personal pride designate the more measured and well-founded phases of feeling. The third class of pleasurable feelings arises in view of the relation of others to our success. We are grateful to those who have aided us. We are sympathetically attached to those who share our triumphs, who enjoy our pleasures with us. Our feelings are made deeper, hence more pleasur- able, by the impulse of kindred feelings on them. Emotional states, like electric conditions, intensify each other, and a 332 THE INTELLECTUAL FEELINGS. movement once established tends to complete itself in part by the reflex influence of one mind on another. The love and complacency begotten by success are as manifest as the impatience and vexation that spring from failure. The moment of achievement is usually seized upon as propi- tious to those who seek either forgiveness or favor. The degree of this satisfaction in others depends on the intimacy of their relations to our success, but extends itself often in a feeble form to indifferent parties. It is expressed under various words according to its character and degree, as grati- tude, good-will, attachment. The fourth class of feelings incident to success are those called out by the achievements of others. They are ad- miration, emulation, honor. The highest, the chief object of admiration is character; though simple power, physical or intellectual, may draw forth the feeling. This emotion inclines to the class of pleasurable feelings, and this, we think, in proportion as it opens a line of emulative action. Wonderful powers shown in fields of effort entirely foreign to our own labors by no means bestow, in the admiration they elicit, the same pleasure as do like triumphs in the familiar paths of our daily exertions. According, then, as admiration carries us from the midway point of indiffer- ence into emulation, does the pleasure become declared and intense even. Let the feeling, by a contrast with our • own weakness, discourage us, and it is painful rather than pleasant. We enjoy honors also the more fully which we can in some measure share; those which have their grounds in our own experience, and which we aid in con- ferring. § 3. * The feelings which accompany the failure of desire • correspond to the opposite classes, but are more intense and more varied. Those which follow directly from the pros- ipect of failure, or from failure itself as occurring, are fear, SECONDARY FEELINGS. 333 discouragement, disappointment, despair, all tending to re- press effort, and to make the effort that is pnt forth pecu- liarly exhaustive. That the activity of the mind is an in- dependent source of strength, is necessary to the highest, most successful development of purely physical strength, is indicated by the very different physical results which ac- company efforts alike in intensity, but unlike in the satis- faction which accompanies them. Hope gives strength, discouragement at once takes it away. As physical life is an independent stimulus to the mind, so mental life is an independent stimulus to the body. The second class of unpleasant feelings, arising from the relation to our own actions of baffled desire, is limited. They are certain forms of shame and humility. We are humbled by failure ; we are ashamed of the ill success which has followed our efforts ; we are mortified that others should be spectators of our weakness. These emotions are disagreeable, and may become excessive, permanently weak- ening the incentives to effort. Humility and shame find their fullest play in the moral field. Like some of the other intellectual feelings, they are adumbrations of emo- tions called forth by moral relations. The third class of painful feelings, those excited to- ward others by opposition and failure, is especially full and varied. Envy, jealousy, dislike, antipathy, resentment, anger, hatred, malice, rage, revenge, are some of the words which express varied phases and stages of feeling exas- perated by the indirect or direct interference of others, by an opposition of attitude, or character, or effort. Envy and jealousy arise from the designed or undesigned displacing of ourselves in position, or in affection, by others. They do not necessarily imply any fault on the part of their ob- ject, but merely an entrance upon ground we had coveted for ourselves. When this entrance is an intrusiou, the feel- 334 THE INTELLECTUAL FEELINGS. ings are proportionately more bitter. Antipathy and dislike express the results of a sense of opposition in character which prepares us for opposition in action, and provokes in a milder form, by anticipation, feelings of repulsion. Resentment, anger, hatred, malice, rage, revenge, mark the more violent outbursts of feeling toward those who di- rectly thwart our efforts,, who stand astride the path of our desires. These feelings, in their extreme form, so blind the mind as to make it almost indiscriminate in its action ; as to lead it to give vent to the pent-up passion on the first ob- ject that offers. The mind, like an electric battery, charged to the full by the irritation and friction of chafing events, is ready to launch a bolt at the nearest point, to blast and splinter in mere wantonness of wrath. It may be doubted, perhaps, whether these feelings of resentment are not in part pleasurable. As simple emo- tions we think not. They give rise, however, to secondary desires, desires of retaliation and revenge, and in the grati- fication of these we experience pleasure. Language recog- nizes this in such an expression as, The sweetness of re- venge. These feelings may also be blended with moral sentiments of indignation, and thus their true character be somewhat disguised. Some have regarded it as a reflection on our constitu- tion, that we should be capable of malevolent feelings. This perhaps it might be, if they were necessary, primary emotions ; if, like the appetites, they found direct, inevitable expression. As secondary feelings, however, they depend for their character on the character of the person who enter- tains them. They arise under the general possibility of transgression, of wrong desires wrongly pursued, and thus are involved in the general problem of sin, and admit of the same remedy that transgression itself suffers. Right de- sires, in their method and measure right, may be attended SECOND AR T FEELINGS. 335 only with right feelings. The holy will may ultimately reach to the correction of these products of the violent, the unsubmissive, the selfish will. The malevolent feelings are simply the evil outflow of an evil purpose. The fourth class of feelings find expression toward others in their failures. They are those of contempt, pity compassion. Contempt is the direct product of a selfish spirit toward weakness. Pity involves sympathy, and leads to consideration in the mishaps of others. It is a feeling that independent spirits accept with reluctance, as it so often bears with it something of the taint of contempt. In compassion the mind goes freely forth and shares with others the burdens of failure. A contemptuous frame of mind, as a personal characteristic, reduces all the incentives to generous action. It is a painful emotion, except so far as, inflaming self-conceit, it finds in the failure of others the food of pride. A low, disparaging estimate of the pow- ers of *men, giving birth to contempt spiced with misan- thropy, will, unless relieved by a marked exception in our own fa^or, depress action and enjoyment. Each newly dis- covered case of weakness increases the bitterness of the heart. This feeling slowly over-clouds the sky, and leaves the soul in a chill, benumbing, disheartening atmosphere, rendering it incapable of pleasure, and indisposing it to the effort by which the spell might be cast off. The contemp- tuous man takes home as guests, sarcasm, satire, unbelief, aversion. He abides in their companionship, lies down and rises with them, and suffers their corrosive breath to tar- nish the brightness of every object. Contempt is the rust of the soul, which eats it up with increasing pain. Nothing can be intrinsically more diverse, or more diverse in their effects, than that intellectual contempt which feeds on the weakness of men, and that moral sentiment which scorns a mean action. The one is the recoil of the soul upward : 336 THE INTELLECTUAL FEELINGS. the other, its gravitation downward, its cynical unbelief in goodness, its despair of strength. The most beneficent of the intellectual feelings, as good- will and compassion, are but feeble sentiments when dis v joined from the moral nature. They are still pleasurable, still indices of action, impulses to a little desultory effort, but rarely have a deeper foundation than that of sympathy, which feebly transfers to us another's feelings ; and play but a secondary part among those towering and dominant passions which drink up the life of the soul. They are re- mote reflections, faint types of those strong affections, those profound sympathies which give to the higher, the moral nature its compass and power, which enable it successfully to confront the appetites and passions, outweighing the good they offer with a greater good. It is the feelings now indicated in this second great class resting primarily on self-interest, and especially liable to ex- cess, that are termed passions. These emotions are fre- quently so strong that we suffer from them, that we seem to be their passive, afflicted subjects rather than their re- sponsible sources. CHAPTER III. The Spiritual Feelings. § 1. The spiritual feelings are so called because they be- long peculiarly to our higher nature. Intellectual action is spiritual action : yet that which gives guidance and gov- ernment to our interior, hidden life, is found in our intu- itions. The intellect is instrumental under these, and, as in the brute, is simply a means to safety and gratification. Our spiritual feelings spring up, then, in direct connection with our intuitions ; those mental elements which make our life truly rational, which give to us a choice of ends, and liberty in the pursuit of them. The only intuitions which draw forth directly feeling, are those of truth, beauty and right. There is in the emotions connected with these reg- ulative ideas the action of the intellect, yet an action differ- ent from that presented by the last class of feelings. In these, it was the observed relation of things to our enjoy- ments which was the ground of desire, with the attendant sensibilities. The mental action intervened between the remote appetites, tastes, passions, and pointed out the means of gratification, and called forth a variety of emotions in prosecuting the labor presented. In the present case, the intellectual action precedes the intuition. Patient inquiry reveals the grounds of belief, the truth : a careful discrimin- ation of qualities, of the symbols of expression, of complex relations, discloses the conditions of beauty : a thorough in- quiry into the nature and results of action, its reflex and progressive effects, lays open its true character, and then 338 THE SPIRITUAL FEELINGS. the intuitive faculty comes in to complete and seal the work in the discernment of a new and distinct quality — that of right. The proposition is said to be true, the statue is seen to be beautiful, the action is pronounced right, and forth- with there arise many sentiments which find their spring in these ideas. These are the spiritual feelings. Their immediate dependence is on the mind's intuitive action ; their secondary dependence, on our intellectual faculties. Our intellectual feelings, on the other hand, find their im- mediate source in mental action, in the conclusions of ex- perience, and their ultimate ground in the appetites and tastes. These feelings again are open to the same division into pleasurable, indifferent and painful emotions. This relation of the feelings to happiness must necessarily be a funda- mental distinction of all the emotions. Their relation to action may be said to be secondary to their relation to en- joyment, since action itself is undertaken or withheld in view of its immediate or ultimate effects on the sensibilities. The feelings can only be classified by their external rela- tions, since, intrinsically, they are all diverse, all simple original states, known in experience only. Of the external relations of the feelings, this relation to happiness is most essential, while that to action comes next in order, both as indicating an immediate purpose served by our sensibilities and their secondary effects on our character and well-being. In their connection with action, the spiritual feelings as- sume a more imperative character than either of the other two classes. In those, feelings enter to stimulate and gratify effort, or check and discourage it ; here, they go before it as well to command as to forbid action. They cease merely to allure, and seek decisively to enjoin and prohibit differ- ent lines of conduct. The middle ground of indication seems narrowed to a point, and to be pressed closely on TRUTH. 339 either hand by dissuasives and persuasives. The spiritual sentiments may be divided into those of persuasion, indica- tion and dissuasion. Their voice is always one of authority, though its authority need not be felt so long as it is kindly and cheerfully accepted. Actual or contemplated resistance provokes a class of penal sensibilities ; and obedience elicits feelings that have the positive character of approval and reward. The weakest of these sentiments, and those therefore which least well represent the class, are the somewhat in- tangible, rare, and uncertain sensibilities which accompany the discovery, the recognition of the truth as truth. Truth is the agreement of a proposition with the facts which it states. Much the majority of truths are received as truths with no emotion. Most of them are matters of interest only as they effect action — only in their relation to our de- sires and tastes, indicating success or failure, or revealing the line of conduct to be pursued. Truths, for the most part, are means possessed of no inherent, emotional force beyond their relation to ends. This negative character of truth seems sometimes to dis- appear, and truth as truth to inspire a certain enthusiasm of mind, by which we feel that this is indeed the food of our spiritual nature. We may breathe the air ordinarily with- out thought, or sensible pleasure. Occasionally, we find it peculiarly invigorating ; we inhale great draughts, and bring our whole physical being into a more conscious and exalted state. Thus is it with the truth, — the daily breath of our intellectual life. We ordinarily overlook it ; at rare inter- vals we, in deep inspiration, feel its pervasive and subtile power, and rejoice in its possession. We travel along the valley, scarcely observing the objects about us, with no ela- tion of feeling ; we pass some crowning summit, take in a wider range, and the before concealed wave of emotion 340 THE SPIRITUAL FEELINGS. becomes sensible to us ; we are lifted on its passing billow as if a breath from another world had stolen suddenly across our path. This is the kind of emotion to which we draw attention — the enthusiasm sometimes felt in truth, more especially in those fundamental, far-reaching truths which seem to suddenly lift the veil of phenomena, of varied colors, and to disclose to us the frame-work of the universe ; the pur- poses which are running through it, and bearing it to its goal. This on-going of a divine plan, when recognized, startles and inspires the mind, lifts truth out of its daily, dry, instrumental ministrations, and gives us the sense of a new inheritance and possession in a universe whose concep- tion we can thus lay hold of, whose secrets we can thus penetrate, whose wisdom and love we can thus interpret and feel. I care not how little or how much of this sentiment we may have felt, how far it may be thought to be confined to the more poetic and penetrative temperaments ; it is suf- ficient to draw attention to it as an enthusiasm for truth oc- casionally felt and avowed, finding expression in the collec- tive use of the word truth, the truth, the truths, as if a cer- tain concealed link and deep unity were to be found in all facts. We do this, too, in the face of those detestible facts, truths, which sin is forcing constantly upon our notice, as if after all, there were some profound fellowship, some one exaltation in all truths, rendering them the truth. This sensibility to the truth, be it more or less clear, be it more or less deep (1) inspires pursuit, (2)^leads to faith in a profound, unfolding plan, and (3) quickens the mind to discover the corrective laws, the compensatory statements for the defects and transgressions which lie on the surface of the world. This sentiment opens up a line of effort, in- spires enthusiasm, sends faith in advance of reason, and rejoices in the slow displacement of accredited by appre- TRUTH. 341 hended facts, of statement by disclosure, of trust by sight, of instinctive belief by the light of comprehensive prin- ciples. It is little more than the exaltation and joy of our spiritual faculties as they enter on, and begin to occupy their inheritance — an inheritance which we are pleased to call that of eternal truth, though on the shifting surface of changing events, everything seems most transitory — of blessed truth, though most horrible and terrible facts are daily evolved before our eyes. Yes, the sense and the rev- elation of deep principles that undergird the world with abiding strength, and gather it up in the embrace of an ex- alted, a blessed purpose, are with us, steal in upon us in our best convictions, and yield the repose which a belief in its ultimate triumph inspires. There is inlocked in our lan- guage and our nature a belief in truth, central, adaman- tine, giving safe, benignant support to the universe of God. In like manner, we carry over to the false, the untrue, a farther concentration of opposition and rejection in the word, falsehood. We personify it as a distinct principle or power of mischief, believe in its weakness, and rejoice in its ultimate overthrow. Eo matter what may have been their character, few of any party have ever espoused false- hood as such, few have not felt that the confession of it would be the admission of ultimate failure. We recognize the vague way in which these words, truth and falsehood, are frequently used ; yet, nevertheless, we claim that there is in this tendency of the mind to recognize the inherent opposition of the true and the false, the ultimate, necessary victory of the one over the other — a latent belief in funda- mental principles and forces, which it is the vain, temporary effort of falsehood to cover up and counter-work. This em- brace of the real, as ultimately involving the ideal, and passing in evolution from excellence to excellence, is the fruit of the mind's discovery of truth and error, its hearty 342 THE SPIRITUAL FEELINGS. acceptance of the one and its rejection of the other; its satisfaction in the external plan of God. § 2. The next group of intuitive feelings, though of a more manifest character, and more prevalent, has yet much of the same subtlety, the same choice of persons and times. Indeed, these are features of the whole class of emotions of which we are speaking. It has, doubtless, been one reason of the difficulty with which the spiritual feelings and the intuitive ideas, on which they are immediately depend- ent, have been recognized, that they are not, like the phys- ical feelings, universally present with approximately equal power, but in many scarcely seem to exist at all, and in their full, intense forms to be confined to comparatively few. Yet the reason of- this is obvious. They are each of them dependent on previous culture, on a faithful, special, dis- criminating action of the understanding. The beauty of the world is not seen, or at least is but very partially and inadequately seen, without an inquiry into its structure and relations, without a discernment of the exquisite perfection of idea and workmanship involved in it. No more is the right understood without a broad survey of conduct, the tracing of actions to their consequences; without rising above the immediate current of the stream to see whence and whither its flow. The intuitive feelings, therefore, can only be strong and clear in the more penetrative and reflec- tive minds. They do not thereby cease to be universal or characteristic when their appropriate conditions are met. The esthetical emotions arise solely under the previous action of mind. Disorder, absolute and complete, can fur- nish no beauty, nothing to be admired, nothing intrinsi- cally to be delighted in. Order, arrangment, is the first step toward beauty, is the first, simplest product of taste. But this order is the result of thought. This arrangement will present itself as beautiful in proportion to the number BEA. UTY. 343 and variety of the ends it meets, and the ease and 'accuracy with which these separate purposes are fulfilled. A little formal order imposed by mere utility, simple convenience in the classification of material, is not sufficient, or suffi- ciently significant to excite and to satisfy the taste. It is not till more feeling enters into our plan, more variety, skill and precision of adjustment, that the elements of beauty begin to be clearly revealed, and the mind takes an additional delight in the work aside from each and all of the ends subserved by it. Gardening, architecture, music, are the arts least imitative — the arts in which the beauty present is most immediately the result of the combining power of the human mind. In each of these, mere order produces scarcely a sensible effect. It is not till the plan discovers high appreciation of the resources at the disposal of the artist, and great power and pleasure in combining and developing them — not till the product becomes thor- oughly emotional, and in its scope and variety betrays a mind and heart alike active, that it begins in turn to com- mand our emotion, and impress us, as the case may be, with the grace, symmetry, harmony, force of the concep- tion. Here, then, beauty throws us into appreciative sympa- thy with the thoughts and feelings of a worker; of one who executes well and powerfully, and delights in such execution — one with whom perfection is a thing esteemed, sought after, and includes far more than the immediate sub- ordination of the means employed to a useful physical end. It is this effort of the mind, without neglecting utility, to lift each of its works out of the mere routine of labor, off from the simple plane of service into an emotional region, — to make it in its excellence, in its skillful or affectionate or grand handling, a source of independent, superior, con- stant pleasure, that is the source of beauty, and of its com- 344: THE SPIRITUAL FEELINGS. mand over the heart. Not merely work, or good work, but superior, expressive, emotional work is its aim. The es- thetical feelings cause us to delight in such labor, and to go, as far as may be, to every undertaking crowned with garlands. If we pass to the beauties of nature, equally do we find that it is thought, aptness of arrangement, skill of work- manship, labor performed with infinite love and faithful- ness, that arrest the mind and gratify the heart. In pro- portion as many adaptations, many powers are gathered into a brief compass, and with a perfect finish and relation of parts united in one organic whole, are we climbing with slow gradations, with a thousand steps of varied progress, from the lowest life to the highest, from the plant to man, delighted with the goodness of the thought, the kind and abundant ministration of faculties to the well-being and ex- cellence of the final product. In each advance of beauty, there is more expression, because there is more and more perfection, more and more beneficent labor, till, in man we find the highest condensation of power and regard ; ser- vice, compactness, symmetry, finish, in their most perfect forms. Everywhere, then, it is the labor of mind and heart, the births of thought and feeling, the rational products of high intelligence and love, that arouse the sensibility of beauty ; and we are so constituted that we can not be indifferent to these qualities when perceived by us. A cold, intellectual apprehension does not exhaust them. They elicit a certain regard, assume a certain relation which we designate as beauty, and so call forth the pleasures of beauty. Such en- joyment on our part is (1) a crowning sympathy with excel- lence ; such perception (2) an additional incentive to high attainment. They are the thirst of an aspiring spirit for that which is beyond, which is above ; for that which it BE A TJTY. 345 knows it can grasp and enjoy. They take all barrenness, all deadness from simple intellectual movement, breathe through it desire, cause it to draw back the curtain between us and the ideal world, and fire us with the zeal of pursuit. While the specific character of esthetical emotions is very pronounced, their minor differences are very great. The same fruits have not all the same flavor. The most exquisite and characteristic tastes complete the circle, with an endless division and change of quality. In works of nature, plants, trees, landscapes, birds, beasts, men ; in works of art, painting, statues, poems; in varied objects, and in their yet more varied combinations, we find a con- stant change of predominant qualities, endless degrees of power, and ever shifting methods of expression. Hence arises in esthetical sentiments every shade of form and force, from impressions scarcely perceptible to those which wholly occupy the soul — overpowering emotions breaking out upon it like a flood. The flow of these enjoyments in the sensi- tive mind may be compared to the movement of music, now gay and cheerful, now common-place, now low and sad, now mysterious, now wild, now sublime, gliding from phase to phase of emotion with perfect ease and inexhausti- ble felicity. The scope, body, variety of feelings which are either in whole or in part of an esthetical character, are in sensitive, poetic temperaments very marked. A large share both of their gentler as well as more exalted pleas- ures springs from this source. The form of action which these emotions prompt is mani- fest. They always afford a mild, often a powerful stimulus to painstaking, emulative and refined action. They pro- mote the finish, the perfection, the beauty of every product of the hand or of the mind. They reveal themselves in the physical results of labor, and certainly not less in char- acter. Tlie restraints and checks of esthetic sentiments are 346 TEE SPIRITUAL FEELINGS. experienced constantly in manners and social customs, and, if the taste is keen and just, in the more deep, personal, spiritual traits of action. Indeed, nobility, magnanimity, the symmetry and proportion of robust, thorough, healthy virtue, can hardly be reached without a large infusion of this esthetic insight, which discerns, delicately and com- pletely, the formal as well as the intrinsic bearings of con- duct. The dependent, complementary relation of the es- thetic to the ethic sense cannot be doubtful. Some may strive to make the first a detached law of action, but it only performs safely and to the full its office, as it accepts the higher law, and aids in its complete application. Perfect beauty in man, its highest subject, is the strong and varied and delicate development of moral power — the infusion of all the members and means of life with this inner, true life of the soul — the flowing outward in limb, lineament, and language of those manifold forces and susceptibilities that spring from wholesome, healthy, physical forces, in the handling of a supreme, spiritual power. Taste rightly de- veloped can no more fail to distinguish morality from im- morality, to work under the one and against the other, than it can fail to discriminate between life and death, health and disease, exalt the first, and hide the second in its de- formity. Beauty stands in the same relation to action as right ; like it, it enjoins and forbids, rewards and punishes. It blows a more silvery trumpet, its notes are less clear, penetrating and decisive than those which break sternly forth from the lips of ethical law, yet they wind their way into many remote places, and persuasively bend into cheer- ful and perfect order the otherwise unpliant recruits of virtue. § 3. We have now reached the feelings which are most central and characteristic in the class to which they belong, the moral sentiments. The emotions just spoken of would VIRTUE. 34.7 lose much of their character were it not for their interpene- tration by those of the moral nature. It is this filtration of the higher sensibilities downward which gives coherence and authority to the recognition of truth, to esthetical feel- ings, which of themselves simply have little binding force. The only imperative voice in man's nature, is that of con- science ; all other authority is but the echo and reflection of this. In one view of the subject our moral nature may be said to be our entire nature ; since a moral quality and moral relation are imparted to all thoughts and actions by the presence of this supreme, supervisory power. In a more strict use, our moral nature includes those emotions which more directly spring from it. Conscience, the per- ceptive faculty, which, in an indivisible act, sees the right and feels the sense of obligation, is the centre of our moral constitution. Without it, we should have no affections, no moral sentiments ; with it, we find the whole atmosphere of our being irradiated, and a thousand colors revealed in objects, tangible, indeed, in the darkness ; endowed with odor and with flavor, but with no direct avenue of approach through the physical night to the intellectual day. Light does not more modify, I may say etherealize matter, mul- tiplying a thousand fold its intelligible signs, crowding them in from all quarters and all distances on the astonished mind, than does a moral perception affect our estimates of character, deepen in meaning, and broaden in time the re- lations of actions. The fundamental moral feeling from which all others spring is that of obligation. This, as regards pleasure and pain, is indifferent. It may give place to one or the other according to the attitude assumed toward the duties desig- nated. The blended, the indivisible nature of the intui- tion and the accompanying sentiment should be carefully marked. A sense of obligation, a mere feeling, with no con- 348 THE SPIRITUAL FEELINGS. viction to which that feeling attaches, is theoretically un- intelligible, and practically nnserviceable. An intuition of right on the other hand, which does not instantly assume the force and pressure of duty, loses its character and slips from the throne of the mind. Intrinsic quality and exterior form, the rational and the emotional elements, are insepar- ably blended, and give us a command, whose unquestionable authority, like that of one born to rule, is in the immediate fact, in tone, attitude, outspoken power. If obedience follows the intimations of our moral sense, there sets in a deep and deepening current of pleasurable feelings, of reward. The force and intensity of these emo- tions will depend very much on the degree in which thejudg- ments which sustain the action of conscience and prepare the way for its decisions have been cultivated ; on the re- lative force which the moral sentiments have secured in our constitution by obedience. Ethical feelings, like esthetical ones, are very dependent on cultivation. The reason of this is obvious, since in neither case are we dealing, as in exter- nal perception, with a direct, immediate faculty, but with one acting on previous intellections, previous conceptions of the mind, and therefore limited in its scope and correctness to them. It is evident that the character of phenomena should be judged by instances in which they are most man- ifest and complete, not by cases in which they are obscure and furtive. A powerful moral nature makes itself at once felt in the pleasures it pours in upon the obedient mind, of such degree and quality that the appreciative heart pre- fers them to all others, and purchases them at any price of suffering which can be exacted of it. Yet these enjoyments are of a tranquil rather than a violent kind ; a deep sense of satisfaction in the choices made, a thorough content- ment in actions done, an inner approval which anticipates a like outward acceptance on the part of the wise and just. VIRTUE. 349 The feelings which follow disobedience, though more irregular and unequal in their action, often dilatory and partial, when compared with those of approval and reward, yet frequently assume a strong, clear, undeniable character. Shame, guilt, remorse, willful opposition, and sullen despair may, in turn, hold sway, and make themselves as distinct as, and more bitter than, any other feelings which the heart ever experiences. For reaching this result, more or less time may be required. Repeated disclosure of the disasters of transgression, the accumulation of physical retributions, a revelation of pervasive law, hemming in and baffling the disobedient, may be needed to instruct the moral judgments and awaken the moral sense. When, however, a pause is given to the career of sin, when reflection and the intuitive results of reflection can no longer be averted, the force and direction of moral emotion are as certain as the pains or pleasures of sense, when things bitter or sweet are on the palate. The pains of indigestion may follow more slowly than disgust from food in itself offensive ; but the conse- quences are no less of a distinct and undeniable character. Moral sufferings may be postponed in more ways and longer than many other emotional issues of action ; yet the de- velopment of causes ripens them none the less certainly to their results. The whole history of the race renders the positive character of the moral sentiments as undeniable as the physical consequences of an unwholesome diet. The fear the cowardice, the apprehension, the boldness, the approval, the confidence ; self-condemnation, self-gratulation ; the re- proaches of conscience, the dismay, the despair attendant on wickedness achieved, the composure of assured conviction, the calm anticipation of suffering, the triumph over it, fill the records of history, are the staple of dramatic and heroic action. Heathen and Christian literature alike breathe in their more profound and earnest moods, one spirit: 350 THE SPIRITUAL FEELINGS. Says Juvenal : H But tell me, why must those he thought to 'scape. Whom guilt, arrayed in very dreadful shape, Still urges, and whom conscience, ne'er asleep, Wounds with incessant strokes, not loud but deep. While the vexed mind, her own tormentor, plies A scorpion scourge, unmarked by human eyes." The history of martyrs especially develops the moral forces in man, since, on these feelings, the struggle has turned. The cruel tossings of such a mind as that of Cran- mer, clear, conscientious, yet timid and distrustful, between fear and conviction, discloses as clearly as any thing can dis- close the nature of the forces at work, unless it be the vary- ing sympathy, the alternate charity and condemnation of succeeding generations, in view of the momentary over- throw and ultimate triumph of the moral sentiments in the fearful, bold saint. § 4. Personal qualities are greatly modified by the moral nature. Meekness, humility lose all servility and are made consistent with the utmost strength ; while moral firmness softens down without weakening the outline of character. Still more is this true of our feelings towards others. These, in the conscientious temperament, receive almost their entire force from the moral sentiments. The affec- tions, a distinct class of sensibilities, are our emotions toward others as moral beings. Admiration, love, sympa- thy, benevolence, forgiveness, charity, patience, indignation, contempt, shame, are feelings which though they may bear the same name with certain intellectual emotions are very different from them. Love a passion and love an affection, the indignation of anger and the indignation of a violated moral sense, are alike diverse sentiments in their relation VIRTUE. 351 both to enjoyment and to action. In the first relation, they may as easily prey upon happiness as promote it ; in the second, they can not fail of being productive of pleasure. In the moral sensibilities, the sharpness and the bitter- ness of the selfish element disappear, and the benignity, composure and patience of a moral impulse take their place. It is the intermingling of so many kinds of feeling, and of the words applicable to them, which confound the charac- ter of action, and the classification of this department. The direction in which the moral sensibilities find full- est play is that of religious sentiments. The relations and duties designated as religious are those which, by the feel- ings and the results involved, are fitted to act most power- fully on the conscience and affections. The religious emo- tions, therefore, seem at times to overshadow other forms of ethical action, since their intensity and scope bear some proportion to the interests covered by them — to the en- nobling, greatly stimulating presentations of the divine at- tributes. The foundation of religion is ethics, yet the ethi- cal form is often swallowed up in the deep, spontaneous play of the religious affections. If we consider the per- manent issues of happiness, of joy and peace, as settled in our own constitution by the moral sentiments, and the rela- tions of actions under them ; if we remember that nothing in our fellow-men is of more abiding interest to us than their character, than the moral purposes indicated, and line of conduct adopted ; and, above all, if we bring to mind that the deepest, the supreme play of feeling is towards God, chiefly known to us as a moral being, we shall see that the class of sentiments now presented are at once the most varied, the most full, the most central and powerful of our emotions. So pervasive are they, that they give coloring to intellectual feelings which they cannot rule, enter in a fragmentary form where completeness is denied them, and 352 THE SPIRITUAL FEELINGS. are brought in to intensify or modify or disguise sentiments intrinsically at war with them. The exact shades of ap- proval and condemnation, of contentment and restlessness, of belief and unbelief in them, are as endless as are the relations which men's actions bear to virtue. Their authority, their retributive connection with pleas- ure and pain, the undercurrent of fear or hope, of repose or alarm, of conscious virtue or acknowledged guilt, which they cause to flow through the soul, obviously assign them the highest rank in the highest class of feelings. It is a convenient use of words, and one we have occasionally employed, to designate collectively the primary incentives of the spiritual nature as tastes, in contrast with appetites and passions. § 5. In the following diagram, feelings are introduced which have not been discussed in the text, and farther divi- sions are made. The enlargements are self-explanatory. We do not present this classification as exhaustive. It aim# simply to define leading directions of the emotions and leading dependences. It serves, also, to show the com- plexity of the feelings, the way in which they blend with each other, modify and pass into each other, and the insuf- ficient, shifting terminology applicable to them. Indeed the word sometimes only implies and does not express the feeling intended. We might easily subdivide the ethical emotions, but no good purpose won d be subserved. They would still prove too subtile and pervasive for us. The one thing we emphasize is the degree in which the higher sink into the lower feelings, and transform them. Innumerable and most complex relations spring up between men by virtue of the moral constitution, and each relation has a new combina- tion of spiritual sentiments. The great facts, too, of relig- ion enter to impart a new elevation and range to these emo- CLASSIFICATION. 353 r General Sensations, Physical J Special Sensations, Feelings, j Appetites, [ Natural Affections. f Contentment. Unrest f Primary. Intellectual Feelings. Wonder. Surprise, Wit, Humor. Desires. (.Secondary. Incident to Wealth, Power, Honor. Truth, Beauty, Virtue. As being achieved. Spiritual Feelings. Incident to failure. Incident J £ le * sur . e ' to truth. 1 Enthusiasm, (Awe. ( Delight, To beauty. •>. Grandeur, ( Sublimity. In reference to ourselves. To right. To others, As achieved. by ourselves. By the aid of others. By others. As occurring. Through ourselves. Through others. To others. Obedience. Disobedience. Obedience. Disobedience. {Hope, Joy, Satisfaction, f Pride, J Vanity, I Courage, t Confidence. ( Gratitude, •< Good-will, ( Attachment. ( Admiration, •< Honor, f Emulation. ( Fear, ■< Disappointment, ( Discouragement. ( Humility, •< Shame, Mortification. 'Anger, Rage, Hatred, Malice, Jealousy, Envy, .Defiance. ( Contempt, ] Pity, ( Compassion, i Self-approval, Peace, Courage. Humility, Repentance, Guilt, Unrest, Remorse, . Despair, f Respect, I Praise, -{ Love, | Faith, t Reverence. ( Disrespect, Distrust, Censure, Justice, Aversion. Patience, Forgiveness, Benevolence. 354 TEE SPIRITUAL FEELINGS. tions. Yet complex or grand as the results may be, their elements are the simple feelings we have given. We have outlined the primary colors of the brilliant spectrum. The beautiful also unites with the right with farther transfiguring force. We only point the way to these fields of poetry and religion, we do not paint them. The posi- tion in which we have put our words of designation must often define the word; we cannot reproduce the atmos- pheric coloring of the spiritual heavens. CHAPTEE IV. DYANAMICS OF THE EMOTIONS. § 1. We have spoken of the three classes of feelings ; the physical, the intellectual, and the spiritual. We wish now to see them more collectively in their relations to each other in the formation of character and the control of ac- tion. The first class spring immediately from physical con- ditions, and, including incidental occasions of pleasure, have primary reference to physical well-being. At points they transcend this object. Taste, touch, smell, are means of simple, intellectual distinctions ; yet, it remains true, that the senses which are the avenues of feeling, the appetites, the sensations indicating special physical conditions, all have primary reference to health, to guiding action in nourish- ing and maintaining the vigor of the body. Even here, it can hardly be said, that " All pleasure arises from the free play of our faculties and capacities ; and all pain from their compulsory repression, or compulsory activity." Much less is this generalization of Hamilton's applicable to the remain- ing classes of emotion. It is the unhealthy and the healthy action, the unwhole- some repression and the wholesome repression, that give pain and pleasure respectively, if not at once, as an ultimate consequence. Pain enters frequently to arrest action, and not as the consequence of arrested action. Mere activity, voluntary though it may be, does not necessarily give the conditions of enjoyment ; these must depend on its relations to health. Neither does repressed exertion, involuntary 356 DYNAMICS OF THE EMOTIONS. though the restraint may be, define the conditions of physi- cal suffering. Overlooking the mental vexation of such constraint, the physical consequences may be agreeable. Physical pleasures seem to depend on the relation which ac- tivity and repose have to health, and not on their relation to the will of the agent. Some forms of disease provoke voluntary, fitful, restless, yet painful effort. Exertion or the want of it by no means explains the accompanying pain or pleasure ; we know through experience their general connection with physical well-being. Our pain and pleas- ures impart a direct stimulus to appropriate effort for the maintenance of the body, still more they instruct us as to its conditions and wants, and thus, in a secondary way, guide our action. They subserve the purposes of intellectual dis- crimination and of gratification. Intellectual feelings have relation to success, are pleas- urable and painful in proportion as this end is secured or lost. Unsuccessful activity, no matter how free and sponta- neous it may be, is always in the intellectual feelings which accompany it disagreeable, often intensely painful. Our physical and our intellectual enjoyments may not always harmonize. Effort in itself wholesome may fail of its ob- ject and occasion disappointment, and exertion crowned with the most flattering success may bring a severe infliction of physical penalties. The mind institutes its own ends, and afterwards finds pleasure, or experiences suffering, by its prosperity or losses in the pursuit of them. As the pri- mary relation of the intellectual emotions is to success in the ends aimed at, the pleasure and pain in this direction experienced act as stimuli to sagacity, and faithfulness in the choice and use of means. This is an instrumental, an intermediate field, and its enjoyments are of a secondary, intermediate character. Spiritual pleasures have reference to the choice of ends, ORDER OF FEELINGS. 357 to the marking out of lines of conduct, to obedience to higher impulses. These again may often fail of concur- rence with intellectual enjoyments. We have the satisfac- tion of success in the attainment of ends which we should never have chosen, and the moral rebuke may thus set in at the point at which the intellectual pleasure is most com- plete. Physical health and spiritual health are ultimate, and the secondary intellectual enjoyments can not avert the consequences of failure as regards either of them. Spiritual enjoyments and sufferings come in to enforce obedience — obedience to the law of spiritual life. They stand in the same relation to this, that physical pleasures do to the low- er life of the body. They are simple, ultimate, the moral sense sustaining them. With self-established authority, the conscience legislates for the whole man, and according as its commands are wisely understood and wisely applied, the minor physical and intellectual enjoyments are gathered up in these supreme pleasures of the soul. Our enjoyments are not thus simply the fruits of activ- ity, they are of such a character as to define its limits, and direct it to appropriate objects. The law of life in the whole man is indicated by them. The ends to be pursued, the limits to be set to activity, even in its right directions, are pointed out, with the accompanying injunction laid upon us of a skillful choice and use of means. § 2. The three classes of feelings now referred to, have a successive, rather than an equal and simultaneous hold on the mind. The physical feelings are most immediate, di- rect, importunate in their claims. The intellectual life is awakened through the physical life, in some sense follows it. The sensations, the appetites, the states of the body, are early and decided means of good and evil — means inde- pendent of thought, with a necessary and irresistible appeal to the sensibilities. The intellectual feelings, as secondary, 358 DYNAMICS OF THE EMOTIONS. involve a previous action of mind, are not strong except in » connection with considerable forethought, a somewhat broad survey of the relations of actions. For this reason, the de- sires do not set in in a deep, strong current, except in more advanced minds, or in the more civilized states of society. In a barbarous community, the immediate impulses are chiefly animal ; in a civilized community the desires come to rule the leading classes, while the appetites still bear sway in the lower ranks. The spiritual feelings are yet more tardy in their full development. For anything like broad, decisive action of our higher intuitions, there is requisite much previous reflection. As beauty involves the union of inner power with perfect form, there must be, for its due perception, a deep, discriminating insight into both. As the universal sway of morality arises from a clear perception of the dependence of individual and general well-being on the form and spirit of conduct in its every manifestation, it is not till faithful observation and protracted reflection have disclosed the character and issues of action, that the ethical impulse can And very complete application. In the outset, it is likely to be confined to a few negative precepts, cut- ting off the individual from gross violations of the right. Ten commandments or twelve tables expounded in the most barren way may seem its limits. Only the latest culture can open these into the pervasive precept of universal love. The most enlightened communities, therefore, as yet pre- sent a very partial government of the spiritual sentiments. When the artistic sensibilities have been awakened, they have hitherto affected but limited classes, and this in a par- tial, one-sided form ; sometimes even in direct violation of the moral sentiment which underlies all high acts. The re- ligious emotions also have been restricted in their action, and fragmentary in their character. The spirit and the force of a higher life have not, in their completeness, been FORGE OF RULING FEELING. 359 grasped, and we have had an ethics more or less at war with esthetics — an intense yet narrowed force, which could not discriminate and command all the elements requisite for its own most perfect expression. § 3. Were it not that communities — that successive generations of men, achieve a collective growth, which the individual is able to receive inductively from them, start- ing at the point they have already reached, this order of development in the feelings would make the condition of mankind comparatively hopeless. But the growth of soci- ety reveals very clearly this progress from the physical to the intellectual feelings, and in an incipient form is disclos- ing that farther movement by which the artistic and ethkal sentiments, under the perfect, harmonious rule of the higher impulses, shall take the supreme position amid the powers and pleasures of the human heart. When any one feeling begins to predominate in the individual or the community, many things concur to strengthen its hold. Take, as an illustration, such a desire as that for wealth. It soon becomes a strong current, plow- ing for itself a deep bed, walled on either hand, and not readily changed. The desire by repetition returns easily as an habitual one. Surrounding objects and pursuits are more and more contemplated in their ability to gratify this feeling, and therefore by their presence more uniformly bring it uppermost in the mind. Kindred pursuits draw together parties in whom the desire is already developed, and by emulation and the confirmation of like judgments, they inflame it in each other. Thus a large commercial city seems a very maelstrom of economic currents, and every individual, a separate particle spinning round and round under the same feverish impulse and waiting to be swallowed up by the same insatiable lust. The brood of feelings also warmed into life by a parent desire, unite at 360 DYNAMICS OF THE EMOTIONS. once in the same clamorous and importunate cries. Yanity, pride, the satisfaction of success, the fear of failure, all quicken effort, and occupy the heart, when for a moment the original impulse relaxes. The circle of secondary de- sires is momentarily enlarged as the means of gratification are placed within their reach, and the wealth acquired is often less and less able to meet the claims laid upon it by feelings which, without law or limit in themselves, become ravenous in proportion to the food given them. Thus external and internal circumstances are increasingly shaped to the ruling feeling, grow up more and more under it, institute claims in harmony with it, confirm the judgments which sustain it, and weaken and remove to a distance adverse emotions. From this household of depen- dents, from this pressure of a prevalent opinion, from these confirmed and consolidated convictions of the soul itself, it is difficult to find an avenue of escape. If we substitute an appetite for a desire, though there is less warping of the judgment, there is in its place a peevish, persecuting habit, not easily to be worn out or resisted. From this confirmed movement which the feelings for the time-being assume, it becomes necessary that the forces which work for progress, should find concentration, and also that long periods should be allowed them in which to possess and fortify the ground they may be able to win. The overthrow of one class of feelings, and their permanent replacement by another in a community, is a truly gigantic work, requiring often the slow eradication and correction of a protracted and varied experience. This fact is especially observable in the development of the spiritual sensibilities. The social, moral law lies be- tween persons, and is, therefore, every moment affected in its form by the parties to it. Gracious affections can not go forth toward ungracious men, any more than from them. FEELINGS OF ANIMALS. 861 Hard facts demand stern principles, and stern principles call for severe enforcement. Only as each heart is softened can all hearts be softened. Only as one sees clearly the path, can he lead others into it ; and no one can see clearly lines of action far ahead of the conditions by which he is surrounded. Each movement must be conjunct with ex- isting circumstances, and, by a series of actions and reac- tions, make way for the next movement. At each stage the animal appetites and the intellectual passions must re- ceive within themselves the modifying force of the spir- itual tastes. The two ultimate sources of impulse, physical and spiritual, do not remain apart under growth ; a vigor- ous interaction sets in; each class softens the other and gives it new conditions. The intellectual feelings, elicited by this double play of incentives, are correspondingly puri- fied and enriched. Time is demanded for securing the en- largement and unity of life, and its diffusion through every function. Time is the one unalterable condition of evolu- tion. § 4. The feelings of the animal, if the view we have presented of his endowments is correct, are almost purely physical. His courage is physical courage ; his fear, phy- sical fear ; that is to say, these states are imposed upon him directly by external objects. The one is the rushing in of nerve power, prompting to conflict ; the other, the deser- tion of the seats of strength by the energies of life — an im- mediate provocation, an inclination, to flight instead of at tack. Memory, giving rise to association, may indeed, in the higher animals, start trains of feeling and thus of ac- tion aside from the power of the object which is more re- motely their cause. Yet these feelings are comparatively limited. Little apprehension is shown except in the pres- ence of danger, and then not according to its real nature, but its sensible form. The alarm manifested by many ani- 362 DYNAMICS OF THE EMOTIONS. mals assumes a direct, instinctive character — the appropriate action evidently follows the sensitive impression without any intervention of judgment. The young of the partridge hide themselves instantly on the first intrusion. Barn- fowls are filled with immoderate and universal alarm as the shadow of the hawk glides by them. The actions of the lower creation assume generally this direct dependence on sensations, with an occasional intervention of the intellec- tual element of association. Having now the emotions completely before us in their relation to the mind and to each other, we are better able to decide on the merits of that theory which recognizes but two classes, resolving the spiritual feelings into the intellec- tual. This will hardly seem possible, if we fairly estimate all that belongs to the intuitive emotions. These higher sentiments so percolate downward, so tinge secondary feel- ings, giving them a new character and value, that it is diffi- cult to analyze out the purely physical sensibilities, and to see how far these, with the action of the intellectual facul- ties upon them, can be made the foundation of our rich, emotional endowments. When, with the utilitarian, we un- dertake honestly to construct our entire spiritual constitu- tion from these purely physical elements, we have a heavy labor laid upon us. Not only must the primary sense of truth, of beauty, and of obligation, be laid aside, all the af- fections which spring from them must be dismissed, and also that esthetical or ethical quality or flavor which inevit- ably pervades intellectual emotions, whose staple is physical pleasure. A rogue will pride himself on a certain honor, whose fiber and force are found in single threads of moral ity. A clown is vain of possessions, whose excellence con- sists largely in beauties hidden in great part from him. Do this work of analysis thoroughly, separate carefully out all but strictly physical feelings, and we shall find re- EVOLUTION OF FEELINGS. 363 maining very inadequate elements to be transformed by in- tellectual combination into the varied and profound sensi- bilities of a truly developed nature. The natural movement of tender sympathies must be made the means by which this vast superstructure is reared. Yet, in the powerful and growing consent of appetite and purely selfish impulses, how quickly and wholly would these feeble sentiments be swept away. How hopeless the effort to stay the actual forces of mischief in the world, not only with no sense of obligation in the mind, but no admiration of virtue, no perception of the beauty of excellence as such, no' delight in any form of intrinsic merit, but always and everywhere, a cold, gross, sensual judgment of actions and their results — the pleas- ure of compassion rated coolly at its scale-mark in a selfish mind, and with nothing farther to commend it, except as it can be shown ultimately to make way for physical indul- gence. Grade these pleasures of the body, give them each their numerical value, put the occasional play of natural sym- pathy with them ; let the intellect honestly, closely adhere to them ; add, subtract, involve, evolve, at pleasure ; and forecast in the long reaches of its calculations such periods as it pleases, and how infinitely short after all must these promises of sagacious action fall of those deep, instant,, noble impulses which our sense of beauty and of virtue- bestow. Virtue is useful because it holds in its right hand peculiar and unmeasured rewards, because it is virtue. It is not virtue because it is useful, because it is laden with baskets filled with fruits plucked from the trees of a sensual paradise. § 5. There are certain laws which control the feelings in their relations to each other. The first of these is, The more intense feelings are transient, the more moderate ones are relatively permanent. This law is more true. of. physi- 364: DYNAMICS OF THE EMOTIONS. cal pleasures than of physical pains. These last usually in- dicate a distinct fact, and are dependent on it for their du- ration. Strong physical pleasures involve a corresponding expenditure of physical force, and so exhaust their resources. This fact is an enforcement of the law of temperance. The law holds still more uniformly in intellectual and spiritual emotions ; in part, for the same reason, and also because ex- cited feelings fall into conflict with ordinary duties, because, seeking immediate gratification in violent action, they are satisfied by that action, and because, as intense emotions, they have less sufficient occasions than more moderate ones. A violent temper, therefore, is wont to be a volatile one ; while one slowly moved is correspondingly firm. Intense grief is followed by comparative apathy ; exciting pleasures by depression of spirits, and vehement anger by relative indifference. The evenly happy life must be fed by the milder, more sustained sentiments ; and the peace, the rest of the soul is found in the balance and correction of its feel- ings one by the other. The moral sentiments yield supe- rior repose, not from their own nature alone, but also from the restraints and rule to which they subject all vexing and exorbitant emotions. Esthetic pleasures are among the most peaceful, since they are among the most harmonized and proportionate, of the sentiments. Even a spiritual feel- ing that is excessive does not escape the lassitude of reac- tion. A second law is, Similar feelings sustain each other, dis- similar ones displace each other. Certain views and states unite easily, flow together and strengthen each other. Others stand in the opposite relation, and exist by mutual exclusion. Harmony is consistent with contrast. Indeed, this is one of the ways in which impressions are deepened and made complete. The intellectual view is made clear .and decided by uniting like with like, and opposing like LAWS OF FEELING. 365 to unlike — by agreement and by contrast. The latter is often the more effective of the two methods of deepening an impression. Harmony, as a condition of feeling, includes the presence of what is concordant, and excludes objects dis- cordant, lying in different parts of the intellectual and emo- tional field. It is opposed to distraction, to diverse emo- tions, and thus divided effects. Living facts are always struggling for equilibrium among themselves — for an organic dependence about a single centre. Hence a pre vailing tendency and a strong feeling strive to subordin- ate all tendencies and feelings to themselves, and to put the character in harmony with its dominant force. To this law also are due the jar of interruption, and the unusual wear of distracting duties. A third law is, New things make a strong emotional impression, old things a weaker one. Novelty not only awakens the feeling of wonder, it enhances all the impres- sions made by intrinsic qualities. The new is impressed upon us by our very constitution with a peculiar force, a distinct wave of sensibility, and is thus enabled to initiate a rapid, tidal flow of feeling, not otherwise possible. In early and in uncultivated life, that which is novel is sought for its immediate emotional character. The grotesque, the odd, the extravagant, the new, the news, give fresh excite- ment, and the intrinsic value or worthlessness of the matter offered to the mind is overlooked. When the powers are more mature or more cultivated, wonder becomes a second- ary, a briefly initiatory impulse, making way for the deeper satisfaction of recognized truth ; and when it fails to yield this pleasure, it drops away almost at once. This law is connected with another law of a reverse nature. Customary things are more pleasing to us, uncus- tomary things less pleasing. This is the result of passive habit — of the silent adaptation of body and mind to their 366 DYNAMICS OF THE EMOTIONS. surroundings. This law is likely to gain ground, as years advance, on the previous law ; and both, as arising from the accident of circumstances, need to be guarded against by a self-contained spirit. We should be servants neither of the old nor the new. A fifth law is, The feelings of men are harmonized and greatly increased by sympathy, they are divided and in- tensified by repugnance. We tend decidedly to share or to reject the feelings of those about us. The one impulse is that of sympathy, the other that of repugnance. A certain contagious force belongs to emotion. The swell of senti- ment among masses, like the surge of the ocean, is heavy, forceful, dominant. It is difficult to maintain feelings which are not shared by those about us ; it is difficult to escape the influence of those which are prevalent. The minds of men flow into each other, and come to feel and propagate, with increasing power, the same influences. Sympathy, strictly so called, does not change the character of a sentiment, it only disseminates it. The inflammable nature of the feelings by which assemblies, mobs, armies are laid open to conflagration, each firing his neighbor, till all are caught up in one uncontrollable frenzy, is a very famil- iar fact. A less significant fact is that of repugnance. If a division already exists between men, as in classes, or in religious faith, that division is broadened by repugnance. All tyranny and cruelty are enhanced by this law. The tears of an enemy beget contempt and aversion, not pity. Laughter we do not share seems to us foolish. If we join in a joke at our own expense, its sting is lost ; if we do not, the amusement of others angers us. A sixth law is, Objects and acts have, in reference to the feelings, intrinsic quality and associative quality. The last quality, which is frequently the stronger of the two, and may be more or less in conflict with native force, LA WS OF FEELING. 367 is due to the part that objects have played or are playing in human history. Our feelings become grouped in memory by repeated experience, and on each recurrence, restore by suggestive power a large class of emotions and incentives with which they have previously consorted. Like feelings are thus con- solidated into varied, powerful classes, which work together on the mind, one feeling never arising alone, but uniformly having present for its aid some of its familiar companions. We shall not understand the force of certain passions with- out comprehending the multiplied echoes which they find in the soul. The transforming power of association is a cardinal fact in the feelings. Still association can create nothing; it simply combines and recombines in many ways the occasions of feeling, working up life into an experience of a definite order. The animation of the feelings is also frequently depend- ent on the power of imagination. Our intellectual emo- tions arise in connection with sensible objects, and the vivid- ness with which these are present to the mind will deter- mine the degree of action in the accompanying sensibilities. The passionate and the poetic temperament are influenced by the images of the fancy. The clear and vivid pictures of the imagination arrest the attention, and arouse the pas- sions, till they come baying along the trail of indulgence, like hounds in full scent. The nature, character, and excitability of the emotions are diverse, but their activity at any one time depends, aside from direct influences, on these mental conditions. Arising out of intellectual action, they are especially affected by the circumstances and conditions of that action. While the pleasurable feelings are evolved, for the most part, in connection with successful activity, and the painful ones in connection with bafiled effort, we are not to suppose that 368 DYNAMICS OF TEE EMOTIONS. this fact explains their very nature, or identifies action with enjoyment. It only indicates the relation of our emotions to the ends of life, but leaves them each to be understood in its simple, intrinsic character by experience. The great diversity and mobility of the feelings, and the subtile way in which they act on the thoughts and are acted on by them, are the points we need most to mark. BOOK III. THE WILL. § 1. We have now to speak of the power of volition — the centre and source of free activity. Willing is distin- guished from thinking and feeling in its positive and pecu- liar character by a reference to consciousness — to that ex- perience in which its phenomenal nature is laid open. It, moreover, bears a different relation to action from that of either of the other two, and this may be pointed out. It stands in the last, the most immediate connection with ef- fort. Exertion is prompted by feeling, is anticipated and guided by thought, is initiated and maintained by volition. While the motive lies back in the emotions, the final deter- mination and executive impulse of free action are found in the will. The intellect is instrumental and interme- diate in its office. It presents objects to the feelings, and inquires into the means of their easiest, safest gratification. The voluntary powers are simple as compared either with those of thought or feeling. Our emotions present by far the most numerons, complex and varied features of the mind. Our intellectual faculties are relatively few, yet ex- ceedingly subtile in their inter-dependence and action. Our voluntary powers are yet more simple, and offer their chief difficulty in intrinsic character, in the problem of liberty. We shall first speak of the nervous system and of the func- tions of its several parts. Next we shall consider executive volitions, and later, the highest form of volition — choice. The first division of volitions is into primary and executive volitions. The ultimate choice is that which presents the 370 THE WILL. more remote objects of pursuit in reference to which other volitions are simply intermediate. The distinction of exec- utive volition and of choice while a very real one is also a very changeable one. Choices may have every degree of generality ; may be very near or very remote. As the primary act becomes more inclusive, a larger number of volitions are simply executive. As a subordinate question becomes more independent, it assumes the character of a choice. One may decide to build a house, and it may still remain to him to determine where, when, with what materials, in what style he shall build. The same volition may be termed in one relation a choice, and in another, an executive volition. Choices rise into more and more gen- eral purposes, and executive volitions sink into and unite with automatic action. CHAPTER I. The Nervous System. § 1. If it were asked, What one fact more than any other distinguishes animal life collectively from vegetable life % we should answer, a nervous system ; not merely because a ner- vous system is confined to animal life, but because it is a controlling condition in all higher forms of that life. This nervous system, though either wholly wanting or very ob- scure in the lower forms of animal life, once present, as- sumes increasing importance with every step upward. In the higher animals it is the administrative system, whose modifications contain a record of the history of evolution. A formal distinction between vegetable and animal life lies in the much greater freedom of motion, internal and ex- ternal, which characterizes animals. The primary function of the nervous system, out of which all others grow, is the facilitating of motion. Motion is made by it more rapid, more varied, more general, more concurrent. Yegetable life, in exceptional cases, exhibits definite and somewhat rapid motion under external stimuli; concurrent movement of parts within itself for an end, and functional activity in prosecuting that end. It thus presents, in the absence of a nervous system, a clear rudimentary expression of the primary offices of that system. But the last attainment of vegetable life becomes the first of animal life, and, as specialized in the nervous system, the means of its enlargement. This system, there- fore, is the pre-eminent structural factor in the new devel- opment. By means of it, functions of many orders proceed 372 THE NEB VO US SYSTEM. concurrently in constant interaction within the body ; every portion of the body is subject to more or less rapid change ; and special members of the body and the whole body are set in motion by external stimuli. Thus motion becomes a striking feature of animal life, and different forms of motion distinguishing features. The nervous system is the means at once of the inside organizing process, and also of the extended interaction with the environment which ac- companies it. A general knowledge of the nervous system, through which all these forms of activity grow up together, is necessary as a means of approaching the highest phase of development, that of voluntary action. Vital action is so closely united with the secondary forms of voluntary effort^ known as executive volitions, that we shall not be able to understand these without some general apprehension of the mechanism they employ, and its methods of play under simple, vital forces. § 2. Life we hold to be a superior, plastic power, work- ing pervasively, yet under one harmonious plan or impulse in all parts of the living body. This life, — this pre-emi- nent, peculiar and inscrutable power, whether we regard it as the immediate presence of the Divine hand, or as a dis- tinct existence, is the maker — the indispensable architect of that most strange and marvelous of structures, a living thing ; be it plant, shrub, tree, insect, bird, beast, or man. Molecular, chemical, electric, thermal forces are the means employed ; but these as much fail to explain the form and relations of the final product, the wonderful manner of its putting up and repair, as do the stone, mortar and timber, the digging, the hewing, and the heaving, the plan and pro- portions of a cathedral. The exact thing to be accounted for is that on which these blind forces cast no light. How came they to work in these marvelous relations to each other ; how to institute these unusual and strange condi- Fig. i. — Nervous System of an Ascidian. Fig. 2.— Nervous System of the Sea Urchin. F1G.3.— Nervous System of the Fresh- Fig. 4. — Nervous System of the Oyster, water Clam. a. a. Ganglia united about the (Esophagus. b. Posterior ganglion. c c. Gills. NERVES. 373 tions of various and complete life, a power which they no- where else exhibit ? We explain the action within the chemist's retort by the chemical properties of the material present, but the retort itself, the application of the heat, the proportion of the ingredients, the experiment as an experi- ment, must find a solution in a new, an intelligent agency. Account as we will for changes that go on in the blood, that there should be veins, arteries, such a fluid as the blood, and the needed combination of powers to propel it ; these and like adaptations which make up the living agent meet with no explanation in simple, molecular forces. Yet these forces always and everywhere intervene between the in- scrutable agent and the phenomenal result. Under a phe- nomenal form, they are the second facts which lie back of . the first — the massive product. Molecular movement is to the living structure, what the mechanical transfer of stone and timber is to the edifice. Many of the changes by which the animal structure is built up and renewed take place locally, by an action there instituted. But as the parts of the body are reciprocally interdependent, the changes of one part must be correlated and exactly har- monized with the wants of other parts. This transfer of vital sympathy is affected by the nervous system. § 3. The essential parts of a nervous system are nerves and a nerve centre or ganglion. It is the office of the nerves to carry stimuli to and from the ganglia. It is the office of ganglia to receive stimuli, and to redirect them so as to secure fitting, harmonious action. These two portions of a nervous system are distinct in structure and color. The nerves are white and fibrous, the dispersive portion of the ganglia is gray and vescicular. The gray centres are usually enclosed in the white matter. Nerves convey im • pressions in two directions, and are hence termed afferent and efferent. They also convey two kinds of stimuli, ' or 374 THE NER VO US S TSTEM. better perhaps are the mediums of changes which give rise to two distinct results, one of feeling one of motion. Hence they are termed sensor and motor nerves. Yet the ac- tion of a sensor nerve is most frequently not accompanied with a sensation. The distinction seems to lie in the dif- ferent termini of the two sets of fibres, closely united as they are in their sheaths and indistinguishable in structure. The sensor nerve starts in a sensor surface, and ends in the gray matter of a ganglion ; the motor nerve starts in the same gray matter, and ends in a muscle. The inscrutable changes which take place in a nerve in the transfer of an impression have no more likeness to the sensor or motor re- sult than have the electric states of a wire of a telephone to the sounds at the termini. The simplest action of a ner- vous system is termed reflex. It is the immediate response along the same bundle of nerves to irritation, by action in the part irritated. Thus the foot is tickled, and immed- iately withdrawn. The response to stimuli is usually much more complex. A blow is aimed at the face, the attitude is changed, the head turned aside,, the arm upraised, the pulse quickened, the eyes closed. There is a consensus of actions in one end. These responses may, without the in- tervention of consciousness, under organic stimuli, become very complex, the distribution being determined by that plastic power accumulated along the entire line of develop- ment. When these actions pertain to the interplay in func- tions of the organs of the body, they belong to the organic life. It is the office of this life to institute and sustain these relations. If they pertain to movements directed toward external objects, and still arise from organic stimuli, they belong to the instinctive life. This life has the same base as the organic life, and is hardly more than it. If consciousness intervenes, stimuli pass into sensations, while the responsive actions become still more extended and com- Fig. 5.— Nervous System of Nudi branchiate. Fig. 6.— Nervous System of a Caterpillar.* 3 Fig. 7.— Nervous System of White Ant. Fig. 8.— Nervous System of Fly. * Orton's Comparative Zoology. FORMS OF NEB VO US SYSTEM. 375 plex. This increase of complexity is due largely to the more distinct introduction of time. Sensations remain in the memory, and the experiences, also, which have arisen under them; they thus serve to combine actions through long periods. Actions and sensations united by the media- tion of consciousness constitute the associative life. By the intervention of reason, still broader areas and longer periods may be included, and the complexity of responsive move- ment becomes incomparably greater. These combinations are those of the rational life. These forms of life do not exist by exclusion, but by the inclusion of the lower under the higher ; by the building of the higher on the lower. § 4. The simplest form of the nervous system is that of nerves united in a single ganglion. An example of this is seen in an Ascidian, belonging to the class of Mollusks, A first step of combination is found in the union of several proximately equal ganglia, each with its own nerves, to each other by other nerves. Each stimulus thus affects not simply its own ganglion, but other ganglia, and the reac- tion is proportionately extended. The star-fish and the sea- urchin, belonging to the Radiates, present examples. In the star-fish, a circle of ganglia are gathered about the central opening, each ganglion being also connected by nerves with the ray to which it corresponds. Thus the stimuli of any ray act on all rays. A third form of the nervous system found in Mollusks, as the fresh- water-clam and the oyster, consists of several unequal ganglia, some- what irregularly distributed through the body, in connec- tion with leading functions, and united to each other by fil- aments or by cords. Thus the oyster has a large posterior' ganglion closely connected with the great adductor muscle,, with the mantle, and with the gills. It has two much' smaller ganglia, situated on either side of the mouth and; united above and below it. 376 THE NEB VO US SYSTEM. In the farther development of this division, the ganglia abont the oesophagus increase in number and size, and be- come the controlling nervous centres. Of this character are the nervous systems of the cuttle-fish and of the Aeoli- dse or Nudibranchiates. The head thus begins to take its final relation to the nervous system. It becomes the centre of the special senses, and the chief source of stimuli. This accumulation first takes place about the mouth, the leading organ of nutrition. As the attainment of food is a primary necessity, this effort is supported by all the senses. In the vertebrates, the mouth, though it is no longer surrounded by the great ganglia, is still supported by the sense of taste, touch, smell, and less directly, by sight and hearing. "While the higher life rises out of the lower, it is united to it by the old bond of service. In the lower Articulates, as larvse, many proximately equal ganglia corresponding to the several divisions of the body are united to each other by a longitudinal cord. The ganglia of the head only slightly predominate. To this bilateral symmetry the nervous system begins to conform, and the nervous cord and ganglia become double. In the higher Articulates, as insects, the ganglia show more subor- dination, and are frequently, as in the bee and the fly, gath- ered into the thorax and the head, the two seats of action and perception. It is here that instinctive life finds its highest development, yet, as in the white-ant, the ganglia are often markedly subdivided. In the Vertebrates, cephilization becomes far more com- plete. If we start with brainless-fish, and close the series with man, we have traveled from a uniform spinal cord to a complete centralization of life, and subordination of it in all its forms to consciousness. When we observe the great complexity and wonderful tharmony of action in the nervous system of man, and also -A Fig. 9. — Brain of Carp. A . Cerebral hemispheres. B. Optic lobes. C. Cerebellum. Fig. — Brain of the Rabbit. Cerebral hemispheres. O. Olfactory bulbs. C. Cerebellum. Fig. ti.— Arrangement of the Brain. The Corpus Callosum {A), the Corpus Striatum {B), and the Optic Thalamus (C) are shown in dotted outline. (Z?) Pineal body. (E) Corpora Quadrigemina. {H) Optic nerve. (I) Olfactory bulb. (L) Medulla. (M) Cerebellum. DEVELOPMENT. 377 observe the slow steps of progress by which this result has been reached, we shall be profoundly impressed by the fact, that all life is gathered up in man and epitomized by him. The several strata of acquisition lie one above the other and directly and indirectly support each other, as certainly as do the several layers in the earth's crust. Organic life pre- pares the way for instinctive life, and the two grow to- gether. These in turn, while sufficient unto themselves, are the needful platform for associative life, and associative life makes way for rational life, ready to use all below it. CHAPTEE II. The Nervous System of Man. § 1. The nervous system, the medium of action, per- vades the entire body of man. An intricate net-work of nerves lies over the surface of the body, spreads through its members, and is gathered in certain lines and centres of nervous communication. There is no pin-point on the skin that does not disclose the presence of nerves ; nor an organ which does not show itself when diseased to be wrought into the nervous web. The collective mass, made up of the cerebrum, the cer- ebellum, the pons Yarolii, the medulla oblongata and the spinal cord, constitutes the nervous centre, the cerebro- spinal axis. " Beginning with the spinal cord, — which we have seen to be a rod or column of white matter or fibres, enclosing a slender core of grey substance — if we trace the fibres of the cord upwards, we find them continuing into the medulla oblongata, the first and the lowest portion of the brain. Of the whole mass of fibres entering the medulla oblongata, the larger portion pass up into the pons Yarolii and the cerebellum, while a part terminates in the grey substance of the medulla itself ; and from that grey substance other fibres take their rise and proceed onward, in the company of the through-going fibres of the cord. Thus the emerg- ing white matter of the medulla oblongata is partly the fibres that entered it as a continuation of the cord, and pa: Ay the fibres originating in the grey central matter of Fig, 12. — Brain of the Cat Fig. 13.— Brain of the Orang-utan. Fig. 14. —Human Brain, side view.* Fig. 15.— Human Brain, upper view. * Orton's Comparative Zoology. STRUCTURE OF THE BRAIN. 379 the medulla, replacing, as it would seem, those that termin- ated there. From the pons Varolii, where we come next, the white fibres advance in various directions, intersecting with transverse fibres connecting the two halves of the cerebellum, and passing upwards towards the cerebrum pro- per. The fibres thus going upwards constitute the crura, pe- duncles or stems of the cerebrum, and seem destined to ter- minate in the grey matter of the convoluted surface of the hemispheres. But in passing through the ganglia of the brain — the thalami optici, and corpora striata — the arrange- ment described above is repeated ; that is to say, while part of the fibres proceed through the ganglionic masses, the rest stop short in the grey substance of those masses, which grey substance gives origin to other fibres to pass out with those that had an uninterrupted course through the bodies alluded to. Both sets together — those passing through and those originating in the grey substance of the corpora stri- ata, or thalami optici, constitute a portion of the white or fibrous substances of the hemispheres, spreading out and terminating in the grey matter, or cortical layer of the con- volutions. They are the first of three classes of fibres, de- scribed above, as constituting the white matter of the cere- brum, that is to say, the ascending or diverging class. "Whatever number of central masses we may calculate as interposed between the spinal cord beneath, and the con- voluted surface of the cerebrum, the manner of communica- tion between them is found to be as now stated. The fibres passing between one intermediate mass and another, are partly transmitted and partly arrested. Wherever grey matter exists, there is the commencement or termination of white matter. The fibres that enter the cerebellum from the medulla oblongata terminate in whole or in part in its outer layer of grey substance, and in that substance a new set of fibres originate to pass to other parts of the brain, 380 THE NERVOUS SYSTEM OF MAN. as the corpora quadrigemina, the hemispheres, etc., and from one half of the cerebellum to the other. The fibres spreading out, as already mentioned, in the hemispheres toward the convoluted grey surface, will have had very various origins. Some may perhaps have come all the way from the extremities of the body, passing by the spinal cord, medulla oblongata, cerebellum, pons Yarolii, thalami optici, etc. ; others have originated in the grey matter of the cord, passing without a break through all the interven- ing centres ; a third class may have had their rise in the grey matter of the pons, a fifth in the cerebellum, a sixth in the corpora quadrigemina ; others in the thalami optici, or corpora striata ; besides other more minute sources. The arrangement may thus be seen to resemble the course of a railway train. The various central masses are like so many stations, where the train drops a certain num- ber of passengers and takes up others in their stead, whilst some are carried through to the final terminus. A system of telegraph wires might be formed to represent exactly what takes place in the brain. If from a general terminus in London, a mass of wires were carried out to proceed towards Liverpool, and if one wire of the mass were to end at each station, while from the same station new wires arose, one for every station, farther on, a complete and per- fectly independent connexion could be kept up between any two stations along the line."* The nerves are divided into two classes, the spinal and the cerebral ; the one passing into the body along the spinal cord, the other directly from the brain. The nerves go forth in pairs from the spinal cord, passing out on either side between the vertebrae. Of this class, there are thirty- one couples. Each of these nerves is divided at its root, into two portions termed the anterior root and posterior * Quain's Anatomy. SPINAL CORD. 381 root. These portions subserve distinct purposes. The cerebral nerves are composed of nine pairs, — sometimes divided as twelve — four of pure sensation, terminating in the special senses, and five motor nerves. Nerves from either lobe of the brain terminate in the opposite half of the body. Thus the dexterity of right-hand work in- volves a discipline of the left lobe of the brain. This lobe consequently tends to greater power. § 2. The spinal cord is the means of sensation and of movement through the entire trunk and extremities. If this cord is cut, sensation and the power of movement by the will are lost in the parts below the point of separation, The power of movement nevertheless remains under local irritation after the division. Superficial irritation will cause a spasmodic movement, accomplished by a reflex action of the spinal cord alone. Movements closely resem- bling voluntary action, of which the individual is uncon- scious, and which he cannot control, will, under these cir- cumstances, take place in the limbs. The circle of nervous action is completed through the spinal cord, independently of the brain. The spinal cord, by virtue of its grey matter, is itself a nerve centre. Including in the spinal cord the medulla oblongata, continuous in structure and functions with it, we find that they, independently of the cerebrum and of consciousness originate and sustain many movements. They institute and harmonize a portion of the automatic action of the body. Of this sort are movements connected with digestion. After the food has passed the lips, been tasted and masticated in the mouth, and thus been fully subjected to inspection and voluntary action, it goes through the remaining processes of digestion, dilution, assimilation, without further consciousness of voluntary action. The contractions of the throat, the peristaltic movement of the stomach and intestines are accomplished by nervous stimuli, 382 THE NER VO US SYSTEM OF MAN. transmitted from the medulla oblongata. This portion of the nervous system chiefly sustains the muscular action in breathing. This is complex and rhythmical, nicely alterna- ting in the states indicated, and in the muscular action in- duced. To receive and combine the indications of the actual state of the lungs, and to distribute to the muscles the appropriate stimuli, so far as the movement is stated and involuntary, is a portion of the office of the medulla oblongata. This action is, moreover, capable of being modi- fied, arrested, or quickened by voluntary effort. In the same way the support and harmonizing of volun- tary muscular movements generally are referred with suffi- cient proof to the cerebellum. By far the larger part of this action is involuntary and unconscious, though voluntary stimuli reach and modify it. A portion of this sustaining influence of the voluntary muscles is received from the spi- nal cord, to wit, that which gives them always a certain ten- sion or tone, distinguishing them from lifeless flesh, and maintaining them in readiness for instant effort. That the harmonizing and co-ordinating of muscular movement are due to the cerebellum, is shown by proof briefly presented in the following passage from Todd and Bowman, page 50. " Animals deprived of the cerebellum are in a condition very similar to that of a drunken man, so far as relates to tneir power of locomotion. They are unable to produce that combination of action in different sets of muscles which is necessary to enable them to assume, to maintain any attitudes. They cannot stand still for a moment, and in attempting to walk, their gait is unsteady, they totter from side to side, and their progress is interrupted by frequent falls. The fruitless attempts which they make to stand or walk are sufficient proof that a certain degree of intelligence remains, and that voluntary power continues to be enjoyed." The cerebrum, on the other hand, is directly connected OFFICE OF CEREBRUM. 383 with all voluntary and conscious action. We present its functions in the words of Bain, to whom we are especially indebted in this connection. " Experiments have been made with a view of determining the characteristic functions of this cerebral mass, so large in the human brain, although dwindling to the most insignificant dimensions in the low- est vertebrate animals, namely, reptiles and fishes. "The convolutions are the portions most accessible to operations. The hemispheres have been seen above to con- sist of an outer layer of convoluted grey matter, and an in- terior mass of white, fibrous, or connecting matter. When irritation is applied to the hemispheres, as by pricking or cutting, we find a remarkable absence of the effects mani- fested in the other centres. Neither feeling nor movement is produced. This makes a very great distinction between the hemispheres and the whole of the ganglia and centres lying beneath them. " Pressure from above downwards, produces stupor. " The removal of both hemispheres in an animal has the following results: " First : Sight and hearing are entirely lost. " Second : Consciousness, including both feeling and thought, seems utterly abolished : so that whatever bodily activity may survive, the mental life is extinct. " Third : All power of moving for an end, all fore- thought, purpose, or volition, is entirely extinguished. This is an inevitable consequence of the preceding fact. For without feeling and the memory of feelings and ideas there can be no voluntary action. The simple act of seiz- ing food implies, besides the power of sight, the feeling of hunger, and the mental association of the appearance of the food with the satisfying of the feeling. " Fourth: The power of accomplishing many connected movements still remains. The actions of flying or walking 384 THE NER V. US SYSTEM OF MAN. may be sustained after the loss of the hemispheres, but in that case a stimulus from without is necessary in order to commence the action. As a matter of course, the automatic actions, those that we have seen to go on in the decapitated or anencephalous animal, may still proceed. " Fifth : The sensibility of the skin, and taste, and smell would appear to remain in a greatly impaired form. Such sensibility, however, cannot be of the nature of true sensa- tion, for to have a sensation is to feel. It may consist in some mode of reflex stimulation, operated through the other centres. By operating energetically on any nerve of sense, we may excite reflex movements extending over almost all the muscles of the body. " Hence it appears that the hemispheres of the brain are indispensable to the exercise of our two highest senses, and to feeling, volition, and thought." — The Senses, and the In- tellect, page 57. That the higher senses, however, aside from the con- nection of their ganglia with the cerebral hemispheres, act purely automatically, is shown by facts of vivisection given by Taine in his work on Intelligence, p. 155. " Here is a pigeon, whose cerebral lobes are entirely removed, but whose corpora bigemina remain. When I suddenly put my hand near it, it makes a slight movement of the head to avoid the threatened danger." This fact, then, presents automatic movement through the highest special sense, en- tirely aside from conscious control, or conscious activity. The ordinary action of the eye in perception is therefore a state superinduced on a purely organic power, and gives a striking illustration of the descent of the conscious into the unconscious life, and its control over it. The offices of the several parts of the nervous system in man may be briefly summarized as follows; all portions save the highest are mediums of general communication ; GRADES OF NEB VO US ACTION. 385 the spinal cord is a secondary centre of muscular action ; the medulla oblongata is an automatic centre of organic action ; the cerebellum harmonizes voluntary action ; the pons Yarolii is a subsidiary automatic centre to the cerebel- lum ; the cerebrum is the source of all conscious activity ; the mesencephalic ganglia — optici thalami, corpora striata, corpora quadrigemina, corpora geniculata — are automatic centres subordinate to the cerebrum. The sympathetic system, not included in the cerebro-spinal-axis, serves the purposes of purely organic harmony in strictly animal life. § 3. We are now prepared to understand that vital, ner- vous action which is not voluntary, but anticipatory merely. Its first most simple form is that of reflex action — super- ficial irritation returned directly from a nervous centre as motor stimulus. This, detached nervous ganglia accom- plish in animal life, and the divided spinal cord in man, An advance on this is seen in continuous, vital movement accomplished by a special nervous centre, like the medulla oblongata, wholly involuntary and beyond the cognition of the mind. A farther progress is seen in that mixed action which is chiefly involuntary, and sustained by a nervous centre as the cerebellum, which is not the seat of conscious- ness, but is intimately connected with a second nervous centre the cerebrum, from which it receives voluntary in- fluences. A still more intimate blending of the higher: and lower life takes place through the mediation of mesem cephalia ganglia. Stimuli which come through the superior senses as sensations are carefully united with fitting organic actions, — as clear articulation in reading — and shortly the two become an automatic combination. The highest action is that which remains throughout under the guidance of consciousness. That we should understand this blending of the automatic and the voluntary is indispensable to a right 386 TEE NEB VO US SYSTEM OF MAN. apprehension of the will. At this point, physical inquiry has been very fruitful in its influence on philosophy. Says Bain, page, 63 : " The conducting power of nerve fibre is attended with nervous waste, and the substance has to be constantly renewed from the blood, which is largely supplied to the nerves, although not so largely as to the vesicles. If now we compare this liability to waste and ex- haustion, with the undying endurance of an electric wire, we shall be struck with a very great contrast. The wire is doubtless a more compact, resisting and sluggish mass ; the conduction requires a certain energy of electric action to set it agoing, and in the course of a great distance becomes faint and dies away. The nerve, on the other hand, is stim- ulated by a slighter influence, and propagates that influence with increase, by the consumption of its own material. The wire must be acted on at both ends, by the closure of the circuit, before acting as a conductor in any degree ; the nerve takes fire from a slight stimulus, like a train of gun- powder, and is wasted by the current that it propagates. If this view be correct, the influence conveyed is much more beholden to the conducting fibres, than electricity is to the copper wire. The fibres are made to sustain or in- crease the force at the cost of their own substance. "The nerve force is propagated more slowly than an electric current through a wire. The rate has been esti- mated at about two hundred feet a second as an average. It is to be remarked, that a nerve is not a simple conductor, but is supposed to consist of a countless number of mole- cules, each of which has playing round it an electrical cur- rent, or currents, which are an obstacle to the simple or di- rect propagation. There is always a certain delay in pass- ing through the nerve centres ; a reflex movement occupies from one-thirtieth to one-tenth of a second under favorable circumstances, which is more time than would be required BRAIN IN MAN. 387 for transmitting an influence through the same length of nerve without interruption. When the stimulus is weak, a proportionately longer time is required to produce the cor- responding movement. We may hence infer that what is called nervous excitement is a quicker rate of the nervous current. The obvious facts bear out this view." These then are the means by which an external force is received, modified and distributed in centres ; by which an internal state directs and secures the succeeding steps in vital movement ; by which an inner impulse of the mind is made in muscular effort to reach the external world ; or by which these vital and mental forms of effort are inseparably blended. § 4. The superiority of the Vertebrates in the animal kingdom is indicated by the presence and growing power of the cerebrum, the organ of conscious action ; and the predominance of the voluntary life in man is disclosed at the same point. The supreme relation of the cerebrum in man to the other ganglia is seen in its absolute size, its rela- tive size, its increase in relative size through all the grades of animal intelligence ; in the number and depth of its cor- rugations, and the amount of blood it receives. Starting down no lower than the cod, we find the cerebral lobes smaller than the optic ganglia, and on a par with the olfac- tory ganglia. As we pass up they increase in absolute or in relative size, or in both, till, in man, they have stretched over and covered all the associated ganglia, multiplied man- ifold their relative dimensions, — relative in reference to other ganglia, and in reference to the size of the body — and become truly the superincumbent, overshadowing hem- ispheres of thought. It is, however, in the combination of relations that man's great superiority is found, and not in any one of them alone. In absolute weight of brain he is surpassed 388 THE NER VO US SYSTEM OF MAN. by the elephant and the whale ; and in weight relative to that of the body by many of the smaller birds. There mnst be farther considered the proportion of the grey to the white tissue.* The four particulars combined, absolute size, size in reference to the body, size of the cerebrum in reference to other ganglia, the proportion between the grey and the white matter, assign man his pre-eminent position. We shall see the importance of the last particular when we remember that the nervous system is an instrument of trans- mitting impressions, and also of thoughts : that physical ac- tivity and mental activity are dependent, the first more immediately on the medium of transmission, the white matter ; the second on the medium of interior activity, the grey matter. Great muscular development, as in the bird, may, therefore, carry with it a relatively large nervous de- velopment. "As we rise through the mammalian series towards man, we find not only a marked increase in the absolute bulk of the cerebral hemispheres, and a yet greater relative excess in their size as compared with the aggregate of that of the sensory ganglia, but an augmentation of their func- tional powers beyond all proportion to their size, which is derived from the peculiar manner in which their ganglionic matter is disposed. In all ordinary ganglia, the nerve-cells, on whose presence their special attributes depend, form a sort of internal nucleus; but in the cerebrum they are spread out on the surface, forming an external or cortical layer. This layer is covered by a membrane termed the pia mater which is entirely composed of blood-vessels, held together by connective tissue ; and thus a copious supply of blood is brought to this important part. But, the ex- tent of the cortical layer, and of its contact with the pia mater is enormously increased by its being thrown into * Journal of Nervous and Mental Disease, Jan., 1876. First Article. BRAIN IN MAN. 389 folds, so as to produce what is known as the convoluted surface of the hemispheres." •■ • • "In the higher orders of Mammalia, the convolutions are well marked ; but we do not find them either numerous or complex in their arrange- ment till we approach Man."* " While the brain in man is about one-fortieth of the weight of the body, it receives from one-fifth to one-sixth of the whole circulating blood, f In the position of the cerebrum, superincumbent and exter- nal, in its enlarged size and greatly enlarged activity, we have a plain physical expression of the degree in which in man the conscious life overshadows, envelops, and takes in charge the organic life. "It may be doubted whether a healthy human adult brain ever weighed less than thirty- one or thirty-two ounces, or whether the heaviest brain of a gorilla has exceeded twenty ounces. "f Yet the gorilla with a brain of twenty ounces would weigh twice as much as the man with a brain of thirty-one ounces. The largest human brain weighs sixty-five or sixty-six ounces. That the two functions of the brain, its motor function and its reflective function, are somewhat separate from each other, and somewhat exclude each other, is plain. Great power of reflection is attended by an indisposition to activ- ity, and great muscular activity by a like indisposition and inability for reflection. A small and active bird, like a snow-bunting, demands relatively a very large nervous sys- tem to maintain the heat and intense life needful to meet its circumstances. Such size, therefore, does not indicate corresponding intelligence. Though our knowledge is not sufficient to enable us to say, that muscular activity is always accompanied with a predominance of white matter, and reflective activity with the predominence of grey mat- ter, yet the relations of the two parts of the nervous system * Mental Physiology, p. 93. f Ibid p. 39. \ Man's Place in Nature, p. 120 390 THE NEB VO US SYSTEM OF MAN seem to indicate this. The conspicuous way in which the least thought interrupts the flow of perceptions is seen in an effort to count the flying posts when a railroad train is in motion. We easily discern them as separate, and can give a tap of the finger for each one, long after the motion becomes too rapid for a distinct enumeration. The distinct recognition of each post, and its addition to the number already reached, is relatively a slow process. CHAPTEK III. Executive Volition. § 1. "We are to distinguish executive from primary voli- tion. Primary volition is frequently termed choice, and there is no objection to the word, if we carefully exclude from it the intellectual weighing of reasons, the balancing of inducements, which often accompany it. The volition is not in these, but in the act which brings them to a con- clusion. A choice initiates a series of acts in reference to an object to be reached by them ; an executive volition re- gards the performance of these acts thus determined on. The primary volition is the true seat of freedom, since subsequent acts flow necessarily from it. This choice may indeed be reconsidered, but so long as it remains in force, so long as it is a purpose of the soul, the acts included un- der it flow directly from it, fixed thereby in their character. An alternative is presented to the first volition, not to those later volitions by which it is completed. These may be looked upon simply as the prolonged force of the first, as much so as the repeated shocks of the ricochetting cannon- ball are the results of an impulse received at once and in the distance. Executive volitions are successive points from which fresh executive impulse is given to a series of actions whose existence and purpose have already been determined. Some have striven to separate widely between volition and choice. The division is a secondary one, covered by this distinction between executive and primary volition. These 392 EXECUTIVE VOLITION. secondary volitions springing ont of consciousness, though properly phenomena of mind, become inseparably blended with those automatic, unconscious movements by which most vital action, and the larger share even of what is termed voluntary action, are sustained. § 2. The voluntary and conscious region of action is evidently very much more limited in the lower animals than in man. We might expect this from the much larger relative development of the secondary, nervous centres, as compared with the cerebrum — the seat or instrument of conscious activity — in the one case than in the other. A command of limbs, a power and discipline of muscle, which with man are the result of protracted training, are spon- taneous in the young of animals. ~No conscious, tentative effort seems to lie back of their powers. They develop themselves spontaneously, with the precision, certainty and rhythm of automatic life. Stimuli, sensations do their work directly, and when as feelings they enter consciousnes, they seem to depart thence with an automatic, rather than with a voluntary impulse, with the decision and certainty of a self -sustained movement, rather than with the hesi- tancy and uncertainty of choice. With primary volitions, secondary volitions also disappear, and the conscious and unconscious feelings, or, more properly, the feelings and unrecognized physical states, blend with each other in se- curing fitting muscular action. In man, in connection with choice, there enters into action a large element of both conscious and voluntary im- pulses, and these mingle with, and modify, and are sustained by, the involuntary action of lower nervous centres. In- deed, the acquisition of skill seems to consist in transfer- ring the nervous impulse from the conscious to the uncon- scious centres, or at least in sustaining the one by the au- tomatic action of the other. The distinct, conscious, volun- VOL UN TAB T AND INVOL UNTAR Y ACTION. 393 tary impulse of each effort in. the combined movement is lost, and the changing conditions developed by the progress itself of action, — be these recognized or unrecognized — with increasing, self-poised force sustain it. Here, we would look, so far as we would look at all, for the sub-con- scious region of Hamilton and others. It is in the case of the will found in purely physical phenomena which trans- pire chiefly in the lower nervous centres, or, if in the cere- brum, in it simply as a nervous centre and not as the agent of mind. Here the physical and the mental are closely united, inseparably blended with each other, and muscular education lies in substituting involuntary for voluntary con- nections — in establishing an independent movement which the mind may at any moment modify or correct, but is not called upon momentarily to sustain. Thus we quicken or check inspiration, though the ordinary action of the lungs proceeds independently of the will. Again, we wink when we will, yet wink constantly also under a purely vital im- pulse. The movements in walking are illustrations of this interlacing of the voluntary and involuntary — the slow dis- placement of the one by the other. A walk determined on, the mind may busy itself with other things, and the muscular play be unconsciously sustained. If, however, any portion of the way presents peculiar difficulties, atten- tion is renewed, and a voluntary stimulus quickens the muscles to the needed effort. The leap made, the embar- rassments overcome, the automatic movement again sets in. There is, perhaps, no more complete example of self- sustained action, reached as the result of protracted, volun- tary effort, than that of reading. In fluent enunciation, the organs of speech are modified each minute so as to express several hundred distinct sounds. These rapid and precise changes go on unconsciously. There is no direct, volun- tary impulse back of them. So far is this true, that it is 394 EXECUTIVE VOLITION. entirely possible to read intelligibly with no conscious rec- ognition, not only of the meaning of the words, but even of the letters which compose them. One, in moments of abstraction, may find himself at the foot of the page, with no proof of having passed over its contents, except the at- tention of others, and the point reached by the eye. Such reading, while it progresses, is as involuntary, as purely automatic, as the inhalation of the breath which makes it possible. Nor is the sensible effect of the images present to the eye on the muscles of the throat in guiding and im- pelling them any more surprising than the declaration each instant, at the nervous centre, of the state of the lungs, and the correlative return of stimulus. Executive volitions, then, are greatly modified by the interplay of voluntary and involuntary action; by the ease with which the second displaces the first, and yet can be restored at option to its former character. § 3. That the cerebrum is the exclusive seat of con- sciousness, or rather, that consciousness is directly asso- ciated with its action alone, has become very plain. Yet in one sense it is the most dependent of the nervous gan- glia, since the other centres minister to it, furnish its data, and its connections are indirect through them. The great mass of action, the automatic action of the body, is sus- tained by lower centres, while conscious and voluntary in- fluences alone pass out from the cerebrum. The will-im- pulse, striking down into this unconscious region, is blind as to the method of its fulfillment, pushing its way ten- tatively through automatic connections. We will an un- usual movement, as the successive separation of the fingers on the hand, or the articulation of a strange sound. We purpose the result while the means are hidden from us ; by some these are hit on at once, by others they are reached with difficulty. There lies below our conscious life a measurably DEVELOPMENT. 395 complete and independent organic life, whose functions are placed partially within the reach of mind, and among whose activities may be introduced many others of a voluntary character. The mind thus executes its will through the body, much as one accomplishes his service by a servant. Nor is the mind left without the means of training this ser- vant to increased efficiency. Its powers and tastes are slowly transferred as skill to the body, and begin to flow on in the current of inheritance as a new exaltation of physical life. Organic action precedes the nervous system. This sys- tem, automatic in its play, enters to enlarge these organic functions, and sustain them by a variety of muscular ac- tivity. In this unconscious circuit appear, in a reflex automatic form, the conditions of general sensation, and, later, the or- gans of special sensation. At length these threads of un- conscious relations are united still more closely by con- sciousness, and a strictly mental dependence begins to ap- pear. Sensations as sensations are associated by memory, and the new connections have the force of a judgment. On this plane, animal life, supported by a sensitive, complex organism and strong instinctive connections, develops itself. In man, there arise, finally, the intuitions, acting as a yet higher and more rational consciousness. The processes which before proceeded as feelings now transpire as thoughts and accept voluntary guidance. The rational life is put in full possession of its powers. The great mass of action in the animal kingdom lies far down in the dark region of or- ganic effects, where most of it, even in man, still remains, yet with him it has assumed quite new dependencies. In reference to the will, we may divide human actions, into four classes : involuntary, voluntary-involuntary, invol- untary-voluntary and voluntary. These actions dimin- ish in number and increase in significance in the several 396 EXECUTIVE VOLITION. classes as we pass upward. The second class includes ac- tion resting on an organic basis, as breathing, but pene- trated by the will. The third class covers action initiated by the will, but passed over to organic connections, as those of walking or reading. In the fourth class, the dependen- cies remain variable and purely voluntary. It is doubtless impossible to find any complex movement which is wholly supported by executive, voluntary effort. Even when we utter our own thoughts, though attention and purpose are constantly present, a greater share of the muscular move- ment, that imparting motion to the lungs for instance, is of an involuntary kind. In struggling to give a difficult sound from a foreign language, the effort seems for a time to ap- proach independent, voluntary exertion, and is often for that reason very unsuccessful. In reference to consciousness there are three classes : unconscious, semi-conscious and conscious activities. Con- sciousness always abides with the mind, and the different degrees in which it penetrates the physical constitution are due to the character of the special and general stimuli. Along the lines opened up by these sensations, the judg- ment and the will are operative, reaching indirectly much beyond consciousness. As the intellectual world as a whole has been slowly built up on the physical, organic world, and receives sup- port from it ; so man erects his voluntary, comprehensive, and spiritual powers on the automatic forces, the nervous connections, the associative combinations, which lie beneath them. Thus in language we are often as much struck with the sort of blind help which words themselves give us, as we are with the fact that they have all grown out of a living thought-process. Indeed, the two impressions are opposite sides of the same thing ; words are infused with life and so yield life. Everywhere the lower grows up to, DEVELOPMENT. 397 gives way to, and enlarges the higher. The primary ma- terial of knowledge are sensations ; to these comes the new element of intuitions, and the true intellectual activity merely foreshadowed in lower relations is initiated. Sensa- tions pass into perceptions, associations into judgments, in- stinctive action into voluntary life, and all into conduct and character. The strictly physical and the organic world include facts only that are borne forward by inherent forces. Upon these are slowly superinduced the experiences of conscious- ness, phenomena that find no expression or measurement outside their own circle. These sensations, appetites, affec- tions begin at once to be united by a concomitant power, new like themselves, that of memory, into a definite expe- rience more and more taking the place of organic, instinc- tive guidance. Herein is a higher animal life, wrought out on its own plane, and constituting increasingly the sig- nificant feature in development. In this associative life, though its data are conscious data, its processes conscious processes, all connections are limited to sensible things, and to the combinations which have arisen among these in ex- perience and been impressed upon the memory. This life, therefore, lies quite below the rational life that, with a marked transition, is built up upon it. The distinctive feature of this new stage of development is the discernment of relations, the penetration into the substratum of forces and powers, and the construction, out of these abstract data, of the world of thought. Sensible impressions no longer occupy exclusively the consciousness, or control its connec- tions. The ears can be closed and the eyes shut, and the most active and productive mental processes go forward with increased advantage. Mind thus becomes spirit, enters a spiritual world, does its work among things invisible, and draws from them its most efficient motives. 398 EXECUTIVE VOLITION. But this new life must be an embodied one, must be put in relation with the world of sensible qualities, spring out of it, and return to it. The medium of this connection, the symbol of these new experiences, is language, whose controlling feature is its power to express relations, to des- ignate and hold fast the abstract. By means of language the mind climbs out of the world of sensations into that of relations, and works there the great constructive labors of thought by which the phenomena about us become the ex- pression of the ancient forces that have borne the creation on their bosom until now, and are sweeping with it placidly by us into the eternity before them. Spaces quite beyond the interpretation of the senses, times into which our time falls as a drop in the ocean, forces which invisibly interlace all visible facts, powers which potentially abide in this field of effort and work out there their own ends, these are the rational elements of a rational life. § 4. That mind is not the product of cerebral action but that cerebral action is the medium of mind in reaching the physical world, seems probable when we consider the facts of evolution simply. The vital power in many and cun- ning combinations precedes the nervous system. This sys- tem has been from the beginning simply the means to farther development in a direction previously indicated. Again, the automatic action of the nervous system has pre- ceded by a long period its conscious action. Consciousness has not been an inseparable function of nervous action, but has entered as a later additional term. Consciousness has been superinduced on a system relatively complete within itself. The higher is not added for the sake of the lower ; but the lower is put to the uses of the higher. So true is this that the organ of consciousness, even after it has been woven into the nervous web below it, can be removed, and a large portion of automatic action remains. That the last MIND AND CEREBRAL ACTION. 399 sensor state, in its passage into the cerebrum, is not united causally to the first motor stimuli, issuing from it, is prob- able ; (1) for if this were true, the cerebrum would simply repeat the functions of lower ganglia ; and (2) in that case consciousness would be a superfluous addition. Plainly consciousness intervenes between the two in a way that in- terrupts simply automatic connections. In this fact lies its entire significancy. CHAPTER IV. Primary Volition or Choice. § 1. We have now reached that central point on which all volition rests, and every effect. In choice we find the home of liberty, the highest expression of power, the un- conditioned support from which hangs all the chain of linked events. Some divisions have been made in choice, which have value in practical morals, but little interest in philosophy. They mark the relation of choices to the action and charac- ter of the person whose they are, and not any inherent dif- ference in the volitional acts themselves. Thus an ulti- mate choice is one which has reference to the most remote, or at least, the most general and inclusive ends of action. Thus a choice of virtue is of this nature, since it at once sets a limit and law to all other volitions, made secondary in their relation to this. A choice of pleasure to be pur- sued directly and everywhere is also of this character. Such choices have more frequently a theoretical than an actual existence. The pursuit of pleasure usually arises under detached, limited choices fastening to some object at no great remove in advance. The universality of such vo- lition is of a quasi, not of a formal character. Even the choice of virtue is often made by a specific surrender to a given duty, rather than by a broad forecast of the entire field of effort — is the settling the struggle of life under an example, instead of a principle. Desultory volitions are also spoken of ; that is, volitions which spring up one side of the leading line of action and CHOICE. 401 are directly or indirectly at cross purposes with it. Thus one whose general pursuit is that of pleasure, gives way transiently to the claims of right, and one usually obedient to duty, for a time, tarns aside under some peculiar tempta- tion. Of these choices, practically there are many ; and while their moral bearing is most important, as choices they present no points of particular interest. Life is more fre- quently expended under the impulse of general choices, — not assuming the character of a single, ultimate choice, though as certainly as such a choice, throwing action into one direction — and under desultory choices, bending with- out reversing the current of the soul. Thus actions flow onward, submitting to a gravitation they may not have rec- ognized, and yet, in never-ending circuits and turnings, betraying the influence of the passing hour. § 2. Passing, then, these distinctions in the relations of volitions rather than in their character, we have only to con- sider simple choice, the primary act of the will, the chief ex- pression of spontaneous power. We shall speak first of what is involved in this notion of free-will, and later, of the proof of its existence. As liberty is a primary, simple relation, we must define it by cutting it off from other things, by denying of it those qualities which have become attached to it from abroad, reflected upon it from the physical connec- tions of the world below it, and then leave it to be under- stood and accepted by the intuitive grasp of the mind itself. Liberty is not, as some would have us believe, found in the absence of outside coercion. If this were liberty, the plant would be free in its growth ; since this proceeds under no mechanical, external impulse, is the result of the action of inner forces. When we say that man is free, we do not, in the higher use of the word, mean to affirm that he is not bound or imprisoned. The ordinary significance of language makes this point sufficiently plain. 402 PRIMARY VOLITION OR CHOICE. By the word choice, we intend to cut off all efficient forces, that is all physical forces, external or internal, me- chanical or vital, from any control over, or direct effect upon the action which is so designated. The commence- ment of the line of effort which springs from a primary vo- lition — a volition, as we shall concisely term it — is absolute and complete. We do not affirm hereby anything concerning the exact manner in which the train of physical forces is set in mo- tion by volition, but only that it does, of its own power, in- itiate the physical movements which follow. These ener- gies may lie in store for it, ready to be used, but the will liberates and controls them. The will, then, in the first place, stands above and beyond the range of causation, even in its most subtile forms as presented by nervous energies and influences. It descends upon and uses these, is not evolved by them. By the limitations now given, all reflex action, all auto- matic action, under the play of the senses and appetites, are as physical states excluded from the realm of liberty, are but the higher forms of physical action. Equally are those executive volitions which have received their impulse from above, those acts which follow directly an intellectual weigh- ing of means, a balancing of probabilities, a deliberative movement which is simply the gathering and eddying of ex- ecutive force looking for a new avenue, the* best avenue for advance, cut off from the freedom which attaches to choice. Having reached a point wholly unaffected by physical force, we are to inquire what are the conditions of liberty. The inducements to action in the will lie before it, not behind it ; they are motives, not causes. There is no opportunity for choice, unless there are two or more of these, — and as by successive rejection they at length assume the typical form — unless there are two motives or lines of action. Neither CONDITIONS OF CHOICE. 403 is there properly opportunity for choice unless these two are distinct motives, subordinate to distinct ends. If the relation is one of means simply, it does not involve an act of choice, but one of intellectual estimates, of judgment. As the word choice is applied both to selection and election, both to the purely mental act deciding on adaptations and to the volitional act deciding between courses of conduct with different and independent moral characteristics, we easily confound the two. These motives, then, must be present, and so present as to furnish a true alternative of action — not a seeming one. Five dollars as opposed to ten dollars, as detached, single considerations, constitute no true alternative. They are exactly of the same kind, and, in or- dinary states of mind, there is no basis of action on which the less can be preferred to the greater, since that which gives value to five dollars gives double value to ten dollars ; and to feel the first inducement without feeling the greater force of the second is simply to disclose a defective estimate, or an abnormal state either of the mind or of social wants. In all cases of which this is a type, there is no proper free- dom. The mind can only choose the less valuable, the less desirable of things, like in kind, by adding to the smaller in- ducement a distinct, factitious consideration, as that of evinc- ing independence, or the exhibition of eccentricity. If, then, all motives are resolvable at bottom into impulse, and measurable on one standard, we assert that there is no real liberty, but only that semblance presented by an intellectual inquiry into the intrinsic value of things not bearing their sale-mark on them. § 3. There is necessary, then, to liberty not only two motives, but motives unlike in kind, resting back ultimately on different principles, revealing different forms of good and phases of character. In other words, there is no choice without the moral element which can alone oppose itself to 404 PRIMARY VOLITION OR CHOICE. all varieties of physical good, and present a distinct ground of action, a reward incommensurable with any sensual pleas- ure. The esthetical element indeed, as infused with ethical sentiment, may furnish a secondary feature in that contrast of action which gives a basis of choice. Two such motives being present, the question returns, What is their relation to choice ? We answer : They in- fluence the will without in any sense controlling it ; here we have reached the final, inexplicable thing, liberty. The will can, by its own power, take either of the two lines of action to the rejection of the other ; can feel motives to any degree, yet refuse to yield to them. The will, with spontaneous independent power, initiates the one or the other of the two courses of action before it. Here is neither fatality nor chance, causation nor fortuity. The will feels, without submitting to motives, and discloses in itself a true beginning of action. § 4. There is one view of liberty which needs to be guarded against, and, in the rejection of it, we shall have defined sufficiently the conditions of choice. It is this. The will always does yield to the strongest motive, not of necessity, but as a fact. In the first place, this theory in- curs all the difficulties of the view, that the will yields to either of the two motives by an impulse or decision rest- ing in itself alone, without its advantages. By motives in this discussion, we understand not simply the outward, ob- jective element, but the inner, subjective one as well, all in short that makes them motives. Influences are influ- ences only through the susceptibilities on which they play, the desires they evoke. The one theory affirms that these motives may be spoken of as stronger and weaker, and that in each case of choice that motive prevails, though not necessarily, which is the strongest. The other theory asserts such a distinction of motives is impertinent, and the will OBEYING THE STRONGER MOTIVE. 405 itself, in its freedom, is the sufficient and entire reason of the volition that follows. The mind, in the act of choice, is no more ruled by its own states than by external conditions. If it were, liberty would as certainly disappear, as if, in the outset, we placed the will within reach of physical forces. We should do with two steps what we had refused to do with one. The present state of the sensibilities would be determined by previous states, and these by constitutional endowments and external circumstances, and thus the threads of influ- ence, the lines of causation, be at length lodged elsewhere than in the will. Each volition would be the frnit of con- ditions which it itself had not determined, and thus be as certainly interlocked with the flow of forces as is the mill- wheel which revolves in the stream. The one theory evades this result by saying, that the stronger motive does control the will, yet not necessarily. The choice may be, though it never is, against it. The other denies the applicability of the conception, greater and less, and affirms an absolute, unqualified freedom, finding and seeking no explanation in forces expressed in motives. This admission, that the will may choose the line of ac- tion supported by what is termed the weaker motive, in- volves philosophically all the difficulties of the view which represents it as alike independent of both incentives, and making either a true alternative to the other. There is no philosophical obstacle to supposing that the will does some- times do what it is admitted that it may do. The state- ment of an action as possible involves the concession of grounds sufficient to render it intelligible, if it should actu- ally occur. No law of mind can be violated by the happen- ing of that which these laws suifer us to regard as possible. We must rely on special reasons, not on general principles, to establish the impossibility in given cases of that which 406 PRIMARY VOLITION OR CHOICE. we have granted to be a theoretic possibility. We can find, therefore, in philosophy alone no sufficient reason for say- ing in the same breath, that a thing may be, and denying that it ever will be. The last assertion must rest on some special, empirical reason; since the first assertion sweeps the ground of philosophy and says, that there is nothing to prevent it. Our philosophy, then, as philosophy, is no more encumbered with the assertion, that the will does cho than with the declaration, it may choose, either alternative. The general principles which admit the one statement, will cover the other. The fact, that an admitted possibility never does become actual, must be established, if established at all, on special reasons peculiar to the case. If there were a general principle or law against the action, it would not remain possible. Moreover this theory establishes an inductive law, of the strongest possible character, against itself. Admittedly, the weakest motive, so termed, never is chosen. There is an absolutely uniform line of action in innumerable and most diversified cases. No law of induction is established on stronger grounds. Yet, when we arc just about to reach the conclusion, that what, under no circumstances, is or ever will be done, is an action excluded by the very nature and method of the forces at work, we are suddenly bidden to face around by the very unexpected assertion, the choice under discussion is one that may constantly be made. On what ground does this odd inversion restS Not on that of induction, for this line of argument prepares the way with well-nigh irresistible power for exactly the opposite state- ment. Not on philosophical principles, for, as previously shown, these principles would show that what may at any time happen, probably, under the inexhaustible variety of circumstances presented by human life, will happen. This assertion, then, that the will may, but never does, choose OBEYING THE STRONGER MOTIVE. 407 the weaker motive, grounds itself neither on experience nor philosophy. It is an ill-gronnded affirmation under either view of it. § 5. Again, to what a mere shadow does it reduce lib- erty. We are free by virtue of a power never put forth. If we could not accept the rejected alternative, we should not be free ; yet, one of the two alternatives always, before choice, stands in such relation to the will, that it never ac- cepts it. The action of the will is practically as fixed by antecedent conditions as any line of causation. One might as well claim that a python should walk because of cer- tain rudimentary limbs hidden under its skin, as to annex all the fearful consequences of sin to such a hypotheti- cal power as this — a power that has never found exercise, subserves no practical purpose, and is only possessed of a metaphysical existence. To sustain the ponderous chain of sin — its interlocked links reaching through all eternity, its galling weight crushing the life of myriads — by so theoreti- cal and fanciful a support, can certainly never subserve the purposes of actual moral government. Farther, a will of this sort, is wholly superfluous. If motives have superior efficiency, and this efficiency is always yielded to, why should any volition intervene ? There is a power present able to secure action, and that does secure the action that actually follows. Why should not this surplus of power, this over- balance of influence, be left in an immediate, precise, inevitable way to reach its own results. Are we to insert another wheel, itself with no practical connections, only that we may band to it the moral universe, and assert responsibility ? If so, let us, in the name of virtue and honesty, give it some other office than that of simply bearing inward a power already existing in com- pleteness in the motive. To deal thus subtly with one's moral judgments, to practise upon them with these e vanes- 408 PRIMARY VOLITION OR CHOICE. cent distinctions and cunning subterfuges of words, itself approaches wickedness. § 6. Whence springs this distinction of motives into stronger and weaker but from a false analogy with the forces of the physical world ? We are not to attach to the word influence a definite, measurable power, capable of nu- merical comparison with like powers. If our pleasures were all referable to one sensorium, something of this sort might be admissible. But they are not. A moral gratifi- cation can be expressed in no terms of greater and less with a sensual indulgence. Were it not for our moral, our rational nature, were we wholly physical, the conditions of liberty would indeed disappear. We might weigh the claims of the senses, assign a numerical value to indul- gences, and trace the rise and fall of motion along this new meter of the appetites and passions. But nothing of this is possible, no approximate estimates of pleasure are possible when the moral nature enters into the calculation ; when the supreme claims of conscience afford a full and fair alternative to every degree and form of self-indulgence. We should have no occasion for freedom, were it not for the self-imposed law of the moral nature, and in issuing a command it also gives the conditions of that liberty which enables us to obey it. There is no such final reference of motives to the same or like sensibilities, by which we are able to pronounce them greater or less. There is no com- mon term or point between mere pleasure and duty. We cannot take the pleasure of a glass of wine from a sense of obligation, and give a numerical remainder. But if there is no antecedent standard by which motives may be measured, it is a mere circle of words to call that the strongest motive which does prevail, and then to repeat the assertion in the form, the will always chooses the strongest motive. There must be antecedent measurement, OBEYING THE STRONGER MOTIVE. 409 — and there is no such measurement — or our language means nothing. This view overlooks the office of the moral nature, the transcendent purchase and power that it gives to choice. It confounds simple, intellectual discrimination between enjoyments, or, .still worse, a certain automatic adjustment and balance between animal impulses, with choice. Liberty keeps aloof from this lower region. It reposes on extended wing in the upper air of our rational, intuitive powers and emotions. There is, and of necessity must be, a moral character to every true act of choice, since the higher im- pulses must enter to break up and rule out these estimates of greater and less, these automatic adjustments of influ- ences essentially one. The sense of guilt which accompanies a moral struggle sustains the view we have presented. If the guilty party could feel that he had yielded to the strongest motive, that a balance had been cast up between motives, and he had ac- cepted the largest sum proffered, the sense of condemnation and shame would be very different from what it now is. In proportion, however, as the transcendent, unmeasurable character of virtue is present to the mind, are the accom- panying moral struggle and the subsequent sense of guilt strong and bitter. The more declared the sin, the more clear the knowledge of the high nature of the things re- jected. It is the increase of light and motive, not their de- crease, which evokes the forces of moral retribution. The mind is not allowed to console itself with the assertion, that at the time and under the circumstances, it actually chose the strongest pleasure, the highest good. Its infinite folly, its unaccountable guilt are enforced upon it, not its sad mistake, its grave misjudgment. § 7. Against the notion of liberty, absolute and com- plete, now presented, it may be urged, that it admits of no 410 PRIMARY VOLITION OR CHOICE. control, that its action can not be anticipated, and hence pro- vided for. Now liberty is limited to the alternatives before it. It can not choose anything, bnt only one of two things ; and it is unsafe to give the opportunity of choice, when we are not ready for the acceptance of either of the things offered. Liberty is simply a larger field of activity, the opening of two lines of effort instead of one, and this is often found very easy even for a man in his control and management of his fellows. It does, indeed, make of gov- ernment a higher art, but does not in skillful hands take away its perfect efficiency, all the efficiency contemplated. Liberty provides for less, recognizes less of a certain sort of control than does slavery. The inevitable, mechan- ical movement of necessary forces is, indeed, lost ; but there is substituted a nobler movement, because it is a freer one, manageable in a different measure, and on different prin- ciples. Those who prefer the clang and ceaseless on-going of machinery may not be pleased ; but the product itself, nevertheless, is every way superior. Moreover, will is constantly declaring itself under its own liberty, establishing a movement and revealing a char- acter more and more manifest to those who have to deal with it. The virtue of a virtuous man does not cease to be free, nor the vice of a vicious man, because the choice of each is not momentarily altered. A free action remains free, no matter how far pursued, and those impulses of the rational life once revealed become more and more declared and fixed in their directions. The conduct of a perfectly virtuous being is among the most calculable forces in the whole universe, and this without the least loss of freedom. We manage events readily which turn on moral evidence, yet the connections are not absolute, are not seen by us to be certain. There is the same difference between causation and liberty as between demonstration and evidence, proof PROOF OF LIBERTY. 411 and argument. Each subserves a feasible, practical pur- pose. It may be farther objected, that liberty so defined is syn- onymous with chance. It is not. The ground of action — and there remains a most adequate and complete ground — is simply transferred from the motives to the will, from the outside to the inside, from secondary and causal agents to a primary and independent one. We must, indeed, give up all hope of conceiving this under forms of the imagina- tion, or, of the understanding through analogical judg- ments ; but let alone, it. is just as intelligible as red,, or sweet, or hard, or as causation itself, which, for some in- scrutable reason, seems to be thought by many to be so per- fectly translucent a notion as to be the only proper solvent for everything else. If we could get over the futile feeling that everything must be like something else, a habit of mind confirmed by physical inquiry, we should have no more theoretical, than we have practical difficulty with lib- erty, claiming hourly its fullest consequences from child and adult, from friend and foe. § 8. What are the proofs of the existence of the power of choice as now defined ? Our analysis, our rejection of this and that explanation as insufficient, have proceeded on the claims of an intuitive notion. We have denied the con- clusions which seemed incompatible with perfect freedom, which furtively subjected the mind once more to the same forces which drive the world. Our proof of the existence of this power is not found directly in consciousness. If it were, the question would hardly admit of dispute. The evidence of consciousness is negative rather than positive. We are conscious of the presence of motives, we are con- scious of volitions, these are phenomenal ; we are not con- scious of the connection between the two, this is not phe- nomenal. We are negatively, indeed, aware of no restraint ; 412 PRIMARY VOLITION OR CHOICE. our volitions seem to be, what we affirm they are, free. But consciousness does not directly settle this question, for the sufficient reason that freedom is not a phenomenon, but the condition or form of phenomena, and hence it does not immediately arise in consciousness, but is only sug- gested from what is there present. The occasion of this suggestion is found in our moral nature. Laws are constantly imposed on our actions by ourselves, by others, and our moral sense justifies them. The record of history and of individual, life everywhere presents them, and hourly, momentarily demands them. Now no law, no command can be imposed on a being that is not free. The only law to which such a being can be subjected is a physical law, working in and through it. A moral law above it, before it, is an absurdity ; and, if fol- lowed by punishment is most cruel. Hence those who logically accept the consequences of their own doctrines utterly subvert the phenomena under consideration, the facts of the moral world, and give an en- tirely new rendering of them as a consequence of their denial of liberty. Says Bain, in The Emotions and The Will; "Under a certain motive, as hunger, I act in a cer- tain way, taking the food that is before me, going where I shall be fed, or performing some other preliminary condi- tion. The sequence is simple and clear when so expressed ; bring in the idea of freedom, and there is instantly a chaos, imbroglio, or jumble. What is to be said, therefore, is that this idea ought never to have come into the theoretical ex- planation of the will, and ought now to be summarily ex- pelled/' Again, " the word choice is one of the modes of designating the supposed liberty of voluntary actions. The real meaning, that is to say, the only real fact that can be pointed at in correspondence with it, is the acting out one of several diiferent promptings. When a person purchases PER VERSION OF FACTS. 413 an article out of several submitted to view, the recommend- ations of that one are said to be greater than of the rest, and nothing more needs really be said in describing the transaction. It may happen for a moment the opposing at- tractions are exactly balanced, and decision suspended there- by. The equipoise may even continue for a length of time, but when the decision is actually come to, the fact and the meaning are that some consideration has arisen to the mind, giving a superior energy of motive to the side that has preponderated. This is the whole substance of the act of choosing. The designation, liberty of choice, has no real meaning, except as denying extraneous interference." In the same vein, he continues, " The term responsibility, is a figurative expression of the kind called by writers on rheto- ric, ' metonymy,' where a thing is named by some of its causes, effects, or adjuncts, as when the crown is put for royalty, the mitre for episcopacy, etc. Seeing that in every country where forms of justice have been established a criminal is allowed to answer the charge made against him, before he is punished, this circumstance has been taken up, and used to designate punishment. We shall find it con- duce to clearness to put aside the figure, and employ the literal term. Instead, therefore, of responsibility, I shall substitute punishability ; for a man can never be said to be responsible, if you are not prepared to punish him, when he can not satisfactorily answer the charges made against him." In another passage, he gives concisely his notion of the method of moral suasion. " There is one form of stating the fact of ability that brings us face to face with the great metaphysical puzzle. It not uncommonly hap- pens that a delinquent pleads his moral weakness in justifi- cation of his offence. The school-boy whose animal spirits carry him to a breach of decorum, or whose anger has made him do violence on a school-fellow, will sometimes defend 414 PRIMARY VOLITION OR CHOICE. himself by saying lie was carried away, and could not re- strain himself. In other words, he makes out a case closely allied to physical compulsion. He is sometimes answered by saying, that he could have restrained himself if he had chosen, willed, or sufficiently wished to do so. Such an answer is really a puzzle or paradox, and must mean some- thing very different from what is apparently expressed. The fact is, that the offender was in a state of mind such that his conduct followed according to the uniformity of his being, and if the same antecedents were exactly re- peated, the same consequent would certainly be reproduced. In that view, therefore, the foregoing answer is irrevelant, not to say nonsensical. The proper form, and the practical meaning to be conveyed, is this : It is true, that as your feelings then stood, your conduct resulted as it did ; but I am now to deal with you in such a way, that when the situ- ation recurs, new feelings and motives will be present, suf- ficient, I hope, to issue differently. I now punish you, or threaten you, or admonish you in order that an antecedent motive may enter into your mind, as a counter-action to your animal spirits or temper on another occasion, seeing that, acting as you did, you were plainly in want of such a motive. I am determined that your conduct shall be re- formed, and therefore every time that you make such a lapse, I will supply more and more incentives in favor of what is your duty." Here is consistency. Mr. Bain has determined that there is no freedom, that the notion is an absurd one, and hence he pushes his theory right over the convictions of men expressed in the most unmistakable, universal and con- stant use of language. He says to himself, the line of my road lies through yonder hill, and he buries his engine up to the furnace in the soil in his effort to drive it through. As we have undertaken only the easier and more modest PER VERSION OF FACTS. 41 5 task of explaining, instead of overthrowing, the universal facts of mind, we must needs believe that the world, wise and ignorant, have not whistled to the wind in talking about freedom, choice, responsibility, and in constructing the frame-work of private, social and religious life upon them. In the above theory, there is the entire transformation of the very familiar facts of hourly life that seek our explana- tion. The language we apply to them is found to be all wrong. There is no proper guilt or punishment, virtue or reward. There is no law, as we use the word in social and ethical discussions ; all is ultimately resolvable into physi- cal force. The man indeed, like the brute, can be reached on two sides instead of one. He can be pushed, guided from behind, and, through the mirror of the mind, can be invited, influenced by things yet before him. As, by in- genious reflection, rays that do not directly fall upon the object are thrown upon it, so forces not yet realized are flung by anticipation on the mind, and become present pow- ers working vital results in the brain. To lay a command, therefore, as conscience does, and furnish, for its execution, no forces, promise no pleasures, threaten no pains, as the immediate results of obedience or disobedience, is, according to the above view, absurd, is to furnish the plan of a noble edifice, and provide no workmen to put it up. There is, on this theory, no other moral law than when I flourish a whip in the face of a restive ox, or apply it to his tough hide. The actions are essentially one ; the first brings the anticipation of pain, and the second, actual pain. In the passages quoted, there is an undersigned confession that the author can make nothing of the true moral phenome- na of moral law, and has, therefore, put in their place a gross caricature, at war with the form and language of our daily life. We do treat the brute and the man very differently, and the more diversely, as we are the more intelligent. 416 PRIMARY VOLITION OR CHOICE. We furnish an influence, an incentive for the one, we claim its existence in the other. We provide for obedience here, we demand it there ; we give the sharp intonation here, we simply state the law in its imperial power there. We accept as complete the service which fear has brought here ; there we despise it, as no solution of the claims of right on the soul. Bain gives the theory of brute life, we are striving to give that of rational life. He takes his ex- amples promiscuously from the causal and the free side of human action. While recognizing the way in which the two laws blend with each other in conduct, we assert the radical distinction between them. If a true, moral com- mand is ever uttered from within, or from without, right- fully to man, liberty, the power to obey it, is implied therein. The second portion of the proof of liberty, without which the first would be incomplete, is the fact, that the mind does spontaneously, inevitably place this notion of liberty back of human, responsible action as its explanation. Our conclusion is the conclusion of the race ; just as cer- tainly, universally, inevitably as is any judgment whatever. We no more necessarily refer an effect to a cause, than we do responsibility to liberty ; and responsibility we univers- ally claim of others. It remains to be shown that any man has ever lived, who has not believed in the guilt of his neighbor ; it is axiomatic in practical morals, that guilt is commensurate with power. Every excuse and apology pre- suppose it. The full form, then, of the proof of the exist- ence of freedom is found in the double fact, that we uni- versally lay moral claims upon others, and that we justify ourselves in so doing by attributing liberty to them. There is a large class of familiar and undeniable facts which the mind pertinaciously explains by an assertion of its power of choice. The difficulty of philosophers in analyzing it, their LIBERTY. 417 perplexities over it, their escape by denials of it, are no more proofs against it, than the like treatment of the mind's action in a dozen other directions. The spontaneous, ever- recurring action of the mind is the proof we have of liberty, and the only proof we have for anything our faculties offer us. We see and see again, till we believe that we see. We think and think again, till we accept our thought. § 9. Liberty, as a primary power, calls for a broader basis in our constitution than we have thus far assigned it, or than is usually seen to belong to it. If one act only, that of choice, is spontaneous, then this act can accomplish noth- ing, enclosed as a single term in the on-going processes of the mind. A volition must have at its disposal spontaneous powers to execute its purposes, powers that can be rescued from the current of previous causes, and be opened up to a new impulse. If the thoughts and feelings are ever mov- ing forward under sufficient and independent forces or causes, then neither can the antecedent conditions of voli- tion be secured, nor the subsequent ones. Antecedently volition would fail, because each state of thought and feel- ing being fixed by forces alien to the will, there would be no opportunity for independent inquiry on any topic ; the will would be found in the midst of intellectual and emo- tional conditions alien to every other than the established drift of thought. The volition would fail subsequently, because an act of choice is nothing unless previous tenden- cies can be checked by it and new ones established ; unless spontaneous powers can be brought forward for its execur tion. One may stand in a mill, its busy wheels revolving everywhere about him, and choose this or that result ;. the choice does not avail unless he can put his hand on. the machinery and modify its action. So far as this is complete within itself, and driven by its own forces, it goes forward heedless of volitions. If one is to oversee men,. his ovei> 418 PRIMARY VOLITION OR CHOICE. sight avails not if lie finds them all occupied in their own ways, and is unable to divert them from their labors, and convert them to his purposes. In other words, an act of will must go deeper and extend farther than itself and lay hold of a whole circle of modifiable powers in the execu- tion of its plans. Unless the thoughts and feelings are spontaneous, unless they are potential, and hold themselves at the beck of the will, in whole or in part, the will remains impotent, having no service because it has no servants. Spontaneity, then, must belong to our intellectual constitu- tion or freedom cannot belong to our voluntary life. Orig- ination at single detached points avails nothing. The will can not be operative if out of harmony with our other pow- ers. It must be able to act by anticipation, to ^accumulate and modify motives ; it must be able to rally to its choices the forces of the mind, and carry protractedly forward its processes ; it must be profoundly in sympathy with the en- tire constitution, and pervade its powers as a life-giving law. Dr. Carpenter, unwilling to abandon liberty, though it is a notion quite out of keeping with his general system, regards volition "as exerted in augmenting the nervous tension of the part of the cortical substance of the cerebrum which is concerned in the formation of the idea of the thing to be done. • ■ • It consists in an intensification of the hypersemic state of the ideational centre." (Mental Physi- ology, p. 425). This voluntary increase of the blood in the brain is open philosophically to all and more than all the difficulties of complete, proportionate freedom, and is most impotent in accomplishing its purposes. If one could at will increase or diminish the steam in the cylinder, and so modify the motive power in any branch of manufacture, this fact would not avail to alter the processes in progress. It would quicken or retard the movement, but not re-direct it. This very narrow volition would be so enveloped in SPONTANEITY. 419 previous states, that it would have no real control. An act of will, to be efficacious, must so pervade the mind as to give new starting-points and new directions. All pure intellectual action is spontaneous, beyond caus- ation, and ready to be played upon by the will. The feel- ings, intellectual and spiritual, are also spontaneous, that is, referable to powers of mind and not to physical forces. These, however, stand in such dependence on the thoughts that they are only indirectly reached by the will. The ac- tions of the mind, though free, first through its spontaneous powers, and second through its choices, none the less stand in determinate constitutional relations, and remain to be opera- ted under their appropriate laws. Thought does not cease to be spontaneous because it is logical, truthful; because the premises contain the conclusion. ]STo more do the feel- ings lose their spontaneity because they are united to the thoughts, and follow after them. It is not the nature of the will to set aside all relations, it is its office to work under them. There is order that is not causal order, to wit, thought- order, emotional order. Will means the power to guide and propel the mind, and this power is no more lost because the mind moves in definite ways, than is the control of the engi- neer because the engine requires a track. The mind works for the truth, not against it, or at random in reference to it. If, then, we deny liberty, our denial will reach propor- tionately deep, and the spontaneity of our intellectual ac- tion will go with it. Our thoughts, as a coherent move- ment of powers, will be dissolved apart. The conclusions will no longer inhere in the premises, and be taken thence by the mind, but impression will follow impression without direct connection, as the shadow of one car pursues that of another in the same train. Intellectual relations thus be- come the relations of shadows ; dependence is found only between things, is exclusively physical and causal. 420 PRIMARY VOLITION OR CHOICE. In strict consistency, however, we cannot hold fast to causation, since the link of force, like the link of liberty, is one supplied wholly by the mind, is one of its notions ; and no verification falls to a single idea which is not common to them all. Nothing can be affirmed of causation as proof which can not also be asserted of liberty. By the disappearance, therefore, of liberty, not only are all the social, civil, moral, and religious facts of the world dissolved, the coherance of thought also disappears, every connection is illusory, and every joint dislocated. We are in a world of films and shadows. Our moral actions first shake off responsibility ; then our thoughts slip their logi- cal connections and pass into shadows ; last of all the visi- ble world becomes a dream. We dream that we dream, and there is no awaking. To come to ourselves, is the very preg- nant phrase of all right-mindedness. When we possess in con- fidence our own powers, then we comprehend other things. Will is germinant, the only germinant thing in the universe ; all else is flow. Is it marvellous, then, that liberty as an idea must be the norm of constructive thought ? Liberty is spontaneity exercised in choice. Spontaneity is self-centered power as opposed to transmitted power. On spontaneity rests the potential — -what may be as opposed to what must be. We cannot go as far as Martineau and others desire to, and extend this dependence of force on will into an identi- fication of them, making an alleged consciousness of force exercised in volition the source of the notion of causation. The subphenomenal power as truly escapes us in mental experiences as in physical ones. The mind is alike pene- trative and constructive in both. What we should learn to do is not to distrust and struggle against our intuitions, but to call them forth and guard them. We shall everywhere lose the substratum of real being, unless we can penetrate to it by special insight. PH00F8 OF LIBERTY. 421 § 10. As the proofs of liberty are of the utmost mo- ment, and as they arise from a variety of scattered consid- erations as well as from central facts, we will for their better apprehension, epitomize and group them. (1) Pure mental facts, more particularly those of thought, must lose their quality without spontaneity. The mind, in its search for truth, must be free to follow it. If secret causal forces control the movement, the distinction between the true and the false is lost ; since every effect is equally necessary and real, the one termed false as the one termed true. If this argument is well taken, it is final ; for an overthrow of this distinction abolishes all denials and affirmations, and among them those of the necessitarian. (2) The moral facts of the world can not maintain their ground as simply causal facts. The distinction as a primary one disappears ; all facts are of one grand order, all are necessary facts. Any division that may remain is one of form simply. The discussion has swept away the phenomena that provoked it. (3) Physical facts, as united by forces and understood through forces, have no basis of knowledge firmer than that which has just given away in morals. The notion of force is one supplied by the mind, and applied for the same rea- sons and in the same way as that of liberty, to wit, to ex- plain certain facts. If the one notion is without proof, so is the other ; if the one has proof, so has the other. If we substitute succession for causation, succession is more allied to liberty than to necessity. And if we strip away depen- dencies, thought must go with them, as thought is one of them, and each instant involves them. If there are no con- nections there are no thoughts, and nothing for thought to pursue. Either the links of thoughts or the links of things must be allowed ; and if one is allowed both must be. (4) The two dependencies involved in necessity and 422 PRIMARY VOLITION OR CHOICE. liberty must remain as the rational correlatives of each other in all knowledge. Necessary movement is not a movement to be known by itself, since such a knowledge would be a superfluous adjunct, a waste of energy. Knowl- edge, on the other hand, being present, must have an ele- ment of freedom, or the simply necessary will remain alien and inapproachable to it. It is along the line of interaction between the fixed and the flexible that rational life is de- veloped. (5.) Language, the most complete and permanent image of mental facts, includes fully the notion of liberty. We refer not merely to the words of moral action and of in- tellectual and political freedom, — though these forms of liberty are so wrought into speech, directly and indirectly, that the most careful expression can not shun them, that a consistent utterance of the doctrine of necessity becomes impossible — but to the presence of such moods as the poten- tial and the subjunctive and the imperative, and the con- stant overshadowing of the actual by the possible and the ideal. This perpetual tenor of speech is illusory, if the world is covered in all its parts by one inflexible movement. Language again, like life, lies in the interplay of the possi- ble and the actual. (6) If pure mental action is not spontaneous, it is the product of realized forces ; no distinct seats of these forces can be given without involving materialism. Motives lie before the mind, and the objects expressed by them may not as yet even exist. Motives are images of possibilities, they are not forces in the mind moving it. The only imaginable seat of forces, efficient in a present process of thought or volition, must be brain tissue. But ner- vous forces working nervous effects bring no explana- tion of thoughts, unless these effects and thoughts are identical. Forces must be physical and must occasion PROOF OF LIBERTY. 423 physical effects, and these alone are what the theory of necessity expounds. (7) Nor does physical investigation disclose any term in the circle of forces which can be regarded as a thought- force. Though we can not trace exactly the circuit of the forces that play into and through the brain, we regard them as complete within themselves on a purely physical basis ; there is no opening, no transition stage, at which to insert a new force, a thought-force. The forces active in the tel- egraph stand on terms of complete equivalence among themselves ; the mind of the operator is not included in them. No more is thought embraced as a recognized part of the nervous circle. We have chemical, thermal, electric nervous forces, but no known thought-force. Not the first fact of this order appears in physical inquiry. Yet will ex- presses itself in forces of great energy, and if it does this by virtue of itself being a force, the fact should be a palpa- ble one. Quite the reverse seems to be the case ; we have a complete circuit of physical forces, with no hint of any deficiency. (8) Nor do motives combine like forces. Forces are not lost in the effects, whether they give apparent form to them or not. Adverse motives are lost. Single lines of ac- tion are taken. Mind does not move along the combining diagonal. (9) If the theory of necessity were correct there could be no successful resistance to it. It is entirely simple, and if it were true, there would be an end. It is its inad- equacy which stirs the mind to perpetual resistance. (10) Nor would the necessitarian, if his view were a sound one, be led so often into fatal concessions. There is no midway ground, we must accept necessity or liberty. Few have the boldness to take necessity, and to consistently defend it. Most, like Dr. Carpenter, are tempted just at 424 PRIM ART VOLITION OR CHOICE. the end to claim something, or to concede something which undoes all their work. (11) The palpable evasions of the necessitarian lead to the same conclusion. Thus Bain explains the action of the school-boy by transferring attention to the teacher, and al- lowing on his part the introduction of new motives. But this involves liberty. If the boy is in the chain of existing motives, the teacher is also, and the committee also, and the community also, and the generation as well. We can not by going backward find a point of modification. The ac- tion of the boy may not be altered, for the motives are not alterable. (12) If there is liberty, and only if there is liberty, can there be a beginning. Causation knows no begiiming, no change of direction, no possibilities. Simple facts without ultimate reasons, inflexible and inexplicable movement, are all that remain to us. If we seem to see reasons in this on-going, something addressed to a rational comprehen- sion, the vision is a mirage of the mind itself. New facts are all that we have. Our vessel sails under bare poles, we know not whither, before an invariable wind. Inquiry is futile, superfluous, a mere accident of the motion itself. The first five of the reasons now offered spring from the very nature of mind. These are followed by three others which show that the doctrine of necessity can gain no .-clear- ness of conception without embracing materialism. Three more arguments spring from the weak way in which the doctrine is centered in itself, and the last argument turns on the relation of liberty to ontology. CHAPTEE Y. Dynamic's of the Will, and of the Mind. § 1. The will is so nearly single, that little is to be said of the form of its activity. There is but one line of exer- tion, executive volitions resting back on a choice. This endures as a permanent impulse, and finds execution in mingled voluntary and vital action. From what has been said, it is evident that animals are destitute of all proper power of choice. Their action is the unconscious or the in- voluntary resolution of physical states and feelings into muscular impulse. The feelings which arise in conscious- ness are as directly connected with action as those physical states which never there present themselves, but, in the darkness and concealment of a purely vital force, accomplish their purpose. The will is strengthened chiefly by use, and that not alone by its own activity, but even more by the restraint and check thus imposed on the appetites and passions. These, allowed control for any length of time, assume so domineering and persistent a form, that the will regains only with the utmost difficulty the ground that it has lost. This anarchy of the soul is, of all forms of confusion, the least susceptible of a remedy, as aid cannot come from abroad, and the chronic weakness of the powers that offer resistance to the mob of impulses, and establish authority over them, speedily passes beyond all cure. Some sudden shock of the moral nature, in rare cases the awakening of a strong desire, is the only spring of hope. In speaking of the activities of the mind as a whole, we 426 DYNAMICS OF THE WILL AND OF THE MIND. are to remember, that these bear by no means the same pro- portion to each other in different individuals. Not only are specific, intellectual endowments and feelings diverse in power, the three classes of activities present various degrees of development. In one, intellectual effort absorbs the mind; in another, the emotions are the chief seats of ac- tion ; while a third is possessed of a will that lapses into stubbornness, through the inefficiency of the thoughts in its guidance. Moreover, different temperaments cause essen- tially the same faculties to exhibit very different degrees of force. The nutritive and the nervous systems are most intimately associated with the mind. Great impressibility and power in the nervous organization ; a preponderance of the nutritive functions giving a full animal life ; nervous power well-balanced and well-sustained by the nutritive sys- tem, constitute the nervous, phlegmatic and sanguine tem- peraments, which greatly modify the measure, hopefulness, and satisfaction of intellectual efforts, even when the natur- al endowments of mind are nearly the same. As the body is at once the medium by which impressions reach the mind, the source whence the strength for their considera- tion is secured, and the instrument by which its practical and theoretical conclusions concerning them are expressed, the importance of the physical conditions of mental activity cannot easily be over-stated, nor be too carefully inquired into. These researches, however, pertain chiefly to physi- ology. It is our task to trace the strictly mental interplay of the faculties, a dependence, not primarily the result of physical connections. § 2. Thought, feeling and volition, express the order in which action ordinarily occurs, the line along which any influence brought to bear on the mind passes through its faculties. Yet these three steps, though usual, are not all neces- RELATIONS OF MENTAL ACTION. 427 sary. Through sensation, feeling may be directly occa- sioned, and activity immediately follow from it, yet this is of an involuntary character. Thought, also, unites feeling and volition, points out the present relation of things, and guides the mind in the right use of means. While the first movement is in the direction now indicated, there are reflex influences of an opposite character. The feelings afl'ect strongly the thoughts. They direct attention to pleasing objects, fasten the faculties upon them, and thus intensify the emotions already established. The candor and fairness of the judgment are lost through this influence of the feel- ings, withdrawing attention from facts displeasing to them, and minutely and laboriously searching out those which maintain and justify their action. Unusual intellectual and moral development is required on the part of one possessed of strong feelings to reach even ordinary impartiality, and to give any considerable weight to reasons for action op- posed to the inclinations. The intellect thus becomes the instrument of the feelings, using all its acuteness, its power of presentation and argument in behalf of conclusions already reached by the heart. When the intellect is thus the sagacious counsellor or the cunning attorney of the emotions, the distortions of truth are proportioned to its strength, and the most powerful thinking is productive only of misleading sophisms. The feelings in the same way frequently engage the will, and the man becomes headstrong and willful in the line of action indicated by them. There is no defence against this but that quick moral sense, which responds with an adequate alternative to the selfish suggestions of the mind, and introduces a calm consideration of the claims of duty in each case. The only sufficient resistance to this domination which strong feelings are sure to assume over the intellect and the will, through the one evoking all the 428 DYNAMICS OF THE WILL AND OF THE MIND. imagery which influences passion, and the reasons which justify it, and through the other imparting a haste and momentum to action which at once clear the way of all ordinary obstacles, and render the onset easy and retreat difficult, is afforded by the moral nature, calming feeling, soliciting candor, and holding the will in the leash of duty. Government in the mind is not then self -evolved, is not the spontaneous inter-action of forces graded to their tasks, but is found in the direct, authoritative claims of a law-giv- ing power. The order of the mind is moral, not natural ; one of command and obedience, and not of self -poised pow- ers ; one to be discerned and pursued, not one to be devel- oped simply. The disorder of transgression discovers it- self, not in faculties lost, not in addition to, or subtraction from the original powers of the mind, but in that dispro- portionate development among them which is the fruit of anarchy, of usurpation on the one side, and overthrow on the other. § 3. If we look at the influences at work on any one character at any one time to make it what it is, there seems to be in it very little power of resistance or modification. The thoughts take such partial and justifying views of ac- tion, so blind themselves to the future results and even the immediate consequences of conduct, so misrender and mis- interpret facts ; the feelings so reward and maintain indul- gence, cast such disfavor and so repulsive an atmosphere over every form of restraint, choke up the path of reform with so many imaginary difficulties, and find the accustomed way so open, so easy, so inevitable ; the will submits so easily where it is wont to submit, is so reluctant to open a new conflict, and so weak to resist the impatient, persistent and domineering passions and appetites, that swarm in troops around it at every suggestion of change, that much modification of character established in its springs and MODIFICATION. 429 conditions of action seems to us impossible. Indeed only the breaking in of an earthquake-power is able to alter and redirect the channels in which the activities of the soul are flowing. If, however, we look at long periods, we see that there is a supreme control of the will over character. Single changes that in the outset are alien to the general move- ment prepare the way for others. ]S"ew thoughts give rise to new feelings, and these slowly displace the old ; the ac- tivity induced in fresh pursuits establishes and strengthens the will, and sets gradually at work varied reflex forces, giving different external and internal conditions, new feel- ings, motives and rewards of effort. At length, the mind accepts spontaneously the changed form of life, and a com- plete transformation is achieved. There is a momentum in mind which prevents its movements from becoming way- ward and fitful, and yet there is present a force which can slowly and certainly bend them in any direction it chooses. " There are a thousand microscopic motives," says Mai- lock, "too small for us to be actively conscious of, which according to how they settle on us, will really decide the question."* This is true only when we leave ourselves to the drift of life. Nothing is provided for, circumstan- ces accumulate, push toward one issue, and make every other issue increasingly difficult. A firm and comprehen- sive purpose anticipates such results, evading them before they are reached. If to such a purpose we add active, broad and candid judgment, pouring into the mind new light, new incentives, and the suggestions of new methods, we have the conditions of a character ripening in strength and in self -guidance. The key of the movement lies in apt and growing knowledge, bringing to the mind fresh terms of activity, and putting a sober will constantly at the helm. * Is Life Worth Living, p. 259. 430 DYNAMICS OF THE WILL AND OF THE MIND. § 4. The feelings are plainly most central and impor- tant in the constitution of the mind. Here is the seat of enjoyment, of all good. Thence spring the motives which influence the will, which offer its alternatives, and thither return the fruits of choice — fresh gratifications with accom- panying incentives to effort. The intellect is scarcely less instrumental to the emotions. It multiplies its resources that these may be nourished ; it fills its canvas with figures that these may be profoundly moved. The emotions are sooner or later endowed with all the treasures of thought, and the painter, the poet, all who can accomplish this trans- fer most quickly, skillfully, perfectly, become the chief artists in human society. The merchant, the inventor, labor for grosser forms of transmutation, the artist for higher, the true hero, for the highest. He alone lifts thought into the moral sublimity of an actual life, that is integral with the triumph of order — the ample victory of the law of freedom in the universe of God. As the impulses to action spring from the feelings, and the fruits of action return to them, it is evident that happi- ness must depend on the predominant emotions. Out of the heart are the issues of life. The physical feelings, the appetites, are primitive sources of pleasure ; yet they are necessarily intermittent, and can be made safely to occupy but a small part of the time which falls to us. Moreover, their permanent enjoyment depends on physical vigor, and this must be maintained by that temperance, by that well- regulated activity which sets these enjoyments still further limits. It is only on the condition of making the appetites secondary, incidental sources of good, that they can at all maintain their position as safe and just means of pleasure. The intellectual feelings are, indeed, capable of incessant activity, yet fail of conferring a sufficient and permanent good. There is not that repose in them, that perfect reac- SPIRITUAL FEELINGS. 431 tion of gratification on the appetitive desire which arrests it in complete indulgence. These intellectual impulses become rather increasingly exorbitant in their claims, fling us ever forward in search of the unattained, and leave us restless and unsatisfied with every acquisition actually se- cured. If we check the desire, we are immediately thrown back on other sources of good ; it fails any longer to main- tain our active powers and call forth our hopes. We must once again put to ourselves the question. What are those grounds, those sources of independent pleasure, of which at length, with the means in our hands that wealth, power, and rank confer, we are to avail ourselves in reaching com- plete and permanent happiness ? If, on the other hand, we steadily inflame and expand the desires, we are fed on prom- ises never realized, we are driven from one round of activity to another. We spread a feast, but have no time to partake of it, or, beginning to partake, are disappointed in its qual- ity. The good is not in it we thought to be there, and we are driven to the hopeless expedient of still farther enlarg- ing our board, enriching our service, and multiplying our viands. It thus not unfrequently happens, that the appe- tites decrease in the ratio in which the means of their grati- fication increase, and, at length, under this ever-returning experience, we discover that desires are wearing us out with unrequited labor; that the coin is indeed paid into the hand, but that it has lost its purchasing power; that we have served for Rachel, and that Leah has been given us. The spiritual feelings, on the contrary, yield adequate and supreme pleasure for several reasons. The higher in- tuitions call forth emotions which are of a primitive and permanent character; unlike the appetites, they may ac- company our every action with subdued pleasure, or with the swell of buoyant emotion. They may give way to out- side, incidental enjoyments, and yet return to us as the un- 432 PRIMARY VOLITION OR CHOICE. dertone of a steady and protracted harmony. Moreover, there is repose in them. The mind is filled with the satis- faction which truth, beauty and virtue afford ; without stim- ulating an excessive activity, they momentarily reward it. It is not merely a good in advance, but one in possession, that gives to the contemplation of beauty, of physical and moral excellence, a supreme and abiding pleasure. The concurrent reward and stimulus of the faculties take from them the intense thirst of desire, the restless, insatiate long- ing of intellectual emotions, expanding the circle each in- stant, and finding it forever made up of the same futile pleasures in greater multiplicity. Again, the rational gratifications increase in scope and in purity of tone. They arise from intellections which, with the growth of mind, become broader, more varied and more just. They yield, therefore, to the esthetical and moral sense more extended, harmonious, and profound im- pressions. ]STo one exhausts art, no one measures the re- sources of virtue, nor makes barren to the contemplation those plans and that providence which are working the world up, with all its stubborn and refractory materials, into a perfect and permanent product of religious art. These pleasures owe their high character also to the ex- tent in which they combine and blend all the activities of the triple powers of man. The intellect is most active in preparing the conditions, in giving the grounds of esthetical and ethical intuitions, while the intuitions combine insepar- ably perception and emotion. To see the true, the beautiful and the good is to feel their power. Nor is the will inac- tive. Under the surface of the mind, fully occupied with these noblest objects of contemplation, there flows a steady purpose to conform all action to them, never to mar them, to win them by becoming a part of them. Here is doubt- less the secret of the repose, the rest of truth, art and virtue. PLA T AND LABOR. 433 that they remove all conflict from our powers, and blend them in satisfied and indivisible activity. This is not as- serted of art as divorced from virtue, but of art as the highest embodiment of rational life, of virtue. The chief difference between play and labor seems to be that the one gives vent to a superabundant power and life in a direction in which it spontaneously flows, and the other demands, in view of a reward, exertion to which the physi- cal or intellectual state does not prompt. Labor approaches play in its character in proportion as the effort becomes spontaneous. Now success stimulates the feelings, and the quickened feelings arouse the active powers in the direction of their gratification. Hence it happens, that those whose labor is abundantly rewarded often take so keen a delight in it, as scarcely to be willing to turn aside for so-called play. The true amelioration of labor is success, the success which expresses power and enhances it, making realization easy and sportive. Drudgery is not so much labor as poorly requited labor. Hence labor that is undertaken under the prompting of strong and gratified desire, is much more easily endured than the same exertion when coerced. Self- directed and prosperous labor will, in proportion as these elements of liberty and power enter into it, assume the character of play, and the ultimate lifting up of the burden of toil will be found in a more spontaneous and successful movement, that is, in one more thoroughly spiritual. Ex- actly in the degree in which higher power is present and prevalent, do we already see the servitude of labor removed. Beauty and virtue must assume this easy, irrepressible char- acter which belongs to the physical putting forth of ani- mal life, before they can lay aside the harsh aspect of toil and struggle, and present the beauty of angelic strength — strength that is no more burdened by the load laid upon it, than the hero of the ring with his own muscle. CHAPTER YL The Relation of the System here offered to the Prevalent Forms of Philosophy. § 1. The inquiries of broadest outside and inside inter- est as regards any system of philosophy are: How does it unite the intellectual to the physical world ? Which, if either, does it absorb in the other ? What are with it the pregnant, ontological principles ? The philosophy now offered strives to maintain, so far as man is concerned, physical and mental phenomena on an independent basis ; so far as God is concerned, it centres and absorbs them both in Him. It thus endeavors to explain the familiar facts of experience, not as a vision, delusive in its form ; but as the substantial, sufficient frame-work of knowledge. It does not by thought abolish that which called forth thought, but retains entire the phenomena it seeks to expound. Indeed, it is difficult to see on what ground the reasoning of a few is to be accepted, while it overturns fundamentally the con- clusions of the many concerning facts of which each is in- dependently cognizant. To yield our faith to such theories seems to be a surrender of the trustworthiness of our facul- ties, since with almost perfect unanimity and endless reit- eration they have reached results exactly opposite to those thus offered. Nor is it an answer to this statement to say, that such a submission of the philosophical to the common mind precludes progress. It does not preclude the addi- tion of new facts, a more careful analysis of old facts, with the correction of opinions that is sure to follow. It does cast suspicion on a movement, that it pre-supposes the TRUSTWORTHINESS OF OUR FACULTIES. 435 entire error and deeeptiveness of all spontaneous convic- tions, denies the validity of every conclusion but its own, and will not go to the common mind for the facts even that seek statement and explanation ; for the facts without which there could be no philosophy. Such theories shake centrally the structure of knowledge, and lead to a complete distrust of those faculties which have been so signally, so universally, so completely wrong in directions wholly open to their action. To make one, two, three mistakes, and re- tain confidence is possible, to affirm that everything thought hitherto has been a mistake, is to reflect the most gloomy uncertainty on our present conclusions, which have no other verification than that they are the last results of faculties hitherto always at fault. That a system of philosophy lies, in the main, in the line of recognized conclusions, gathering up, harmonizing and expounding them, furnishes the same evidence of its truth as that afforded to a physical theory by the fact, that it easily includes and explains the facts under discussion ; or to a social theory by the fact, that it recognizes and makes clear events of hourly occurrence. Nor is it suf- ficient to give alleged reasons why men have been mis- taken ; universal and complete mistake is an impeachment of the mind whose consequences cannot be evaded. We postulate in the system now presented, the trust- worthiness of all our faculties in their careful, corrected, legitimate exercise; and accept as proof of a faculty or power, steady, reiterative action in any direction, yielding fruits of knowledge. What we see and hear, we accept as seen and heard, because our faculties are self -consistent and persistent in the affirmation. They renew the impressions in the same form on each like occasion. For a like reason we accept the conclusions of judgment. If we reached a different result each time we reviewed the proof of a propo- 436 THE RELATION OF FORMS OF PHILOSOPHY. sition, we should trust no one of our conclusions. We be- lieve what we believe, because the mind, on repeated in- quiry, arrives again and again at the same convictions. Thus is it with memory. We are uncertain when we find inconsistent and changeable impressions ; we are certain when the faculty restores the same image on each occasion. We start with no a priori theory as to what faculties the mind can have. We recognize as a fact that it does do what it seems to do, and take as a sufficient and ultimate proof of its power to impart and impart correctly any knowledge, the observed fact, that it does do this repeatedly and consistently. We cannot, therefore, accept the exist- ence of the notion of causation, and recognize the constant use which the mind makes of it, and at the same time affirm it to be illusory. The admitted fact establishes a power of mind to discern and employ this notion, and is thus a suffi- cient proof of the correctness of such a notion. We should as soon say, the mind insists that it sees, but the vision is fanciful ; as to say, the mind persists in assigning causes, but it has no ground for such assignment. The simple fact, that it does persistently assign them, is all the proof we are resting on in any department of knowledge. We postulate, then, the assertions, that the mind does what it does by virtue of a power of doing it, and that the habitual conclusions of a power are sufficient evidence, and the only possible evidence of its existence and their own truth. If the mind supplies ideas, in a uniform way, ideas which the senses alone can not reach, then this fact is satis- factory proof, that these ideas rest back on a distinct fac- ulty, and are sufficiently verified by that faculty. The phi- losophy here presented bridges the chasm between mind and matter, not by direct sensation, but indirectly, by in- tuitive ideas, whose presence gives occasion to the discus- sion, and makes it intelligible to us. In pronouncing so KNOWLEDGE OF PHYSICAL WORLD. 437 authoritatively, as some do, that matter is cut off hopelessly from mind, that there can be no communication between them, they seem to contradict their own statement ; since the mind is dealing with matter in the very affirmation by which it declares matter to be unapproachable. It is not, then, with the idea of matter, that the mind finds difficulty. This it works with in all its theories, and discovers nothing in it self-destructive, or destructive to the notion of mind. Whether, however, this idea, so manageable within the mind, has any outward thing that corresponds with it, is a question of simple proof, and, if such proof be present, yields no new perplexity. If the mind can in thought handle things so unlike itself as natural objects, it can also recognize their actual being on sufficient evidence. But it is said, there can be no such evidence, for such evidence implies not an ideal, but an actual influence of matter on mind. Is there, then, a clear a priori impossibility, that there should be found in the phenomena of mind such traces of the influences of matter as to furnish the grounds for an inference of its existence ? To the ordinary mind this question presents not the least difficulty. To it, sensa- tions, perceptions, are plainly such traces. But, says one who has longer contemplated the problem, is not space the condition of all material being, and is not this the one form which has no actual relevance to acts of mind ? Is not con- sciousness the essential characteristic of thought, and does not this in turn exclude altogether physical forces ? How then shall a material force strike within consciousness, or how shall a mental activity leave it to appear in space? Here undoubtedly our powers of explanation are at fault. The inquiries put us lie too deep in the secret nature, the unphenomenal nature of things to admit of that phenomenal statement or explanation which is sought for. Indeed, in the very language in which our queries are urged, we have 438 THE RELATION OF FORMS OF PHILOSOPHY. over-leaped the limits of clear thought. In speaking of a mental activity as leaving consciousness, or a physical force as entering it, we have subjected to the conditions of space that which is wholly foreign thereto. Yet these embarrass- ments should be no ground of disquiet, since, sooner or later, whatever path we take, we reach the unphenomenal, and thus the ultimate, The how of pure thought is as un- intelligible as the how of pure matter, and the inter-depen- dence of the two is no more obscure than the manner of the existence of either. The nature of thought is as unknown to us as anything can be. We discover easily the relations of things that lie in its light, but what that light is in which they are seen, what is the sub-phenomenal nature of the ac- tivity whose product we retain as a judgment, is wholly in- scrutable in the sense of being capable of any other phenom- enal rendering than that through which we actually know it. When we reach the bounds of events, we also reach the limits of a certain form of explanation* Yet we cannot deny the forms of existence that lie beyond, since such a denial is itself the source of greater perplexities than those it seeks to escape. Moreover, that space is not directly or indi- rectly penetrable by the activities of mind, is a proposition whose conditions are too obscure to suffer it to be ranked as an a priori conception. Were it not for our belief in the actual existence of the external world, and our connection with it, there would be no problem, since ideally the mind deals freely with space. If matter did not exist, if powers to apprehend it did not belong to us, there would be noth- ing to call forth the question which perplexes us. The very query itself thus becomes proof of the fact. We are not alone in an inability to solve ultimate prob- lems, pertaining to matter beyond the bounds of experience. Indeed an experience that should commence with a com- plete knowing, that should even know how it knew, would SPACE AND CONSCIOUSNESS. 439 be an eye that saw itself, an ear that heard itself. Mind is not such an organ. It reveals thought, not the nature of the thinking powers ; its phenomenal, formal character, not the very essence of the act itself. All that we claim is, that there is no a priori impossibility discoverable by us, making a transfer of influence from mind to matter, from matter to mind, an absurdity, Our last traces of physical force in the movement inward are found in the brain, our first traces in the movement outward are also met with at the same point. Thus far only can the eye follow material changes ; here is it first able to pick them up. How the last nervous impulse is linked to the play of consciousness, or how a pure volition breaks forth and liberates a physical force without ' itself becoming such a force, we cannot ex- plain. We only affirm that our ignorance is so complete as to cut us off as perfectly from a denial of the possibility of such a transfer, as from an exposition of it. We simply do not see that the realms of space and consciousness any- where over-lap, or even touch each other. We are pro- foundly ignorant of the nature of any connection between the two. We therefore satisfy ourselves with denying the existence of any a priori proof against such a dependence ; while experience, under the spontaneous interpretation which the human mind everywhere gives it, constantly affirms it as a fact. In the ideal world, the mind freely contemplates physi- cal being and forces. It moves at liberty among them, re- gards them as modified by its own activity, and is, in turn, modified in its thinking by them. It thus far recognizes no incompatibility between the two realms ; but is prepared to accept those actual relations which give occasion to these ideal ones. If an a priori necessity, ingrained in mind, divided the two fields, how could the mind so easily escape it in its own spontaneous movements ? It does not, can not 440 THE RELATION OF FORMS OF PHILOSOPHY. regard lines as at once parallel and intersecting; a relation of space as equivalent to one of time ; how, then, can it practically accept the communicability of matter and mind, and theoretically pronounce it impossible ? § 2. An increasingly prevalent form of philosopny, held crudely by some and subtily by others, is materialism. In its most logical, yet most naked and repulsive forms, it re- solves all thought into the mere action of nervous centres, induced in a purely physical way by physical forces. This system is for the most part the product of scientific inquiry, a study of the laws of the material world simply. While affecting great contempt for a priori systems, and claiming experience as the only source and test of truth, in its phil- osophy — by courtesy so called — it presents an example of the most unreasonable a priori method anywhere found in the progress of knowledge. The entire organum, the scheme of inquiry and instru- ments of thought, with which it approaches the intellectual world, have been gathered in departments utterly alien to the one to be contemplated. Far from being ready to ac- cept new facts under their own laws, philosophers of this school approach the science of mind, with the antecedent conviction, that physical laws reign everywhere, that there is the same fixed dependence of events in the realm of thought as that which they have found in matter. They thus, with the blindness of a limited system, push up the stream of causes as far as they can go, and then deny there is anything new beyond. As this theory fails, not merely to explain, but even to accept, the new and diverse phe- nomena of consciousness ; and feebly substitutes for them some connected, but very different phenomena, to wit, those of the nervous centres, we feel at liberty, giving its scien- tific inquiries due praise, to pass it very lightly as a philoso- phy. It deals with the external conditions and accompani- MATERIALISM. 441 ments of mental activity, and not with the inner forms and laws of those activities. Under that fatal certainty which causes equivalent errors to follow each other at opposite ex- tremes, it strives to stand outside in space and expound con- sciousness, as formerly the hasty philosopher inclosed in consciousness constructed his outside facts. The last gate which this school suppose themselves to have opened, at which the powers of the physical world are to rush in and submerge those of mind, is that known as the correlation of forces. All material forces are converti- ble and indestructible. Hence it is concluded, that those which are at play in the living nervous organism mutually replace each other, and evolve from within themselves all the most subtile and the most palpable of the activities of rational life. Accept this relation of forces in the body, and we yet need the independent, spontaneous power of mind. The mind avails itself of a stream of forces that flow incessantly through the physical organization; into this it dips its wheel, and with it works out its purposes. This admission by no means closes the argument. We have here a telegraph, we discover that the electric, chemical, thermal, mechanical forces liberated are so far equivalent as to induce us to believe that they are perfectly so. We stand in an office ; we behold an intelligible cypher rapidly appearing on the ribbon before us ; does the equivalence, the indestructibility, the convertibility of the forces in the mechanism we have investigated, explain the message we have received? We may say, that nothing has been lost or added to the sum of forces concerned in the transfer of these words. "Very well, the words in their intelligibility still seek solution. These are explained by the constant interference of a higher power, a remote operator, above the circle of self-balanced forces which have transferred the motion from the indicating to the inscribing index. 442 THE RELATION OF FORMS OF PHILOSOPHY. I may never see the hand that plays the remote key, but I cannot fail to believe in its existence, nor in the independ- ent, intelligent character of the force that presides there. I know not how the key is touched by w T hich the self-poised, nervous forces of the brain are set in motion ; but in the product wrought out, I do see unmistakably the evidence of such initiation, guidance, arrest. The continuity and equality of the forces in the nervous circuit, if fully estab- lished, do not weaken or embarrass the conviction. They simply leave us where they found us, ignorant of the way in which the mind employs the current of material forces ; these still yield the clearest evidence of being at some point of their circuit intersected by another and higher circle of influences. To say that the only force which can modify physical forces must itself be a physical force, betraying its presence among them as a new, additive power, is not merely to affirm what we do not know, but is to make the assertion that the intelligence of the products momentarily evolved by these nervous centres does not indicate a like quality in the ultimate agency, an assertion in flat contradic- tion of the principles of reasoning on which we habitually proceed. How little this form of philosophy can accomplish is evident from the fact, that it itself must admit, that some kinds of matter are intelligent, self-conscious, spontaneous, and others are not. Thus having laboriously swallowed up mind in matter, it is compelled to re-include under matter, distinctions in every way as perplexing and inscrutable as those displaced. The facts remain, and either matter is self-conscious, or that which is self-conscious is mind. Words, rather than ideas, are thus offered as explanations in this deceptive resolution of two distinct elements into one. If an adversary of this theory chooses to add the farther affirmation that this self-conscious matter is also MATERIALISM. 443 free, the point can only be fairly settled by re-opening the entire discussion ; for it is antecedently no more improbable that matter is free, than it is that matter is conscious, and intelligent. The forces concerned in intellectual action are either conditioned from within to all the facts of mind, and we are remitted to consciousness to determine what these facts are in their entire complement; or these forces are conditioned to their action from without. If we accept the first statement, we have recognized two kinds of forces or activities utterly distinct from each other ; if we accept the last, we have used two words, and called one set of forces appearing in space, material; and the same forces arising in consciousness, mental ; thus overlooking the distinctions between them. What possible explanation is there in this % Do not the fundamental differences between matter and mind, open to all our faculties, remain as before \ It would be well for philosophers to remember that theories can not reflexively wipe out facts, and that those of mind are of the most primitive and undeniable order. If either of the two classes of facts is to be merged in the other, physical ones necessarily yield to those of mind, as in their nature second- ary, being known only as they affect consciousness. As the material world is at best reached inferentially, it can not logically displace the very faculties that know it. The knowing must have precedence of the thing known. If either is to be found to contain the other, it must be the first the second, not the second the first. Materialism does not always assume the crude form now controverted. It has sometimes a more mixed and subtle character, one in which it is partially blended with ideal- ism. Mr. Mill, while deriving all knowledge from experi- ence, and declining to recognize any intuitive elements, nevertheless leaves the existence of matter in doubt. Sen- sations and perceptions are accepted in an ideal form, and 444 THE RELATION OF FORMS OF PHILOSOPHY. the outside world of realities, which lies back of them, is left unapproached. Such a system is beset with more diffi- culties than either materialism or idealism. Sensations, whose existence and influence lie wholly within the mind, can with less reason be made to control and give form to the mind, than matter conceived as wholly outside and independent of the intellectual powers. Indeed it is not easy to see how a perception can occupy this anomalous position, on the one side giving law to the mind, on the other, cut off from all known, exterior dependence, and rest- ing back on the very faculties whose form it controls. This system exhibits the same defective analysis which belongs to all materialism. Space and time are evolved from experience, though they are the conditions of experi- ence. They are made to spring from sensations, though themselves utterly beyond sensation. Those ideas on the other hand, that are admittedly in the mind, yet admittedly beyond experience, are pronounced delusive. Of this char- acter is that of causation. Breaking this cord of connec- tion, the external world swings loose from this philosophy. There lies against it concisely these difficulties. Claiming experience to be the source of knowledge, it elaborates a system far removed from ordinary conviction, and subver- sive of many of its most cherished opinions. It knows nothing of matter, while mankind knows this chiefly. It gives sensations, perceptions control over the mind, while the opinions of men divide control between outside and in- side conditions. While employing it makes illusive, in its intellectual basis, the notion of causation, which above all has universal sway in the practical world. It denies moreover necessity to any ideas whatever, while the whole history of pure mathematics shows the contrary. It is compelled to refer to experience the recog- nition of such facts as this, that straight lines, parallel DIRECT PERCEPTION. 445 through a portion of their extent, are so through their whole extent. Its analyses are inadequate, and it rejects without reason the ideas for which it can find no place in its system. That is to say, it makes its method the test of the facts, and not the facts the test of its method. There are two forms or tendencies in materialism ; the one is found in the identification of the phenomena of matter and mind, and the other in the identification of the laws of mat- ter and mind. The last is the method of Mill and Spencer, and is not less destructive of the facts than the first. § 3. The next system of which we shall speak, is a mixed one, that presented by Hamilton. Its most striking feature is, that it makes matter itself the direct object of perception, and thus, losing one occasion for intuitive ideas, accepts a part of them, perverts a part, and neglects a part. Among those resolved into powerlessness, are causation, lib- erty and the infinite. We need only to speak of its central characteristic, the direct perception of matter. Against this there holds, we believe, the very generally accepted axiom that nothing can act save where it is. The introduction of the adverb where, shows this statement to be limited to physical forces, since these alone appear in space, alone have locality. Physical forces must be where they are ex- ercised. This will hardly be denied by any one. For a force to show itself as a force where it is not, would be for it to be and not to be at the same point at the same time. Mind, thought, have no reference to space, and hence it conveys no very intelligible idea to say, that the mind must be, a thought must be where it acts. Their objects of con- sideration may come from any quarter, and any distance ; conclusions may strike out into the most remote regions, and such words as come and go, near and distant, have only a figurative signification. Now perception, till the brain is passed, is a thing of physical forces, and each organ and 446 THE RELATION OF FORMS OF PHILOSOPHY. nerve can only be affected by that within it, not by that without it. It is against the above axiom to say, that I feel the stone, meaning thereby that the sensation is outside the organ — conversant, less or more, with the very essence of being in the stone. The organ is affected by what is with- in itself ; till the contour of its own states is penetrated, the object might as well be miles, as inches, or fractions of an inch distant. Physical effects lie as content in each organ of sense, and are as localized within it as is the object with- out it. If, then, these physical changes which accompany perception, sensation, were perceived by the mind, the very object, the source of these, would not thereby be directly known. But these states are not perceived, we know noth- ing about the states either of the eye, the ear, or tongue, in seeing, hearing, tasting. Sensations enter consciousness, and lose at once their special organic grounds. When we have reached the last physical change in our nervous organ- ism, we have not reached the first thing that the mind is conscious of in sensation. No organ of sensation is revealed by its own sensations, but by other sensations of other organs of which it is made an object. If, then, the mind knows the object in perception, it is not by the movement inward from the object, since this finds change, when from it, as a cause, there arises a nervous affection in an organ of sense ; and this again meets with a second inexplicable change, when there is a transfer to consciousness, and the true con- tent of the mind lies within it, divested of physical quali- ties. We might as well say, that the first ball is in the second ball moving after concussion, as to say, that the very object of perception, or any portion of it, is in this its latest effect. No, the second and third ball move through a change within themselves ; the organ becomes a condition of perception — itself distinct — through a distinct condition of its own nervous substance. IDEALISM. 447 It is, then, by an outward movement of the mind, that matter is known, and this is not perception but inference, the interpretation of sensations. Through the notions of ex- istence and causation and space, the mind establishes the ex- ternal world. Sensations, till interpreted and expounded by judgment, are the crudest possible conditions of knowledge. If it be said, that the act of perception itself is the result of an outward, not an inward movement, that it takes place at the exterior tip of the nerve, and not as the consequence of physical effects traced to the nerve centres ; we say, that the mind must either pass perceptively beyond the sensa- tional organ, or the perceptive act is still within the human body, and thus removed from the object perceived. More- over such a theory neglects the obvious ministration to perception of all the chain of nervous influences passing in- ward. If these are means to sensation, they must intervene between the presence of the object and the perception of it. The inscrutable transition from a nervous state to a feeling is the last result of the inner current. Of an outer ner- vous current, there is no proof. The mind has no further perceptive connection with the object than that in which these several conditions of the organ intervene between the mind and the object. The last nervous condition remain- ing, the sensation remains. § 4. The last system of which we shall speak, is ideal- ism. Idealism has peculiar excellences and defects. It seizes the most fundamental element of the universe, and evolves all else with consistent logic from it. It does not humble mind under the laws of matter, but makes. it the source and law of all things. As all known existences must in some way enter consciousness, or be productive of phe- nomena there, it is evident that idealism has no occasion to lose or overlook any part of knowledge. Neglecting that inferential action of the mind by which it recognizes the 448 THE RELATION OF FORMS OF PHILOSOPHY. objective validity and relations of the various sources of its perceptions and sensations, idealism is able, by limiting the attention to the phenomena of the internal world alone, to trace the inherent relations among these, and develop a purely ideal system of purely ideal objects. Herein there is opportunity for great subtlety, profoundity, consistency, and even breadth of thought ; since everything, outer and inner, finds representation here. If the images of all the objects and events of the external world were brought to the eye of a spectator on a transparency, it is plain that he might form a very inclusive, and, in some of its aspects, cor- rect philosophy concerning them. Consciousness is such a screen, and the philosopher, confining his attention to this, may evolve a very harmonious system. Idealism more signally than most other theories fails of being a science, a knowing as actual that which is con- ceived of as theoretical. It matters little, that the inherent connections are necessary, unless the premises from point to point of the argument are verified as real. The difficulties of idealism are much the same as those of the a priori proof of the existence of God. An ideal conclusion is evolved from ideal premises, but as the last do not take hold of the world of facts, no more does the first. Philoso- phy is not merely philosophy, but science as well. It pos- sesses inductive, united with deductive, elements. It re- sembles mixed rather than pure mathematics. It does not start with definitions of ideal objects, but with facts. Idealism, on the other hand, while contemplating thought, contemplates it as thought merely, in its formal relations rather than in its actual, phenomenal character and force. It deduces the individual from the general. It inquires not so much what is given actually and practically with inde- pendent testimony by the several faculties of mind; as what can be evolved from the mere fact of thought. IDEALISM. 44:9 The result is, that no system is as far removed from gen- eral belief and faith as idealism. None so signally fails to recognize and expound the phenomena of mind under the form they actually assume in experience. It seems rather a field of intellectual gymnastics than of sound, sober in- quiry concerning things, corrected and guided each instant by an observation of facts. Idealism starts with assuming the least possible. It would commence with nothing if it could. It accepts only activity known in consciouness. It must not even say " an action," lest there should thus be implied something which is active. From this it proceeds to develop matter and mind, activity and divided activity; recognizing itself inconsci- ousness by opposing to the naked knowing the conscious- ness of knowing. Thus it moves onward, spinning a world out of its own bowels, and with little more of actual corres- pondence of results to the notions of men than there exists between the threads of a spider's web and the actual forces which hold the world together. Yet the idealist relishes his own system none the less for being so stuffed with the ego. Scientific philosophy does not inquire how little it may assume, but how much it may consistently accept ; at how many points it has reached ultimate facts. If the idealist is at liberty to regard the connections of thought not as fanciful and chimerical, but, as they seem to be, logical and coherent ; in short to accept thinking as a valid and reliable act ; if he is at liberty to assume memory, these necessary assumptions involve the fitness of still farther assumption.. Are not these, portions of a set of powers, and if the phi- losopher avails himself of two, can he do better than to* avail himself of all ? Does he trespass any more on sound; principles in using the entire group than in using a part ?: Indeed, does he not act absurdly in employing thus adroitly a part, and neglecting the remainder, equally fitted, for 450 THE RELATION OF FORMS OF PHILOSOPHY. another and specific purpose ? Should it be one's object to see how much can be done with the least possible means, or to see how much can be accomplished with all available means ? Having a clue ought he not as a thinker to follow it as far as it will carry him, and does it not carry him logi- cally to a faith in all his faculties, since he must have a faith in a part of them ? Possibly, one can hop a little dis- tance painfully on one foot ; is it, therefore, wise to sling up the other? The scientific philosopher at once sets to work to determine by observation and analysis all his facul- ties, and accepts the testimony of them all, as each necessary to the right understanding of the peculiar and independent facts rendered by it. Thus the idealist and the less cunning but more wise inquirer begin at once to diverge. The one constructs a system of remarkable connections, subtile and sagacious, but altogether airy and unsubstantial ; the other acquires classi- fied knowledge, with many lines of causation and deductive relations in it ; often presenting, indeed, inscrutable points, yet always having the ring and firmness of facts. Idealism is ideal ; science, the philosophy we seek, is actual. § 5. The system we have now presented aims fully to recognize the different, independent kinds of knowing. 'Each of these is ultimate, and, therefore, inexplicable under other forms of knowing. To carry one faculty into the •province of another, is to displace that other, and with it the information it is fitted to give. Knowledge, in its last -analysis, has always a certain mystery about it for the very reason that we can go no farther. There is a mystery in a color, as green ; in a taste, as sweet ; in an odor, as fragrant ; in a judgment, pronouncing the stone to be hard ; in every intuition, as that of a cause, of liberty, of the infinite. We must not expect to expel mystery, but to reduce it to a iminimum, and place it at the right points. LIBERTY AND CAUSATION. 451 One of the chief labors of the philosopher is to keep in- dependent faculties, so recognized on adequate grounds, from devouring each other; from making incursions into fields alien to them, from refusing to accept what has not been submitted to themselves, and received their peculiar seal. The imagination and the understanding belong es- pecially to these intrusive faculties, while the intuition of cause, having swept through the entire physical world, is ever bent on a raid into spiritual realms. To be ready to recognize, in their unrestricted forms, the facts of conscious- ness as revealed in the mind, in language, in history ; to analyze these cautiously, without bias and perversion, for the discovery of the simple activities or faculties they re- veal ; and afterward # to hold fast to every affirmation of these faculties, is the duty of the wise cultivator of mental science. "We have a love for science above that for phil- osophy, because of this inductive element it so obviously includes. The confusion which arises from undue emphasis laid upon a single faculty is seen in our knowledge of the thing- in-itself, of mind-in-itself . These, we are said, not to know, because we can not conceive them. To wish to conceive them is the futile desire to put a second circle of pheno- mena below the first circle. Such a movement could ad- mit of no pause. The noumena must either be resolved in- to phenomena, or always escape the imagination. But do we not know the noumena, things-in-themselves ? Do we not know the whole circle of effects which express them ? What more could we wish to know; unless forsooth, we we are dissatisfied with our circle of senses and wish a larger one ? To desire to know a thing in one way, when its very nature only allows us to know it in another way, is certainly an irrational impulse. The independent validity both of causation and of lib- 452 THE RELATION OF F0BM8 OF PHILOSOPHY. erty has been recognized. Each idea is present to the mind in the spontaneous explanation which it offers to a certain class of facts. They divide the universe of events between them. In the one moiety, we have necessity, in the other, liberty ; in the one, movements already conditioned by the forces at work, in the other, movements then and there conditioned by the power that initiates them. In their relations to each other, liberty is primary, and causation is secondary. Causation marks dependence, a dependence which, on its own level, can find no arrest, no matter how far we trace it. Events, follow them backward, forward, on either hand, are conditioned one upon another ; forces are already at work accomplishing the tasks assigned them. But a first, an independent, an unconditioned force nowhere appears. Causal action, therefore, necessarily presents a fragmentary and partial character. Of it alone, there can be made up no whole, no universe ; since the more we have, the more we demand to explain what we have. The events before us, like the section of a river, must flow into and flow out of the horizon. We can reach no beginning and no conclusion, nor even find diminution as we go backward, or increase as we go forward. The boundaries of our vision enlarge themselves in all directions, but are always illusory. Liberty, on the contrary, to the extent of the events which spring from it, affords a complete commencement. We need go no farther back. An arrest is found in it, and the events which flow thence are explained by the form, impetus, and direction which it has imparted to them. Causation is necessarily finite in its manifestation ; since it inheres in a power already put forth, and is conditioned to a given number and form of products. Liberty rests back on the agent, never goes forth from him, and partakes, in its possibilities, of the breadth and the limitations of his facul- ties. It commands more than the actual, to wit, the poten- LIBERTY AND CAUSATION 453 tial of being. Infinite power can inhere in a free personal- ity, and in no other form of existence. Causation is closely connected with space. It may be questioned whether it ever acts in any other connection- It inheres in forces, and these are put forth into separate, spacial existence. The phenomena of mind which involve cause and effect do so through material dependencies. The mind's own action would seem to be always either spontaneous or free, that is its spontaneity is revealed under the three forms of thought, feeling and volition. Liberty, in contra-distinction, remains always in connection with con- sciousness. We can only choose consciously. Matter can only be the source of causative action. Mind is the source of spontaneous and of free action ; of spontaneous action, that is action springing independently from it, though often evoked by conditions not supplied by it; of free action, that is action held within the conditions which are its occasion but not bound to any one of them. Liberty, again, lies back of all causation, because the whole flood of forces with which human liberty plays springs from the choice of God, is but the executive power with which he momentarily sustains and accomplishes his purposes. Here we reach another ultimate fact. We know not how the mind affects these secondary physical forces, that in the human body play beneath its touch. ]STo more do we understand how these imperishable and uniform forces on which the universe is buoyed, of which it is fash- ioned, go forth from the will of God. Their wholly finite? yet rational character compels the reason thus to refer them to an independent, self-sufficient and wise Source ; and therein to complete the conception of the universe in time as in space. A cord of great length is no more self-sup- porting, no more explicable in itself, than a shorter one. The only idea which is, as it were, spherical, self-centered, 45± THE RELATION OF FORMS OF PHILOSOPHY. demanding nothing, suffering nothing outside of itself, is that of an Infinite, Personal God, a sufficient source of all things ; whose spontaneity and liberty require no explana- tion, and bring explanation to all beside. On this ground, and on this alone, the reason accepts the idea, as one by which it does see — as a sun that does spread its light through the whole heavens, leaving nothing which is not sought out by its rays. The final proof of truth is the fact of light, the very fact of light admitting no controversy and no denial to those who receive it — to whom it gives the power to be- come the sons of God. The real efficiency of every word is found in the disclosure of itself as the light which comes down from heaven. § 6. While the division of schools of philosophy into materialistic, realistic and idealistic is simple and funda- mental, it does not express with any fullness and exactness the very variable facts before us. We offer, therefore, a second division more suggestive of this complexity.* Constructive Idealism. Idealism. Materialism. Constructive Materialism. We draw attention to the direct contrasts of the inner circle, of realism to nihilism, of materialism to idealism ; * Consult recent British Philosophy. SYSTEMS OF PHILOSOPHY. 455 and also to the relations of the outer circle. This circle turns, in its divisions, on the force given to intuitive ideas. By realism we mean what is sometimes termed natural realism. It is the realism of Reid and Hamilton. Each act of perception is said to include, as an indivisible constit- uent, the direct knowledge of both matter and mind. Ma- terialism is the identification of matter and mind as two forms of the same experience ; this is the method of M. Taine. Idealism is the absorption of all being into proces- ses of mind, after the manner of Fichte and Hegel. Ni- hilism is the denial of all knowledge, phenominal and con- structive. The impressions we call knowledge are merely impressions afloat with other impressions, reflections in a stream which may disappear at any moment. Constructive realism is realism reached as the result of the combined action of our perceptions, intuitions and judg- ments. It is the realism oliered in this work. Construc- tive materialism is the materialism of Spencer. It gathers the constructive laws of thought from the phenomena known as physical, and builds the universe, both physical and intellectual, by means of them. It affirms nothing as to ultimate, substantial being ; but phenominal being it puts exclusively under the laws of matter. Constructive ideal- ism declares that all the forms of knowledge are purely mental, and cannot be said to be forms of matter. Thus matter, when its existence is allowed, is not known under its own types. Substantially and phenominally it is hidden from us. Constructive idealism rests chiefly on the works of Kant. A measure of agnosticism is involved in opposite direc- tions in both constructive materialism and constructive ideal- ism. When these two incapacities are united, we have agnosticism ; and agnosticism may easily lapse into nihilism — the antithetical point to realism. Hume is the best expo- 456 THE RELATION OF FORMS OF PHILOSOPHY. nent of nihilism. The circuit on either hand through ideal- ism or materialism involves a decay of power, and* so leads on to nihilism. Pure Positive Philosophy is agnosticism. It directs attention wholly to phenomena. A ninth central school might be added, that of identifi- cation ; but it has not gained, and can not readily gain, any clear expression. Matter and mind are to be united in one concurrent line of evolution. This is the goal of Hegel. It has been our purpose to draw cardinal lines firmly, and this farther division is added simply as provocative of historical inquiry. INDEX A. Abstraction, 153. Actions as voluntary, 395 ; as conscious, 396. Affections, 311 ; natural affections, 320 ; affections, 350. Analysis, 153. Animals, their powers, 295 ; tricks, 298 ; training, 298 ; language, 300 ; manner of judgment, 300 ; character, 301 ; sagacity, 302 ; feelings, 361 ; actions, 392. Appetites, 317. Association, subconscious facts, 43 ; memory, 133, 138, 175 ; laws, 172, 174 ; beauty, 240 ; power of mind, 291 ; animals, 301 ; feelings, 366. Attention, what, 291 ; to more than one thing, 291. Automatic action, 385, 395. B. Bain, Prof., imagination, 143, 144, 147 ; judgment, 162 ; existence, 181 ; space, 196 ; cause and effect, 210, 213 ; right, 228, 233 ; feelings, 314 ; nervous system, 383, 386 ; liberty, 412, 424. Beauty as an idea, 240 ; feelings, 342. Bentham, J., morals, 239. Berkeley, Bishop, perception, 111. Brain, relations to mind, 56 ; functions, 383 ; superiority in man, 387, 394. Bridgman, Laura, 276. 0. Carpenter, W. B., unconscious cerebration, 50, 53 ; brain, 388 ; will, 418, 423. Cause and effect, an idea, 209 ; nature of, 210, 218 ; growth of mind, 281 ; relation to liberty, 420, 421, 451. Cerebration, 49, 55. Chance and liberty, 411. Choice, 31, 346. 458 INDEX. Classification, 153, 157. Conceivable imagination, 151 ; judgment, 170 ; infinite, 250. Conception, 153 ; conceptionalist, 158. Conditioned, law of, 214, 247. Conscience, 231, 239 ; custom, 232 ; feelings, 847. Consciousness, nature of, 20 ; truthfulness, 21 ; relation to intelligence, 69 ; to facts, 123 ; to judgments, 161 ; to existence, 181 ; as an idea, 201 ; to action, 396 ; to liberty, 411, 420. Correlation of forces, 442. D. Darwin, imagination, 145; gemmules, 212. Deduction, 283, 288. Descartes, perception, 108, 111. Desires, 323 ; classes, 325 ; relation to pleasure, 327 ; strength of, 328 dependent feelings, 329. Dynamics of intellect, 268 ; of feelings, 355 ; of will, 425. E. Equilibrist, subconscious states, 41. Esthetical feelings, 342. Ethical feelings, 346. Existence, an idea, 179. Extension, a primary quality, 116. External world, knowledge of, 112. F. Feelings, 306 ; distinctions, 307 ; divisions, 309 ; Physical Feelings, 312 ; general sensations, "313 ; special sensations, 315 ; appetites, 317 ; relation to pleasure, 319 ; natural affections, 320 ; Intellectual Feel- ings, 322 ; primary, 322 ; wit, humor, 325 ; desires, 323 ; secondary, 329 ; success, 329 ; failure, 332 ; Spiritual Feelings, 337 ; divisions, 338 ; nature, 338 ; truth, 339 ; beauty, 342 ; right, 346 ; obedience, 348 ; disobedience, 349 ; modifications, 850 ; diagram, 352 ; relation to action, 355 ; growth, 857 ; collective growth, 359 ; animal life, 361 ; relation to right, 362 ; laws, 363 ; mind as a whole, 426 ; pleasures, 430 ; restfulness, 431. Ferrier, D., 64. a. Generalization, 153. INDEX. 459 H. Habit, memory, 134. Hamilton, Sir W., phrenology, 29 ; desires, 30 ; consciousness, 33 ; sub- consciousness, 34, 38 ; mind always active, 73, 77 ; perception, 87, 94, 96, 100 ; qualities of matter, 110, 121 ; memory, 128; imagina- tion, 143 ; judgment, 160, 161, 165; association, 172; conscious- ness, 204 ; causation, 214 ; infinite, 245, 247 ; logic, 285 ; emotions, 355 ; subconsciousness, 393 ; his system, 445, 455. Hartley, association, 173. Hickok, L., 259. Hypnotism, 67. I. Idealism, 86 ; perception, 91 ; as a system, 447. Ideas, regulative, 177 ; how arise, 178 ; what, 179 ; criteria, 254 ; charac- ter, 256 ; relation to inheritance, 260 ; to instinct, 261 ; grouped, 263 ; are these all ? 263. Imagination, 142 ; theories, 143 ; influence, 147 ; strength, 151 ; incon- ceivable, 151 ; condition of feeling, 367. Induction, 283, 288'. Infinite, an idea, 243 ; in connection with space, 243 ; with time, 244 ; with God, 246 ; nature, 249. Intellect, divisions, 79 ; dynamics, 268. Intellection in morals, 227. Intelligence and consciousness, 46. Irritability, 321. J. Judgment, perception, 84, 102 ; a power; 153 ; processes included, 154 ; kinds, 164, 166, 169 ; inconceivable, 170 ; office, 171 ; association, 175. K. Kant, desires, 30 ; categories, 266 ; feelings, 306. Knowledge relative, 114. L. Language and philosophy, 24, 27 ; perception, 95 ; animals, 301 ; relation to thought, 396. Lewes, G. H., 45. 460 INDEX. Liberty, an idea, 221 ; what, 401 ; conditions, 402 ; strongest motive, 404; motives, 404 ; objections, 409 ; proof, 411 ; spontaneity, 417 ; proofs summarized, 421 ; causation, 451. Life, what, 372 ; forms, 375. Locke, 12 ; perception, 110 ; qualities of matter, 121. Logic, 285. M. Mallock, W. H., 429. Man contrasted with animal, 295. Martineau, J., 420. Materialism, 441. Matter, primary and secondary qualities, 116 ; existence, 438. Maudsley, Dr., 37 ; memory, 128. •Memory, subconscious states, 38 ; a power, 125 ; physical states, 127 ; theories, 128; association, 133, 138 ; relations, 134; qualities, 136; phases, 137 ; cultivation, 140. Mental science, limits, 19, 71 ; difficulties, 22 ; aids, 27 ; divisions, 30. Mesmerism, 67. Mill, J. S., 285 ; causation, 210 ; right, 233 ; logic, 285 ; materialism, 444, 445. Mind, relation to brain, 56 ; mastery, 59 ; constant activity, 72 ; growth, 269 ; power over itself, 290 ; relations to cerebrum, 396, 398 ; gov-. ernment, 428 ; nature, 430. Minimum, visible, 40. Mnemonics, 142. Moral nature, 224 ; sentiments, 346. Motives, 404 ; stern and weak, 405. Mfiller, imagination, 144. Murphy, J. J., intelligence and consciousness, 46, 204 ; consciousness, a feeling, 48. Mystery, 450. N. Nervous system, 371 ; office, 372 ; parts, 373 ; forms, 375 ; in man, 378 ; offices, 381 ; modifications, 385. Nominalism, 158. Number, an idea, 183. 0. Ontology, 453. Order of development, 268, 426. INDEX. 461 Perception, nature, 80 ; organs, 82 ; objects, 82 ; judgments, 84, 101, 104 ; statement of doctrine, 92 ; consciousness, 94 ; data, 97 ; not direct, 99 ; eye, 101, 103 ; ear, 104 ; importance, 107 ; history, 108 ; rela- tion to philosophy, 113, 445. Philosophy, advantages, 2 ; progress, 8 ; postulates, 13. Phrenology, 29. Plato, 158, 185. Play and labor, 433. Postulates, 13, 435. Power of mind, 145, 436. Quain's anatomy, 380. Qualities of matter, 116. Q. R. Realism, 158 ; as a system of philosophy, 435, 451, 453. Reason, 176 ; what it gives, 179, 397. Reasoning, 284 ; kinds, 286, 288. Reid, 13 ; perception, 87. Relation as an idea, 260. Relativity of knowledge, 114. Resemblance as an idea, 185 ; growth under, 279 ; science, 280. Right as an idea, 224. s. Sagacity in animals, 302. Science and resemblance, 2S0. Sensations, 80 ; growth, 269 ; feelings, 312. Sight, 101. Sleep, 67. Somnambulism, 67. Space as an idea, 188. Spencer, mind and brain, 63; judgment, 162; association, 173; space, 186, 191, 196, 198 ; time, 205 ; causation, 212 ; right, 238 ; infinite, 251 ; ideas objective, 257 ; inheritance, 260 ; dynamics, 268 ; feel- ings, 315 ; systems, 455. Spinal cord, 378, 381. 462 IJVBEX. Spontaneity, 221, 419. Subconsciousness, 34, 70. Synthesis, 153. Systems, 434 ; materialism, 441 ; Hamilton, 445 ; idealism, 447 ; realism, 450 ; arrangement, 453. Taine, M„ 384. Temperament, 426. Time as an idea, 204. Todd and Bowman, 382. Training- in animals, 299. Truth as an idea, 224 ; feelings, 339. IT. Unconscious cerebration, 49, 55. Understanding, 125. Utility, 226. V. Vertebrates, 387. Vision, monocular, 106 ; stereoscopic, 106. Vital action and volition, 392. Volition, 31 ; divisions, 391, 400 ; executive, 391 ; primary, 400. w. Whitney, W. D., animals, 286. Will, 369 ; control, 425. THE END PUBLICATIONS OF G. P. PUTNAM'S SONS. A History of American Literature. By Moses Coit Tyler, Pro- fessor of English Literature in the University of Michigan. Volumes I and II, comprising the period, 1607-1765. Large 8vo, about 700 pages, handsomely bound in cloth, extra, gilt top, $6.00 ; half calf, extra, $11 00 The History of American Literature, now offered to the public, is the first at- tempt ever made to give a systematic and critical account of the literary development of the American people. It is not a mere cyclopaedia of literature, or a series of de- tached biographical sketches accompanied by literary extracts : but an analytic and sustained narrative of our literary history from the earliest English settlement in America down to the present time. The work is the result of original and independent studies prosecuted by the author for the past ten years, and gives an altogether new analysis of American literary forces and results during nearly three centuries. The present two volumes— a complete work in themselves— cover the whole field of our history during the colonial time. " An important national work." — New York Tribune. " The literary event of the decade." '— Hartford Courant. " A book more interesting than half the new novels." — The Nation. II A work of great and permanent importance." — New York Evening Post. " One of the most valuable publications of the century."— Boston Post. " A book actually fascinating from beginning to end.'' — Prest. J. B. Angell. " As the work stands, it may rightfully claim a place on the library table of every cultivated American." — New York Times. "No work of similar scope and magnitude and erudition exists, or has been at- tempted in this country."— New York Evangelist. " A unique and valuable work " — Chicago Tribune. " A work which will rank with those of Sismondi, Ticknor, and Taine."— New York Evening Express. " It is this philosophical character of the work which brings it not far distant from the works of Taine, of Buckle, and of Lecky." — Buffalo Express. u One can hardly speak too strongly in* praise of these conscientious, careful and successful volumes, which deserve to be studied alike by scholars and patriots." — Rev. Henry Martyn Dexter, D.D. " But the plan of Professor Tyler's book is so vast and its execution so fearless, that no reader can expect or wish to agree with all its personal judgments. It is a book truly admirable, both in design and in general execution ; the learning is great, the treatment wise, the style fresh and vigorous. Here and there occurs a phrase which a severer revision would perhaps exclude, but all such criticisms are trivial in view of so signal a success. Like Parkman, Professor Tyler may almost be said to have created, not merely his volumes, but their theme. Like Parkman, at any rate, he has taken a whole department of human history, rescued it from oblivion, and made it hencefor- ward a matter of deep interest to every thinking mind." — T. W. Higginson, in The Nation. " The work betrays acute philosophical insight, a rare power of historical re- search, and a cultivated literary habit, which was perhaps no less essential than the two former conditions, to its successful accomplishment. The style of the author is marked by vigor, originality, comprehensiveness, and a curious instinct in the selection of words. In this latter respect, though not in the moulding of sentences, the reader may perhaps be reminded of the choice and fragrant vocabulary of Washington Irving, whose words alone often leave an exquisite odor like the perfume of sweet-briar and arbutus." — George Ripley, in the Tribune. u Professor Moses Coit Tyler's ' History of American Literature,' of which the first two volumes have just been issued, will take rank at once as a book of lasting value, even though the author should advance no further than he has already done in the scheme of his work. We are not unmindful of the eminent historians this country has produced, when we express our opinion that his history is the best study of Ameri- can historic material that has been written by an American. There has been manifestly no limit to the enthusiasm, conscientiousness and industry with which he has possessed himself of the entire body of the literature of which he treats, and at the same time he has displayed the qualities of a true literary artist in giving form, color and perspective to his work." — David Gray, in the Buffalo Courier, PUBLICATIONS OF G. P. PUTNAM'S SONS. VAN LAUN. The History of French Literature. By Henri Van LAUN, Translator of Taine's "History of English Literature," " the Works of Moliere," etc. Vol I. From its Origin to the Renaissance. 8vo, clotli extra . . . . . . . . . . . $2 5c Vol. II. From the Renaissance to the Close of the Reign of Louis XIV. 8vo, cloth extra $2 50 Vol. III. From the Reign of Louis XIV. to that of Napoleon III. 8vo, cloth extra $2 50 The set, three volumes, in box, half calf, $15.00 ; cloth extra, 7 50 " We have to deal with a people essentially spirited and intellectual, whose spirit and intellect have been invariably the wonder and admiration, if not the model and mold of contemporary thought, and whose literary triumphs remain to this day among the most notable landmarks of modern literature." * * * — Extract from Author's Preface. " Mr. Van Laun has not given us a mere critical study of the works he considers, but has done his best to bring their authors, their way of life, and the ways of those around them, before us in a living likeness." — London Daily News. " This history is extremely interesting in its exposition of the nterary progress of the age, in connection with the social and political influences which helped to mould the character and the destinies of the people."— Boston Daily Globe. " It is full of keenest interest for every person who knows or wishes to learn anything of French literature, or of French literary history or biography. Scarcely any book of recent origin, indeed, is better fitted than this to win general favor with all classes of persons." — N. Y. Evening Post. THIERS (Louis Adolphe) Life of. By Francois Le Goff, Docteur-es-lettres, Author of a " History of the Government of National Defense in the Provinces," etc. Translated, from the author's unpublished manuscript, by Theodore Stanton, A.M. Octavo, with Portrait, cloth extra. . . . . . $2 00 This book is written especially for the American public by M. Francois Le Goff, of Paris, a French publicist of the Conservative-Republican school, who knew Thiers personally and who is thoroughly conversant with the history and 'politics of France. Besides the biographical narrative, which is enlivened by many fresh anecdotes, the writer attempts to present such a connected view of French political history for the last fifty years, as will throw light upon the present crisis in France, so incomprehen> sible to most Americans. The work will also be interesting as an able defense of the unity of Thiers' political life, a position rarely assumed by even the most ardent friends of the great statesman. It is illustrated by a fac-simile of his handwriting and auto- graph, a view of his home, etc. BRIEF BIOGRAPHIES. First Series. Contemporary States- men of Europe. Edited by Thomas Wentworth Higginson. They are handsomely printed in square i6mo, and attractively bound in cloth extra. Price per vol. . . . . . . . $1 50 Vol. I. English Statesmen. By T. W. Higginson. " II. English Radical Leaders. By R. J. Hinton. " III. French Leaders. By Edward King. " IV. German Political Leaders. By Herbert Tuttle. These volumes are planned to meet the desire which exists for accurate and graphic information in regard to the leaders of political action in other countries. They will give portraitures of the men and analysis of their lives and work, that will be vivid and picturesque, as well as accurate and faithful, and that will combine the authority of careful historic narration with the interest attaching to anecdote and persona* delineation. 41 Compact and readable * * * leaves little to be desired."— N. Y. Nation. ^^f<