r LIBRARY UNIVERSITY OT- CALIrOKNIA SAN D1f?GO 1101 A SHORT HISTORICAL ENGLISH GRAMMAR SWEET .Eonbott HENRY FROWDE OXFORD UNIVERSITY PRESS WAREHOUSE AMEN CORNER, E.G. Qtw MACMILLAtf & CO., 112 FOURTH AVENUE SHORT HISTORICAL ENGLISH GRAMMAR HENRY SWEET, M.A., PH.D., LL.D. Formerly President of the Philological Society Editor of 'The Oldest English Texts,' Alfred's ^Cura Pastoralis' and ' Orosius ' Author of 'An Anglo-Saxon Reader' 'A first' and 'A Second Middle-English Primer' 'A Primer of Spoken English' 'A History of English Sounds' 'A New English Grammar' 'A Primer of Phonetics] 'Shelley's Nature-Poetry,' etc. AT THE CLARENDON PRESS 1892 jrforb PRINTED AT THE CLARENDON PRESS BY HORACE HART, PRINTER TO THE UNIVERSITY PREFACE FOR some years the want has been felt of a short historical English grammar up to date, especially as regards phonology, dialectology and chronology, the last implying careful dis- crimination between what is really in living use and what is obsolete. The present work is an attempt to supply this want. It is an abridgment of the historical portions of my New English Grammar. It does not include syntax, for the good reason that a grammar which attempted to deal even if only superficially with such a vast and difficult subject as historical English syntax could not possibly be designated as a ' short ' grammar. But within the limits of phonology and accidence, including composition and derivation, it will, I hope, be found to contain all that is really essential to the beginner. Some teachers will be disappointed at not finding here any exposition of that time-honoured generalization ' Grimm's Law,' and the still more popular ' Verner's Law.' I have for the present excluded them, because they do not belong to historical English grammar, but to comparative Arian philo- logy ; because, if studied adequately, they are too difficult for beginners; and because, without a detailed knowledge of b vi PREFACE. Sanskrit, &c., they are of little use for etymological purposes. But although most of those who have kept pace with the recent developments of Comparative Philology admit all this, some of them still plead for the retention of Grimm's Law on the ground of its being so interesting, and having such a stimulating effect on pupils. The answer to this is, By all means teach it then, but teach it as an extra, not as a part of English grammar, any more than you would include French, Latin, and Greek etymology in English grammar ; although, of course, English grammar undoubtedly leads up to all these subjects, and is more or less directly connected with them, in the same way as it is connected with the political, social, and literary history of England. The study of this grammar requires no preparation except a knowledge of the ordinary grammatical terms. It does not even postulate any practical knowledge of Old English, although I should advise every teacher of historical English grammar to let his pupils go through a preparatory course in Old English with the help of such a book as my Anglo- Saxon Primer. . Additional grammatical details and illustrations that may be required will easily be found in The New English Grammar and my History of English Sounds, in which latter will be found a concise statement not only of Grimm's and Verner's laws but also of all the other sound-laws by which English is connected with the older Arian languages. HENRY SWEET. SOUTH PARK, REIGATE, 7 Sept., 1892. CONTENTS PAGE HISTORY OF ENGLISH . i Periods i. Cognate Languages 2. Old English 3 Characteristics 12. Latin Influence 13. Celtic Influence 14. Scandinavian Influence 15. French Influence 17. Middle English 6 Dialects 21. French and English 24. Rise of the London Dialect 26. Scandinavian Influence 28. French Influence 29. Latin Influence 31. Modern English 1 1 Influence of other Languages 36. Periods 40. PHONOLOGY. PHONETICS. Analysis .......... 14 Breath and Voice 44. Nasal Sounds 45. Consonants 46. Vowels 47. Vowel-like Consonants 48. Synthesis .15 Quantity 50. Stress 51. Intonation 53. Glides 54. Syllables 56. Diphthongs 57. b 2 viii CONTENTS. PAGF. Vowels 17 Rounding 59. Tongue-retraction 60. Tongue-height 62. Vowels in Detail 63. Consonants . ... . .21 Form 81. Place 86. Rounding, Fronting 92. The Aspirate 94. R in English 97. LAWS OF SOUND-CHANGE . . . . . .25 OLD ENGLISH SOUNDS. Orthography 26 Pronunciation 27 Stress '..... 29 Quantity 31 Vowels . 31 Mutation 131. Consonant-influence 134. Consonants 35 Gradation 36 MIDDLE-ENGLISH SOUNDS. Orthography .......... 37 Stress 41 Quantity 42 Vowels ........... 43 Consonants 47 MODERN ENGLISH SOUND-CHANGES. Orthography .......... 53 Vowels 58 Consonants 63 PBESENT ENGLISH. Stress ........... 67 Quantity 70 CONTENTS. ix ACCIDENCE. PAGE NOUN'S. Old English 7* Gender 266. Cases 269. Early Middle English 76 Late Middle English ........ 80 Modern English 81 Irregular Plurals 315. Foreign Plurals 320. ADJECTIVES. Inflections . . . . . . . . . -87 Old English 328. Middle English 334. Modern English 338. Comparison . . 90 Old English 339. Middle English 340. Modern English 341. Irregular 344. PBONOTTNS 95 Personal 97 Old English 359. Middle English 362. Modern English 375- Possessive 107 Self 109 Demonstrative 112 one, a ; none, no 115 Interrogative and Relative . 117 Definite 119 Indefinite 119 Quantitative . . . . . . . . . .121 x CONTENTS. PAG It NUMERALS. Cardinal . .' . . . . . . . .123 Ordinal 126 VERBS. Old English 128 Inflections 473. Strong 486. Weak 498. Preterite- Present 507. Middle English 140 Early 510. Late 535. Modern English 149 Present English 579. Irregular Verbs in Modern English 159 CONSONANTAL VERBS 160 With vowel-change : pret. vowel e 586 ; ea 589 ; o 590 ; ou 591. With t instead of d 592. With t in- stead of d and vowel-change : pret. vowel e 597 ; o 602. With t instead of -ded 604. With consonant-loss 607. \Vith consonant-loss and vowel-change : pret. vowel 88 608 ; o 609. INVARIABLE VERBS 168 aa 616 ; ai 617 ; v 618 ; e 621 ; 89 625 ; i 627 ; o 633 ; u 634. VOCALIC VERBS i ;o pret. vowel au 635 ; u 637 ; as 646 ; e 651 ; ei 655 ; i 658 ; ij 661 ; o 662 ; ou 667 ; o 681 ; u 6^9; uw 692. MIXED VERBS . 183 ISOLATED FORMS 184 quoth 713 ; hight 714 ; iclept 715 ; wont 716. ANOMALOUS VERBS ... .... 185 can 718; dare 719; may 720; must 721 ; ought 722 ; shall 723 ; will 724 ; wot 725 ; need 726 ; be 727; have 730; do 731. CONTENTS. xi PAGE PARTICLES. Adverb-endings ......... 191 Adverbs from Nouns and Adjectives 194 Pronominal Adverbs . . . . . . . -197 Correlative Particles . 199 Pronominal Conjunctions 200 Negation and Affirmation ....... 201 Comparison of Adverbs .202 PREPOSITIONS 204 Old English 764. Middle and Modern 'English 770. INTERJECTIONS . . .207 COMPOSITION. Old English. 208 Modern English . 210 Meaning of Compounds . . . . . . .212 DERIVATION. Native Elements. PREFIXES 214 a- 803 ; seg- 805 ; be- 806 ; for- 807 ; ge- 808 ; mis- 809 ; of- 810 ; on- Sn ; to- 812 ; un- 813. SUFFIXES 220 Agouti-forming, (a) Concrete: -end 822 ; -ere 823 ; -estre 824; -ing 825 ; -ling 826; -en 827. (b) Abstract: -nis 828; -u 829; -u)>, -J> 830; -ung, -ing 831 - -dom 832; -had 833; -rseden 834; -scipe 835- Adjective-forming: -ede 836 ; -en 837 ; -ig 838 ; -isc 839 ; -sum 840 feald 841 ; -full 842 ; -leas 843 ; -lie 844 ; -weard 845. Verb-forming -na 846 -sian 847 -laecan 848. xii CONTENTS. PAGE Foreign Elements. PREFIXES 230 ab- 853; ad- 854; amb- 855; amphi- 856; an- 857 ; ana- 858 ; ante- 859 ; anti- 860 ; apo- 86 1 ; bi- 862 ; cata- 863 ; circmn- 864 ; cis- 865 ; com- 866 ; contra-, counter- 867 ; de- 868 ; demi- 869 ; di- 870; dia- 871; dis- 872; en- 873; endo- 874; epi- 875; ex- 876, 877; exo- 878; extra- 879; hyper- 880; hypo- 881 ; in- 882, 883 ; inter-, enter- 884 ; intro- 885 ; meta- 886 ; ne- 887 ; non- 888 ; ob- 889; para- 890; per- 891 ; post- 892 ; pre- 893 ; preter- 894 ; pro- 895, 896 ; pros- 897 ; re- 898 ; retro- 899 ; se- 900 ; semi- 901 : sine- 902 ; sub- 903 ; subter- 904 ; super- 905 ; supra- 906 ; sus- 907 ; syn- 908; trans- 909; ultra- 910. SUFFIXES 241 Noun-forming, (a) Personal: -ee 912; -ar, -e(e)r, -ier 913 ; -or 914 ; -ard, -art 916; -ess 917 ; -ist 918 ; -ite 919 ; -trix 920. (t>) Diminutive : -ule, -cule 921 ; -et, -let 922. (c) Abstract: -y, -ey 923; -ice, -ess, -ise 927 ; -cy. -sy 928 ; -ad, -id 929 ; -ade 931 ; -age 932; -ment 933; -ion 935; -ana 936; -nee 937 ; -ncy 938 ; -o(u)r 939 ; -ory 940 ; -ry 941 ; -ure 942 ; -ism 943 ; -icism 944 ; -ate 945 ; -itude 946 ; -ty 947. Adjective-forming: -ble 948; -bund, -bond 950; -ic 951 ; -ical 954 ; -iac 955 ; -id 956 ; -oid 957 ; -al, -ial 958; -il(e) 962; -an(e) 963; -can 966; -ian 967; -in(e) 968 ; -nt 9/0; -lent 971 ; -ar 972; -ary 973 ; -ior 974 : -ese 975 ; -ose, -ous 976 ; -esque 97 8 J -K e ) 979 J - ive 9 8 3- Verb-forming : -fy 984 ; -ish 985 ; -ize 986. SHORT HISTORICAL ENGLISH GRAMMAR. HISTORY OF ENGLISH. PERIODS. 1. The name ' English language ' in its widest sense com- prehends the language of the English people from their first settlement in Britain to the present time. For the sake of convenience we distinguish three main stages in the his- tory of the language, namely Old English (OE), Middle English (ME), and Modern English (MnE). OE may be defined as the period of full endings (mono., sunne, sunu, sidnas), ME as the period of levelled endings (mone, sunne, sune, stgnes), MnE as the period of lost endings (moon, sun, son, stones = stounz). We further distinguish periods of tran- sition between these main stages, each of which latter is further divided into an early and a late period. The dates of these periods are, roughly, as follows : Early Old English (E. of Alfred) Late Old English (E. of yElfric) Transition Old English (E. of Layamon) Early Middle English (E. of the Ancren Riwle) . Late Middle English (E. of Chaucer) . Transition Middle English (Caxton E.) Early Modern English (Tudor E. ; E. of Shakespere) Late Modern English 2 HISTOR V OF ENGLISH. [ 2. to which may be added Present English, by which we understand the English of the present time as spoken, written, and understood by educated people, that is, roughly speaking, ipth-century English. COGNATE LANGUAGES. 2. English belongs to the Arian family of languages, descended from a hypothetical Parent Arian language, the chief of which are given in the following table, different periods of their development being separated by dashes : (A) East- Arian, or Asiatic : (a) Sanskrit, the sacred language of India Pali Bengali and the other Gaurian languages of India. (6) Iranian languages: Zend or Old Bactrian. Old Persian, which is the language of the Cuneiform inscriptions Modern Persian. (c) Armenian, which is really half-way between East- and West- Arian. (B) West- Arian or European : (d) Greek Romaic or Modern Greek. (e) Latin the Komance languages: Italian, Proven9al, French (Old French, Modern French), Spanish, Portuguese, Roumanian. (/) Celtic languages. Gaulish. The Goidelic group : Irish, Manx, Gaelic. The Cymric group : Welsh, Cornish, Breton (introduced from Britain). (g) Slavonic languages. Old Bulgarian Russian, Polish, Bohemian, Servian, Bulgarian. (h) Baltic languages. Lithuanian, Lettish. (i) Germanic languages. 3. The Germanic group, to which English belongs, con- sists of the following languages : 8.] HISTOR Y OF ENGLISH. 3 (A) East- Germanic : (a) Gothic. (Z>) Scandinavian languages. West-Scandinavian group : Norwegian, Icelandic. East- Scandinavian group: Danish, Swedish. (B) West- Germanic : (f) Low German languages. Old Saxon Dutch, Flemish. Anglo- Frisian group : English, Frisian. (d) High German, or German. 4. English is then a member of the Anglo-Frisian group of the Low German languages. Old English. 5. In the fifth century or perhaps earlier Britain was partially conquered by a variety of Germanic tribes from the other side of the German Ocean, the chief of which were (a) Saxons, from the country between the Elbe and the Rhine. (3) Angles, from the district still called Angeln in the South of Schleswig. (c) Jutes, from the North of Schleswig. 6. The first settlement is said to have been that of the Jutes, who took Kent and the Isle of Wight. 7. The Saxons occupied the country south of the Thames ; except Cornwall, where the Britons still kept their nationality. Some of the Saxons settled in Sussex; some north of the Thames in Middlesex and Essex ; the remaining portion of the tribe being called 'West-Saxons/ whence their state is called ' Wessex.' 8. The rest of England was occupied by the Angles. Suffolk and Norfolk were included under the name of B 2 4 HISTOR Y OF ENGLISH. [ 9. ' East-Anglia.' Another tribe of Anglians occupied what are now the Midland Counties, between the Thames and the Humber. These were called Mercians, and their country is called ' Mercia.' The country north of the Humber was occupied by a variety of Anglian tribes included under the name of Northumbrians. Ancient Northumbria extended up to the Firth of Forth, and thus included the greater part of what is now the Lowlands of Scotland. 9. All these tribes spofce the same language with slight differences of dialect. These differences increased by degrees, so that already in the 8th century we can distinguish four main dialects : Northumbrian and Mercian, which together constitute the Anglian group; and West-Saxon and Kentish, which together constitute the Southern group. 10. All these tribes agreed in calling their common lan- guage English, that is, ' Anglish,' because the Angles were for a long time the dominant tribe. The supremacy after- wards passed to the West-Saxons, and their capital, Winchester, became the capital of England ; and West-Saxon became the official and, to a great extent, the literary language all over England. The West-Saxons still continued to call their language English, the name ' Anglo-Saxon ' being used only as a collective name for the people, not the language. 11. In this book OE words are always given unless the contrary is stated in their Early West-Saxon forms ; that is, in the dialect of King Alfred. CHARACTERISTICS OF OLD ENGLISH. 12. The characteristics of OE are those of the other Low German languages. It was, as compared with MnE, a highly inflected language, being in this respect intermediate between IS.] HISTORY OF ENGLISH. 5 Latin and Modern German. In its syntax it closely resembled Modern German. It also resembled Modern German in having an unlimited power of forming new words by deriva- tion and composition, as when it made Scribes and Pharisees into ' bookers and separation-saints ' (OE boceras and sundor- hdlgari). LATIN INFLUENCE. 13. Nevertheless it adopted many Latin words, some of which it brought with it from the Continent, such as strat ' high road,' ' street/ mil ' mile,' cdsere ' emperor ' from Latin (via) strata, mllia (passuum), Caesar; while others were learnt from the Romanized Britons, such as ceaster 'city,' laden 'language' from castra, (lingva) Latlna. These are all popular words. There is another layer of learned words which came in after the introduction of Christianity in 597. Such words are deofol ' devil,' mynsler ' monastery,' fers ' verse,' from diabolus, monasterium, versus. CELTIC INFLUENCE. 14. Very few Celtic words came into OE, because the Britons themselves were to a great extent Romanized, espe- cially the inhabitants of the cities, who were mainly the descendants of the Roman legionary soldiers, dry ' druid,' ' sorcerer ' is an example of a Celtic word in OE. SCANDINAVIAN INFLUENCE. 15. Towards the end of the 8th century Scandinavian pirates chiefly from Norway, but also from Denmark, all being indiscriminately called ' Danes ' by the Anglo-Saxons began to harass the coasts of England. By the end of the next century they had conquered and settled East-Anglia (in 6 HISTORY OF ENGLISH. [ 16. 870), Mercia (in 874), and Northumbria (in 876); although in the next century they were forced to acknowledge the supremacy of the West-Saxon kings. In 1016 the whole of England was conquered by the Danes, and England was ruled by Danish kings till 1042, when the Anglo-Saxon royal line was restored in the person of Edward the Confessor. 16. It is not till the close of the OE period that Scandi- navian words appear. Even Late Northumbrian (of about 970) is entirely free from Scandinavian influence. FRENCH INFLUENCE. 17. With the accession of Edward the Confessor in 1042 Norman influence begins ; and in 1066 the battle of Hastings made the Norman duke William king of England, although the actual conquest was not completed till 1071. 18. The Normans were Scandinavian by race, but their language was a dialect of Old French. 19. The influence of Norman French on OE was of course even slighter than that of Scandinavian, so that it does not become a factor of importance till the ME period. Nevertheless several French words passed into literary OE even before the Conquest, such as c astel ' castle,' capun ' fowl/ Middle English. 20. In its Middle period English went through much the same changes as the other Germanic languages, though at a quicker rate. Many of the sounds were changed, most of the old inflections were lost, their place being supplied by form-words prepositions, auxiliary verbs, etc. and many words became obsolete. 23] HISTORY OF ENGLISH. 7 DIALECTS OF MIDDLE ENGLISH. 21. The Norman Conquest, by depriving the old West- Saxon of its literary and political supremacy, gave free play to the development of the dialects. Although the ME dialects are continuations of the OE ones, it is convenient to call most of them by different names. The main divisions are Northern, corresponding to the Old Northumbrian, Mid- land, corresponding to the Old Mercian, Southern, corre- sponding to the old West-Saxon, and Kentish. We include the first two under the term ' North-Thames English/ the last two under ' South-Thames English.' 22. Of these dialects the Midland was the predominating one. Its commanding position in the heart of England enabled it to exercise a direct influence on all the other dialects, while Southern and Northern were completely cut off from one another. Hence even the earliest Southern of about 1 200 shows considerable influence of the Midland or Old Mercian dialect. 23. It is to be observed that the changes which distinguish one period of English from another went on much faster in the North of England than in the South. In fact, the Old Northumbrian dialect of the roth century had already entered on its transition period characterized by a general confusion in the use of inflections, and was thus almost on a level with the Early Southern Middle English of about 1200. Again, the Northern dialect in its Early Middle period had got rid of nearly all the inflections that are not preserved in MnE, being thus several centuries ahead of the South-Thames dialects. The Midland dialects were more conservative than the Northern, though less so than the South-Thames dialects. It will be seen, then, that the criteria of full, levelled, and 8 HISTOR Y OF ENGLISH. [ 24. lost endings by which we distinguish the periods of English (1) apply only to the South-Thames dialects. STRUGGLE BETWEEN FRENCH AND ENGLISH. 24. For a long time the two languages, French and English, kept almost entirely apart. The English of 1200 is almost as free from French words as the English of 1050; and it was not till after 1300 that French words began to be adopted wholesale into English. 25. Meanwhile English was steadily gaining the upper hand. In 1258 we find it officially employed in the Procla- mation of Henry III. In the next century French gradually fell into disuse even among the aristocracy. In 1362 English was introduced in the courts of law instead of French. About the same time English took the place of French as the vehicle of instruction in schools. RISE OF THE LONDON DIALECT. 26. In the ME period the dialects had diverged so much that speakers of the extreme Northern and extreme Southern dialects were no longer able to understand one another, and the need of a common dialect became pressing. Such a common dialect can be formed only in a centre of intercourse where speakers from all parts of the country meet constantly. Such a centre was London, which now was not only the capital of England, but also a place of great and growing commercial importance. 27. The London dialect, as we find it in its earliest document, the Proclamation of Henry III, shows such a mixture of Midland and Southern forms as we might expect from its position on the border-line between these two 29.] HISTOR Y OF ENGLISH. 9 dialects. The Midland dialect was intermediate between the two extremes, Northern and Southern, not only geographi- cally but also linguistically; so that speakers of Midland could understand both Northern and Southern much better than Northerners and Southerners could understand one another. Hence the Midland element in the London dialect made the latter peculiarly fitted to serve as a means of general communication. Hence also the Midland element in the London dialect became stronger and stronger in the course of the ME period, till at last even Northern forms passed into it through the medium of the Midland dialect, while Southern influence became weaker and weaker. SCANDINAVIAN INFLUENCE. 28. Although the Norwegians and Danes spoke different dialects, the difference between these dialects was very slight. The Scandinavian words imported into English seem to be mostly Danish. Although the Scandinavian dialects were not intelligible to the Anglo-Saxons, yet the cognate languages English and Scandinavian were so similar in structure and had so many words in common, that the languages blended together with the same facility as the races that spoke them. English got the upper hand, but Scandinavian nevertheless left its mark on every English dialect, especially the East- Midland and Northern dialects, where the population was half Scandinavian. Ill, fro in 'to and fro,' bound in 'bound for a place/ are examples of Scandinavian words in English (Icelandic ill-r l bad,'frd 'from/ buinn 'ready'). FRENCH INFLUENCE. 29. The Norman French introduced into England was not a uniform dialect, but was itself split up into local 10 HISTORY OF ENGLISH. [30. varieties or sub-dialects, which in the Norman spoken in England the 'Anglo-Norman' or 'Anglo-French' language were mixed together indiscriminately. The loss of Nor- mandy in 1204 put an end to the influence of Continental Norman; and henceforth Anglo-French was influenced only by the literary French of Paris, this Parisian French having the same predominance among the French dialects as London English had among the English dialects. At the time when the influence of Anglo-French on English begins to be important that is, in the late ME period it was, therefore, a mixture of Old French of different periods and different dialects, modified by changes of its own, and also by the influence of English itself, especially in its pronunciation. 30. French influence on English is most marked in the vocabulary. Soon after the Conquest English ceased for several centuries to be the language of the higher purposes of life, and sank almost to a mere peasant's dialect. So when English came again into general use, it had lost a great part of its higher vocabulary, for which it had to use French words, such as sir, duke; captain, army, battle; sermon, preach. Even when the English word was kept, the same idea was often expressed by a French word, whence numerous synonyms such as work and labour, weak and feeble. LATIN INFLUENCE. 31. In Old French itself we must distinguish between popular and learned words. The popular words in Old French, such as sire ' lord,' from Latin senior ' older/ are simply Latin words which have undergone those changes which take place in every language whose development is natural and unimpeded. But as Latin was kept up as an 3aO HISTOR Y OF ENGLISH. 1 1 independent language throughout the Middle Ages, Latin words were imported into Old French as well as the other Romance languages, being used first in books, then in ordinary speech. These learned words were kept as much as possible unchanged, being pronounced as they were written. It often happened that a Latin word which had assumed a popular form in French, was re-imported direct from Latin, so that chronological doublets were formed, such as caitif ' wretched ' and captif, both from Latin capfivus, whence the English caitiff and captive. 32. These learned French words were introduced into ME in great numbers. Hence when Latin words came to be imported directly into English, they were put into a French shape on the analogy of those Latin words which had really been brought in through French. Thus when a word in -Ho, such as nomindtid, was taken direct from Latin, it was made into -tion (MnE nomination) on the analogy of the older importations, such as nation (ME nacioun). Modern English. 33. In the Middle period literary English was still dis- tinctly an inflectional language. In the Modern period it became mainly uninflectional, with only scanty remains of the older inflections. 34. The Modern period is that of the complete ascen- dency of the London dialect, which henceforth is the only one used in writing throughout England. Henceforth the other dialects of England continued to exist only as illiterate forms of speech confined within narrow areas. 35. The spread of Modern London English or ' Stan- dard English,' as we may now call it was greatly aided by J 2 HISTOR Y OF ENGLISH. [ 36. the introduction of printing in 1476. The publication of Tindal's translation of the New Testament in 1525 paved the way for the Authorized Version of 1611, which made Early Modern London English what it has ever since been the sacred or liturgical language of the whole English-speaking race. INFLUENCE OF OTHER LANGUAGES. 36. In the Early Modern period, the Renascence the revival of the study of the classical authors of Greece and Rome led to the adoption of an immense number of Greek as well as Latin words, the Greek words being generally Latinized, just as the Latin words imported into Middle English were Frenchified. 37. As the first prose writings were mostly either trans- lations from Latin, or else the work of scholars to whom Latin was in some respects a more natural means of ex- pression than English, it was inevitable that Early MnE prose was greatly influenced by Latin, not only in vocabu- lary, but also in grammatical structure and idioms. In a few generations many Latin and some Greek words and expressions which were at first purely learned and technical passed into the language of everyday life ; while, on the other hand, many others became obsolete. 38. As the relations of England with other countries became more extended, many words were imported into English from almost every European language, especially Dutch, French, Italian, Spanish, and Portuguese, and from many other languages besides, such as Arabic, Persian, and Turkish, and the native languages of America. 39. Standard English has always been influenced by the different English dialects. The literary revival of Broad 40.] HISTOR Y OF ENGLISH. 1 3 Scotch which is really Modern Northumbrian at the end of the last century by Scott and Burns has introduced many Scotch words into literary English. PERIODS. 40. The main general difference between Early and Late MnE is that the former is the period of experiment and com- parative licence both in the importation and in the formation of new words, idioms, and grammatical constructions. The Late MnE period is, on the other hand, one of selection and organization. The most marked differences in detail are the great sound-changes undergone by the spoken language changes which have been completely disguised by the fixity of the orthography. PHONOLOGY. PHONETICS. 41. Phonetics is the science of speech-sounds. 42. As the ordinary spelling does not always show the real pronunciation, it is necessary to use a phonetic spelling, which, to prevent confusion, we enclose in ( ). Analysis. 43. The foundation of speech-sounds is breath expelled from the lungs, and variously modified by the vocal organs throat, nose, mouth, lips. THROAT-SOUNDS : BREATH AND VOICE. 44. The first modification the breath undergoes is in the throat. If the vocal chords, which are stretched across the inside of the throat, are kept apart so that the air can pass through with but little hindrance, we have breath, as in ordinary breathing or sighing, and in the consonant (h), as in high. If the chords are brought together so as to vibrate, we have voice, as in murmuring or in the word err. NASAL SOUNDS. 45. If the passage into the nose is left open, we have a nasal sound, such as (m) in am. In the formation of all 5 o.] PHONETICS. 15 sounds that are not nasal, such as the (b) in amber, the nose-passage is closed. CONSONANTS. 46. If the mouth-passage is narrowed so as to cause audible friction that is, a hissing or buzzing sound or if it is completely stopped, a consonant is produced. VOWELS. 47. If the mouth-passage is left so open as not to cause audible friction, and voiced breath is sent through it, we have a vowel. Every alteration in the shape of the mouth produces a different vowel. VOWEL-LIKE CONSONANTS. 48. Some consonants have hardly any friction when voiced, and are called vowel-like consonants. Such con- sonants are (1), as in little (litl), and (m). Synthesis. 49. We have now to consider the synthesis of sounds, that is, the different ways in which they are joined together in speech. QUANTITY. 50. By quantity, sounds are distinguished as long, half- long or medium, and short, ' long ' being often used to include half-long as well. In phonetic notation long and half-long vowels are doubled, short vowels being written single, as in (maama) murmur. The length of consonants is only occasionally marked by doubling. 1 6 PHONOLOGY. [51. STRESS. 51. There are three main degrees of stress or loudness : strong, half-strong or medium, and weak. Thus in con- tradict the last syllable is strong, the first half-strong, the next weak. We mark strong stress by (), half-strong by (:), these marks being put before the sound on which the stressed syl- lable begins, weak or unstressed syllables being left unmarked : (:kontr3'dikt). Weak, stress is marked when necessary by prefixing (-), as in (-it reinz) ' it rains.' 52. Sounds which occur only in unstressed syllables, such as the short (a) in (maama) murmur, are called weak. INTONATION. 53. Intonation or tone is either level, rising, or falling, marked respectively (",'/) The level tone is not much used in speech. The rising tone is heard in questions, such as what', the falling in answers such as no\ GLIDES. 54. Glides are sounds produced during the transition from one sound to another. Thus in (kii) key we have the glide from the (k)-position to the (ii)-position, which does not, however, require to be written, as it is implied by the posi- tion of (k) and (ii). 55. Consonants are often joined together without any glide, not only in such combinations as (nd) in hand, where the (d) is formed by continuing the (n), the nose-passage being closed at the same time, but also in such words as the English act (sekt). SYLLABLES. 56. A syllable is a vowel, either alone or in combination with consonants, uttered with a single impulse of stress. 59-] PHONETICS. \>] Every fresh impulse of stress makes a new syllable, the be- ginning of the syllable corresponding with the beginning of the stress. Thus (a'taek) attack has two syllables, the first syllable consisting of the vowel (9) uttered with weak stress, the second of (tsek) uttered with a new impulse of stress beginning on the (t). Vowel-like consonants often form syl- lables in the same way as vowels, as in <$#/// =(bset-l). DIPHTHONGS. 57. If two vowels are uttered with one impulse of stress, so as to form a single syllable, the combination is called a diphthong, such as (oi) in oil. Most diphthongs have the stress on the first element, A simple long vowel, such as (aa), is called a monophthong. We now have to consider sounds more in detail. Vowels. 58. As every alteration in the shape of the mouth pro- duces a different vowel, the number of vowels is infinite. Hence what we call the vowels, (a), (i) etc., are really groups of an indefinite number of vowels differing very slightly from one another. ROUNDING. 59. The shape of the mouth-passage by which vowels are formed depends partly on the position of the tongue, partly on that of the lips. If the lip-opening is narrowed while the tongue is in a certain position, the resulting vowel is said to be rounded. Thus (y) in French lune is the round vowel corresponding to the unrounded (ii), which is nearly the sound in English he, both vowels having the same tongue- position. c 1 8 PHONOLOGY. [ 60. TONGUE-RETRACTION. 60. The tongue-positions depend partly on the degree of retraction of the tongue, partly on its height or distance from the palate. 61. If the root of the tongue is drawn back, we have a back vowel, such as the (aa) in father. If the fore part of the tongue is advanced, we have a front vowel, such as (ii). If the tongue is left in its neutral position, intermediate between back and front, we have a mixed vowel, such as (39). TONGUE-HEIGHT. 62. If the tongue is raised as close to the palate as is possible without making the vowel into a consonant, a high vowel is formed. Thus (i) is a high-front vowel, (u), as in full, a high-back-round vowel. There are two other degrees of height, mid and low. For convenience we may include mid and low vowels under the common name ' un-high ' vowels, distinguishing them as close and open, according to the degree of openness of the mouth-passage. We denote open vowels, when necessary, by italics. French / in ett is the mid- front-close vowel, or, more briefly, the front- close vowel, for when a vowel is not expressly called high, we assume it to be un-high. English (e) in men is the corre- sponding mid-front-open vowel. Very open vowels are called broad, (ae) in man is a broad front vowel. The distinction of close and open applies also to the high vowels. Thus French (i) infini is the close high front vowel, English (?) in finny is the open high front vowel. 70.] PHONETICS. 19 THE VOWELS IN DETAIL. The following are the most important vowels. (A) Unrounded vowels. 63. (a) 'clear back': (a'haa) aha I 64. (B) ' dull back ' : (sim) son, sun. 65. (a) ' mixed ' or ' neutral ' vowel : (maamg) murmur. 66. (i) ' high front/ Close (i) in French fini, the short E. i, as in finny, being always open. Long close (ii) is the older E. sound in such words as see, sea, receive, machine, and this sound is still preserved in Scotland and the North of England. In the South of England it is diphthongized into (z) followed by very close (i), which is nearly the sound of the consonant (j) in you, so we write (sij), etc. 67. (e) ' front.' French / is close front. The E. vowel in men is open front (e). Before (a) with which it forms a diphthong it is still opener, as in (fea) fare, fair. The long close front (ee) is still preserved in Scotch in such words as name, day, where Standard E. has the diphthong (ei). 68. (&) ' broad front' : (msen) man. (B) Round vowels. 69. (u) '^igh back round.' Close in French sou, the E short (u) in/ull, good being always open. The older close (uu) in such words as moon, move, you (juu) is still kept in Scotland and the North of England, but in the South of England it becomes (w) with a distinct (w). Weak open (u), as in value, is the high mixed round vowel, which, when necessary, we write (ii) (vgeljii). 70. (o) ' back round.' Close in French beau (bo). Close (oo) in Scotch no, know, where Standard E. has the diphthong c 2 20 PHONOLOGY. [ 7'. (0u). The (o) in the diphthong (ed) is red pronounced with a rounded (r). 93. When a consonant is modified by raising the front of the tongue, it is said to be front-modified or fronted, which we express by adding (/). Thus the lip-open front- modified consonant is the sound in French huit (/3/It) ; it is almost a consonantal (y). THE ASPIRATE. 94. The aspirate (h) is partly an open throat consonant, partly a breath vowel-glide. Thus (h) in hook is mainly formed by unvoicing the beginning of the (u), almost as if we were to write the word (w^uk). PHONOLOGY. [ 95- 95. The following is a table of the chief consonants. Those marked * do not occur in E. BREATH. Throat. Back. Front. Point. Blade. Blade- Point. i>. Lip- Back. Lip. Teeth. Open h *x *9 *rA,) s J V VfA f Side *\h Stop k *c t P Nasal *rjA *KA *n/* *m VOICE. Open *5 j r, 5 z 3 * w V Side 1 Stop g *q d b Nasal i) *fl n m 96. We generally write (rh), etc., instead of (r^) for the sake of convenience. R IN ENGLISH. 97. (r) in E. occurs only before a vowel following it without any pause, as in here he is (hiar ij iz) ; before a consonant or a pause it is dropped, leaving only the preceding (9), as in here she t's, he is here (hia Jij iz, hij z hia). This (a) is ab- sorbed by a preceding (aa, aa), as in err, erring, far, far away (aa, aarirj, faa, faar awei). After (o) the (a) is kept finally, but dropped before the (r), as in pour, pouring (poa, porirj), being also dropped before a consonant in the same word, as in poured (pod). 102.] LAWS OF SOUND-CHANGE. 25 LAWS OF SOUND-CHANGE. 98. Sound-changes fall under two main classes internal and external. 99. Internal changes are either organic or acoustic. Organic changes are due to the natural tendencies of the organs of speech, as in the change of OE start into MnE stone through the natural tendency to pronounce a back vowel without opening the mouth fully, and so to round it. 100. Acoustic changes are the result of the impressions which sounds make on the ear, as when one sound is sub- stituted for another because of their likeness to the ear : thus children often make through (Jjruw) into (fruw), and point (r) is changed into back (5) in French and other languages. These are imitative changes. 101. External changes are those which are independent of organic and acoustic tendencies. Thus the change of spake into spoke in MnE is not the result of any tendency to change a into o in MnE, but of the influence of the preterite participle spoken (566). 102. Internal changes are further distinguished as isolative and combinative. Isolative changes, such as that of OE a into MnE o, affect a sound without regard to its surround- ings, while in combinative changes one sound is modified by another one close to it, as in the change of ME (au) in saw into MnE (soo) through (sou) or (sou). Here we have two distinct combinative changes : first the rounding of the (a) by the influence of the following (u), and then the lowering of the high (u) till it is merged into the (o). We see that the influence of one sound on another is either backwards, as 26 PHONOLOGY. [ 103. in the change of (au) into (ou), or forwards, as in the change of (ou) into (oo). 103. All combinative changes are, besides, either conver- gent or divergent. Convergent changes, as of (au) into (ou) are organic, being due to the tendency to save trouble by making the passage from one sound to another as short and easy as possible. 104. Complete convergence or assimilation in diph- thongs makes them into monophthongs, as when (ou) be- comes (oo), and in this case is called smoothing. 105. Divergent changes are often partly acoustic, being due to the striving for distinctness, as when the half diph- thong (ou) in no is made into full (au) in Cockney E. But cleaving, by which a long vowel is made into a diphthong, is an isolative organic change; it consists generally in forming the first half of the vowel with greater openness ' either of the mouth- or the lip-passage than the second. We see the beginning of cleaving in the E. change of (ii, uu) into (ij, uw), which by divergence could easily become (ei, ou) or (ai, ou) and then (ai, au). OLD-ENGLISH SOUNDS. Orthography. 106. The Anglo-Saxons brought with them to England their national Runic alphabet, which was founded on one of the Old Greek alphabets or possibly the Latin. On their conversion to Christianity they adopted the Latin alphabet in its British form, to which they afterwards added the two Runic letters J>=/^ and p=z#. In the British-Latin alphabet i io.] OLD ENGLISH. 27 and consequently in the OE alphabet as well several of the letters had peculiar forms, g for instance being written 5. 107. Each letter of the Latin alphabet was used to denote the OE sound nearest to that which the letter had in the pronunciation of British Latin, which was more archaic than that of the Continental Latin. 108. Spelling in OE was purely phonetic : the OE scribes wrote as they spoke, as far as the defects of their alphabet would allow them to do so. 109. In this book we supplement the defective distinctions of the OE orthography by adding diacritics, which gives the following new letters /, g, c, g, a, etc., (~) denoting vowel- length. Pronunciation. 110. The vowels had the same sounds as in our phonetic notation, the unmodified vowels being all close except a. a=(a), as infaran ' go/ ' travel' ; long in stan ' stone/ e ' he who.' 126. In Anglian, short vowels were lengthened before vowel-like consonants followed by another consonant ' group-lengthening ' as in did ' old,' Ijjng ' long/ blind 'blind/ dumb ' dumb '= Early West-Saxon eald, Igng, lang, blind, dumb. These lengthenings appear also in Late West- Saxon. Vowels. 127. a (9), ae, ea. These vowels all correspond to Germanic a, still preserved in Modern German ; thus OE mann, feeder, heard= German mann, vater, hart. Germanic a in the Oldest E. was kept only before nasals, as in mann, hand, lang. Everywhere else it was fronted to a, as in wees ' was/ cecer ' field/ feeder. Before ' group r and // that is, before r and / followed by a consonant, and before strong h the voice-glide (a) was developed, as in E. (hiarirj), which afterwards by phonetic divergence developed into full (a), as in heard, earm ' arm '; eall, eald ( old ' ; ge-seah ' saw/ eahta ' eight/ weaxan (112). Before a back vowel in the next syllable & became the back vowel a, as in dagos ' days/ dagum ' to days ' dat., compared with dceg ' day/ gen. doges. These are the West-Saxon forms. In Anglian a before nasals became g as also often in Early West-Saxon and ce 32 PHONOLOGY. [ 128. before group / became a, so that the Anglian forms are mgnn, hgnd (126), Igng ; heard, etc. ; all y did (126). 128. i, e, eo. In Germanic, e before group-nasals became z, whence OE bindan 'bind,' singan 'sing' compared with helpan ' help.' In OE itself e also became i before single nasals, as in niman ' take ' compared with stelan ' steal.' The vowel in such words as witan ' know ' is Germanic and Arian i. In OE e before group r became eo much in the same way as a became ea (127), as in steorra ' star,' eorpe. e, i became eo, io before a back especially a back round vowel in the next syllable, as in heofon ' heaven,' cliopian ' call,' the forms hefon, clipian also occurring. 129. u, o. In Germanic, o became u before group-nasals, and in OE itself o became u before single nasals, whence OE gebunden ' bound ' compared with geholpen ' helped,' genumen ' taken ' compared with gestolen ' stolen.' In such a word as sunu ' son,' the u s are Germanic and Arian. 130. The Germanic vowel & is preserved in West- Saxon, as \T\fizr 'danger,' izfen 'evening,' being narrowed to e in Anglian and Kentish -fer, efen. MUTATION. 131. Mutation is the influence exercised by a vowel on the vowel of a preceding syllable, by which the first vowel is modified in the direction of the second one. Thus in OE gecoren ' chosen '= Old High German gikoran, compared with OE c urun later curon ' they chose/ u has been lowered to o by the influence of the a. This is therefore an 0-mutation of u. 132. But the most important mutations in OE are the front mutations, caused by Germanic i and /, which after 1 330 OLD ENGLISH. 33 they had caused the mutation were generally lost or modified inOE. 133. The following are the mutations in their Early West- Saxon forms : e . . . i. beran ' carry,' birep (Oldest E. birifi) 'carries'; cweban ' say,' cwt'de (Oldest E. cwtdt] ' saying,' ' speech.' a (se) . . . . faran 'go,' ' travel,' fgrian 'convey'; mann ' man/ mgnn (Germanic manni} ' men.' a . . . f. hdl ' whole,' ' sound,' hfzlan ' heal ' ; an ' one,' tenig ' any.' This ' mutation & ' remains in the non- West- Saxon dialects, which change Germanic ce into respectively, as in cwddon, cwide compared with cweban, slcegen 'struck,' slaga ' slayer ' compared with slean [from *sleahan\ ' strike/ ' kill.' These changes are the result of weak stress of the syllable containing s, />, h in Early Germanic. Hence we call the resulting r 'weak r' to distinguish it from r= Ger- manic r, and so with the other consonants. 147. b in the combinations, tj>, dp, sh becomes /, to which a preceding d is assimilated, giving the combinations //, s/, as in Early West-Saxon bitt=.blteb ' bites ' and bideb ' waits/ clest ' chooses ' from ceosan. 148. Double consonants in OE often represent a Germanic single consonant +/, as in s^llan 'give/ sc^bban 'injure/ sultan ' set '= Gothic saljan, skabjan, saljan, the single con- sonant appearing in such forms as sgleb, scgbeb, sgteb ' he gives/ etc., which point to older *salip, etc. Germanic kj, SJ'fJ appear in OE respectively as cc, eg and bb, as in wrgcca ' one exiled/ l/cgan ' lay/ hgbban ' raise ' compared with wracu ' state of exile/ lag ' he lay/ hafen ' raised.' Germanic rj, on the other hand, appears as ri in OE. as in dorian ' injure ' [cp. daru ' injury ']. 149. In OE itself, c, /, p are often doubled before r and /, as in biter, bitter ' bitter ' [cp. bitan ' bite '], appel ' apple ' [cp. apulder 'apple-tree'], nikke ' thick,' cw being expressed by the Early Old French qu, as in quene ' queen ' = OE ewe, n. c was kept before back vowels and generally before consonants, as in cumen, comen, cljne ' clean.' The ME development of OE c having nearly the sound of French ch t this digraph was used to express it, as in chirche=QiE cirice. c=(s) was used only in French words, such as face. 160. In ME the difference in form between the OE 5 (106) and the French g was utilized phonetically. The letter g was assigned to (g), as in god ' good/ and the soft French g, as in geste ' exploit/ and also to the ME develop- ment of OE stopped g, which had nearly the sound of (d^), as in sengen 'singe/ brigge ' bridge '=OE s/ngan, brycg. Hard g was also expressed by the French gu, as it still is in iongue= OE iunge. j=(d%) was written only in French words, such as juggen ' judge.' 5, on the other hand, was restricted to the open sounds, both back and front, as in days, %ung=O~E. dagas, geong, the latter sound being after- wards expressed by_y, as in MnE : yong, young. 161. After much fluctuation OE strong h was written gh, as in right, doghter. 162. Latin z still kept its sound (dz) in Early Old French where it was also used to express (ts), as in douz ' sweet' and did not become simple (z) till a later period. Hence it is not till the end of the ME period that they began to write z instead of s=(z) in E. words, as in wezele 'weazel/ generally written wesele. 163. The Latin sound (w), which was expressed indiffer- ently by the angular v or the round u, became (v) in Old French, the old symbol being kept, so u, v became the symbol of voiced OE/in ME, as in /# was used throughout the ME period, but the digraph th soon came into use to express the voice as well as the breath sound of }>, as in brjben, brethen (br^tSan) ' breathe/ brjfi, breth (breep) ' breath/ In Old French th was written only in learned words, proper names, etc., and had the sound (t), which it often kept in ME as well ; we still pronounce such words as Thomas with a (t), as in ME. Old French ph=(f) was also used only in learned words and names, f being often substituted for it ; it was used in ME in such learned words as phisik ' physic,' also written fisik. Stress. 165. In ME the noun- and adjective-prefixes al-, mis-, un- throw the stress forward, as in almihti, misled, un'cup ' un- known '=OE '(zlmihtig, -misdeed, -uncuj>. 166. In Old French the stress generally fell on the same syllable as in Latin, as in na'ture Latin nd'turam. Through the dropping of final Latin syllables many French words thus came to have the stress on the last syllable, as in o-nour =.ho-norem,prte=pieiatem. When first introduced into ME French words kept their original stress: nd'tUre, o-nur,prte; 42 PHONOLOGY. [ 167. but such words afterwards threw the stress back on to the first syllable by the analogy of the native E. words, such as fader, 'bodi, becoming -nature, etc. 167. In longer French words, where it would have been inconvenient to throw the stress back to the first syllable, it was drawn back from the end to the middle of the word, as in sove-reynete, condicioun (kon-disiuun) and the other words in -ioun= Latin -ionem. 168. Many words of French origin compounded with particles, such as crvow (a*vuu), defense, disuse (dis'^za), keep their original stress by the analogy of native words such as a'risen, be-cumen. Quantity. 169. The first quantity-change that took place in ME was the lengthening of OE short consonants after a short strong vowel, so that OE in ' in ' and inn ' dwelling ' were levelled under the latter form ; and as it was no longer neces- sary to mark the distinction, the OE double consonants were written single, as in a!, man=QE, eall, mann. But double consonants before vowels were kept in ME in pronunciation as well as spelling, so that, for instance, sunne *sun' = OE sunne was kept distinct from sune ' son '=OE sunu, these two words never rhyming on one another in verse. 170. The OE group-lengthenings were kept up in ME, as in gld, Igng, blind, dumb, douml>=Q\d Anglian did, Igng, blind, dumb. Otherwise OE long vowels were generally shortened before two consonants, as in askien, wisdom [compare ME wis ' wise '], kepte ' kept ' pret. = OE ascian, wisdom, cepte. But length was often preserved before st, as in Ijsf 'least,' prest ' priest ' = OE last, preost. 1 74 .] MIDDLE ENGLISH. 43 In the transition from ME to MnE the long vowels before ng and mb were shortened, whence MnE long, young (j^n), dumb compared with old (ould), blind (blaind). Hence also OE -anc, -gnc appears as -ank in MnE, while OE -ang, -gng appears as -ong, as in lank = OE hlanc compared with long = OE lang. 171. In Late ME short vowels before a single consonant followed by another vowel were lengthened, as in name, mjte ' meat/ brgken ' broken '= Early ME name, mete, ibroken=.Q]L nama, mgte, gebrocen. We call these lengthened vowels ' new- longs ' as opposed to the ' old-longs ' in such words as win 'wine'=OE win. But the high vowels t, ii, u were never lengthened, as in writen 'written,' dude 'did,' j=OE gewriten, dyde, sunu. 172. Vowels were not lengthened in final strong syllables, as in smal, swan, y of ^ gave,' GW=OE smal, swan,geaf, God, because the final consonants had already been lengthened (169). 173. Short vowels are often preserved in Late as well as Early ME before a single consonant followed by the full vowel /', as in mam', peni, bodi, or weak e + a vowel-like conso- nant (r, 1, n, m), as in hamer^feter, coper; sadel, hovel; seven, troden, all of which still have short vowels in Present English. This is called back-shortening. Originally long vowels are sometimes back- shortened in ME, as in laper from OE leapor. But there are several exceptions to the general principle of back-shortening, as in Late ME dker t crddel, stplen=O~E =.Q~E. smifi, but also to OE^, as in sinne, dtde. But (y) was still preserved in the Southern dialect, as in siinne, diide, being represented by e in Middle, as well as Old Kentish, as in senne. The London dialect generally has r=OEj/, but some words have the Southern, and the few the Kentish forms : sinne, bust, kernel=. OE synn, bysig ' occupied/ cyrnel, ' kernel.' In some words (y) was broadened to (u), especially after lip-consonants, as in worien ' worry/ moche 1 much '=OE wyrgan, my eel, micel. 179. e. OE close (e) became open (e) in Early ME, so 1 8s.] MIDDLE ENGLISH. 45 that OE e and / were levelled under the latter sound, which we write simply e in ME, as in helpen, eten, rest, meteQfc helpan, elan ; rgst, mgte. OE eo also became open e in Late ME, as in erbe, hevene. All these es are liable to be lengthened in Late ME (171), as in gten, mgte. 180. u. OE u was kept unchanged in ME, as in sum. 181. o. OE close o became open in Early ME, as in folk, nose, bodien ' proclaim '=OEye> lc, nosu, bodian, being liable to lengthening in Late ME, as in nose, bgdien. 182. The OE long vowels z, e, ce, u, o were generally pre- served unchanged in ME, e, & being also the representatives of OE eo, ea respectively (176) : win, kene ' bold,' dep, s/ 'sea/ hgved ' head,' hits, hous, god ' good '=OE win, cene, deop, so;, heafod, hus, god. So also M.Efmden,feld ' field/ hund 'dog/ word 'word '= Anglian findan, feld, hund, word (126). i is sometimes the result of raising Anglian e before open g and front h, as in ie 'eye/ hih ' high '= Old Anglian ege, heh, West-Saxon eage, heah, the open^-=(j) being absorbed. So also open g was absorbed in ME by a preceding u or u, as in fuel 'bird/ buen later bowen ' bend ' = OE fugol, bugan. It is to be observed that ME e represents not only the com- mon OE e in cene, but also the Anglian )-sound. 213. We are now able to answer the question, Why is English spelling unphonetic ? The main reason is that it has not followed the changes of pronunciation. The present English spelling represents not the sounds of Present English but those of Early MnE or rather Late MnE. Such a spelling as knight is not in itself unphonetic ; on the con- trary, it is a phonetic representation though an imperfect one of the sound-group (knict), which in ME was the pro- nunciation of one of the words which we now pronounce (nait), the other one having been pronounced (nift) in ME, and written accordingly night. Such a spelling as island is, on the other hand, unphonetic from every point of view, because it inserts a letter which is not pronounced now, and never was pronounced. Such a spelling as author was also origin- ally unphonetic, though it has now become phonetic but only by corrupting the pronunciation and obscuring the etymology of the word. 58 PHONOLOGY. [214. Vowels. 214. The most convenient way of dealing with the MnE vowels is to take each Late ME vowel separately, and trace its history down to the present time. 215. a was gradually advanced to the broad (as), so that such words as man, sat had exactly their present pro- nunciation in Second MnE. But in First MnE the old (a)-sound was still kept by many speakers. Before / not followed by a vowel a kept its back sound and the glide between it and the / developed into an (u), so that such words as fall, calm became (faul, kaulm), being sometimes written/a#// etc. (a) was also kept after (w, wh), as in was, what, where it was rounded in Second MnE, whence the present (woz, whot), although there was no rounding when a back consonant followed, as in wax, wag. In Second MnE (se) was lengthened before (s, J>) and in some other cases, as in glass, path (glseses, psese]?). At the end of the Third MnE period this (sese) was broadened into (aa), which is the present sound (glaas, paa]?). 216. i, e have generally remained unchanged. But in First MnE er final or before a consonant became (ar) as in star, hart, /foar/=ME sterre, hert, herte. Not in the weaker. 217. u was preserved in First MnE, as in full, come (kum). In Second MnE it was unrounded to (A), which was afterwards lowered to its present sound (B) (fel, kern). But before this lowering took place the (A) was generally rounded back again to (u) between a lip-consonant and (1), as in full, wool= ME wolle, and in other words after lip- consonants, as in wood ME wode, put. 218. li generally appears as i in MnE, into which it had already been unrounded in the London dialect of ME. Thus 223.] MODERN ENGLISH. 59 MnE has///, sinQ&gefyllan, synn. But (y) was preserved in First MnE in some words still written with the French u, such as buy, fatty =OE bysig^ byrgan. 219. o kept its ME sound (0) in First MnE, as in top, ox, and was broadened to its present sound in Second MnE, being lengthened before the same consonants which lengthen (ae), as in froth, cross, off. In Early MnE a glide-(u) developed between (o) and / not followed by a vowel, as in bowl (b and became (iu) in Second MnE, all the three ME sounds #, zV/wr^, which are now vulgarisms, were in general use. As (B) was very similar in sound to (3), there was a tendency to make (a) the general weak vowel, although the older clear weak vowels were still kept in many cases, as in (nsej^ngel, naefonsel) national, now pronounced (naefanal). In Second MnE weak initial vowels were often dropped, especially in long words, as in apprentice (prentis), estate ), opinion (pinjan). We still keep the short form of 234-1 MODERN ENGLISH. 63 the first word in the expression 'prentice hand, but the vowel has generally been restored by the influence of the spelling. Consonants. 232. During the transition from ME to MnE the hisses/, s,f, became voiced in weak syllables, especially in inflectional -es, as in the gen. sing, mannes and the plur. stgnes, whence MnE (maenz, stounz), the breath sounds being preserved in strong monosyllables such as ges, pens=MriE (gijs, pens) contrasting with pem'es=M.nE. (peniz). The same change was carried out in weak monosyllables, so that numerous doublets were formed. Thus the emphatic adverb 0/"=MnE off preserved its (f), while the preposition ), and a 4 o.] MODERN ENGLISH. 6.5 in many words the lip element was exaggerated in Second MnE till it became (f) (Isef, laesef, >oft, Ipoot, in^f) which in draft by the side of draught both from ME draght has been adopted in the spelling. 238. r was kept unchanged in First MnE, being after- wards gradually weakened till it lost its trill everywhere. Towards the end of the Third period it began to be dropped everywhere except before a vowel, as in the present Standard E. 239. Already in First MnE (r) had developed a glide before it in such words -as. fire, flower (faiar, fl6uar)=ME fir, flur, and had broadened a preceding e into (a), as in star (216). In Second MnE it began to modify preceding vowels in the direction of (a), so that er, ir, ttr came to be levelled under (ar) or (w), as in her (hBr) fir, bird, fur, turn. In Third MnE it modified preceding (ee)=a, at, et'to (ee), as in care (keer), fair, their contrasting with name (neem), fail, veil; and towards the end of this period it broadened a pre- ceding (ae) into (a), as in star, hard. ME jr, or appear in Third MnE sometimes as (iir, uur), as in fear, moor, being sometimes broadened into (eer, or), as in there, bear, floor. In the present century (r) has been dropped everywhere except before a vowel, r final or before a consonant being represented only by a preceding glide-(9), as in (faia) = Early MnE (faiar) = ME fir. This (a)=r has broadened preceding (ij, uw) into (i, u), as in here (hia), poor, cure (kjua) contrasting with he (hij), pool (puwl). The glide-(a) before (r) was finally absorbed by a preceding mixed or broad vowel, (er) in her etc. passing through (sa) into (aa), (aa, oa) into (aa, o), as in star, floor. 240. 1. Already in First MnE (1) began to be dropped between (u) and a following consonant, as in half (haulf, K 66 PHONOLOGY. [ 241. hauf), folk (foulk, fouk); also in should (Juuld, Juld, Jud), would, could, where the (1) was at first dropped only when these words were weak. 241. s, z. In Second and Third MnE the combinations (sj, zj) became (j, 5), as in nation (n^jBn)=Early MnE (naeaesjun)=rME nacioun (naasi-uun), sure (siur, sjuur, Juur), usual (iuziuael, juusuael), such words as nature, verdure passing through (naeaetjur, neetpr, verdjur, verdjar) into the present (neitfa, vaadsa). 242. w in First MnE was kept before (r), which it rounded, and was then dropped itself, as in write (rz^ait), the (r) being afterwards unrounded. 243. In Second MnE w was dropped in weak syllables, especially in -ward, -wards, as in Edward (edard), backwards (baekardz). We still drop the w in towards (todz), but it has been restored in the other words through the influence of the spelling, except in vulgar speech. The weak ending -wich drops the w in all familiar place-names, such as Greenwich (grinids). 244. k was kept initially before (n) in First MnE., as in know [compare acknowledge}, the (n) being unvoiced, and the (k) afterwards dropped, so that in Second MnE (kn0u, kn^ou) became (n^oo), this (n/fc) being afterwards levelled under the more frequent (n) in no, etc. 245. g was dropped before (n) in Second MnE as in gnaw. 246. In First MnE medial (rjg) was shortened to (n) in such words as singer (sinar), singing=M^ (singer), etc. by the analogy of final (n) in sing ; but (rjg) was kept in the comparison of adjectives, as in longer, longest. 247. t, d. In Second MnE (t) preceded by the hisses (s, f) and followed by the vowel-like consonants (1, n, m) 252.] PRESENT ENGLISH: STRESS. 67 was regularly dropped, as in thistle (]>isl), fasten (faesesn), chestnut, Christmas, often. 248. In First MnE (d) preceded by a vowel and followed by (r) was opened into (t$) in many words, such as father, together, hither -=OE fader, Late ME fader, fader (173), OE td'gcedre, hider. Conversely (S) often became (d) in First MnE in combination with (r) and (1), as in murther, murder, rudder, fiddle ~QE morfior, robor,fi}>ele. 249. b. In First MnE final (b) was dropped after (m), as in lamb. Hence b was added in writing to words which in ME had only m, as in limb, num&=M.JL Urn, inumen 'taken,' ' seized ' =. OE genumen. PRESENT ENGLISH. Stress. WORD-STRESS. 250. The characteristic features of Present English stress are some of them of OE origin, while others developed them- selves in ME and in the different periods of MnE, some being apparently of very recent origin. 251. In Present English, as in OE, the most general principle of stress is that subordinate words especially form- words have weak stress. Thus in he is a man of the world, the subordinate words he, is, a, of, the all have weak stress. Hence the weakened stress in a .-piece of bread, and the distinction between -some bread and 'some 'people. 252. The OE principle of putting the stress on the first syllable of a word generally resulted in the principal stress being on the root-syllable of inflected or derived words. F 2 68 PHONOLOGY. [ 253. This principle is still maintained in MnE in native words, as in fearful, fearfully, fearless, fearlessness, fisher, fishery ', fisherman (fijaman). 253. We have seen that already in ME many long words of French origin with the stress on the last syllable threw it back on to the first syllable by the analogy of the native stress (167). In MnE this tendency has become stronger and stronger, so that the first-syllable stress in such words as honour, pity, emperor, justify, which in Late ME was only occasional, has now become fixed. Even in the present century many of these words have thrown back their stress to the first syllable, such as balcony, crystalline, recondite, which in the last century were stressed on their second syllables. 254. Native words which had weak stress on the first syllable in OE and ME, such as arise, become, forgive, to-day, still keep this stress in MnE, as also those French words which preserved a similar stress in ME through their resem- blance to the above native words, such as avow, defend. 255. Many other foreign words have also preserved their advanced stress. There are many foreign derivative endings chiefly Greek and Latin, often modified in their passage through French which regularly take the stress, such as -esque, -tion, -sion etc., -bility, -graphy, as in picturesque, grot- esque, imagination, position, possibility, photography, in all of which the stress is taken away from the root-syllable, on which it falls in the shorter forms imagine, possible, photo- graph etc. Many words which were imported from French and other foreign languages in the MnE period keep their advanced stress even when the analogy of other words points to throwing it back on the first syllable, such as machine, caprice which show their French origin by the pronuncia- 258.] PRESENT ENGLISH: STRESS. 69 tion of i as (ij) champagne, canoe, gazelle. Words which were imported straight from Latin generally keep the Latin stress, as in pa-pyrus, even when the final syllable is dropped, as in cre'ate, severe. Words of Greek origin follow the Latin accentuation as well as the Latin spelling, so that the original Greek stress is preserved in English only when it happens to be preserved in Latin also, as in genesis, museum = Greek ge'nesis, mouseion. 256. But foreign words even of recent introduction are always liable to have their stress thrown back on to the first syllable, or, at any rate, towards the beginning of the word, as soon as they become popular, which in Latin words is generally shown by their shortening or dropping their endings, as in auditor = Latin auditor, disciplme=~Lz\.m discip'lina, phi-losophy= Latin phik'sophia from Greek philosophid. 257. When a foreign word is used in different senses, it often happens that in its more familiar meaning it throws the stress back, keeping the original stress in the less familiar meaning. Thus we keep the original Latin stress in the adjective august and the name Au'gustus="Lztm au-gustus, but throw it back in the month-name 'August. So also the adjective mrnute keeps its Latin stress, which is thrown back in the more familiar noun 'minute. 258. In many cases where the same foreign word is used both as a noun and a verb in English, it keeps its end-stress when used as a verb by the analogy of the native verbs which have the same stress, while the corresponding noun- or adjective-form takes the stress on the first syllable, so that the distinction between such words as the noun -accent and the verb to accent is really ultimately due to the analogy of the OE pairs 'forwyrd, forweorpan etc., which analogy was greatly aided by the fact that many verbs of French and 70 PHONOLOGY. [ 259. Latin origin also threw forward their stress; thus the con- trast between the foreign verbs tn-duce, invade etc. and the native nouns 'income, -insight etc. led to the distinction between the noun 'insult and the verb in-sult from Latin insultare. The following are additional examples of such pairs : 'absent to absent 'abstract to ab'stract 'affix to a'ffix 'object to ob'ject 'present to pre'sent 'compound to coni'pound 'extract to ex-tract frequent tofre'quent 'Produce to pro'duce 'rebel to re'bel In some cases, however, the noun- and adjective-forms keep the verb-stress, as in ad-vice (to ad-vise), ce-ment. 259. The normal stress of a word is always liable to be changed by considerations of emphasis, even a weak word or syllable being capable of taking strong stress if emphasized, as in that is -the thing to do, especially in cases of contrast, as in to give and -forgive, not -subjective but -objective, against the normal stress forgive, subjective, ob-jective. Quantity. 260. In MnE there is a general tendency to shorten long vowels. As we have seen (222, 224), long vowels are often shortened before certain consonants in native words, as in blood (bhed)=OE and ME blod. 261. There is also a tendency to shorten long vowels or keep strong short vowels from being lengthened when followed by a single consonant and a weak vowel, in words of French origin, whether popular or learned, as in cavern, cavity compared with cave ; gratify, gratitude compared with grateful; perish, method, benefit, relative, astonish, philosophy. 265.] PRESENT ENGLISH: QUANTITY. 71 astronomy, pleasure (pi 639) compared with please, courage (kBrid5), flourish. 262. But when the consonant is followed by two weak vowels, the preceding strong vowel is often lengthened, as in atheist, radiant, patient, tedious, especially in the derivative endings -tion, -sion, etc., preceded by a strong vowel, as in nation, admiration, adhesion, notion, corrosion, although i is not lengthened under these circumstances, as in hideous, petition. Short vowels are also preserved when the two short vowels are preceded by certain consonants, such as n and sh, as in companion, fashion. 263. There is also a variety of other exceptions, especially before certain endings, such as -al, -ive, -y, -n and -r preceded by weak vowels, as in fatal, decisive, navy compared with navigate, bacon, paper, labour, those in -n and -r being probably the result of the influence of native words, such as the preterite participles taken, shaken, etc., and the numerous derivatives in -er, such as maker. 264. But some of these words with long vowels shorten them when another syllable is added, as in national compared with nation, tyrannous compared with tyrant. 265. In words which have been imported direct from Latin and Greek, the vowels are generally long under the circumstances described above, as in basis, ether, regent, crisis, focus, strophe. But there are several exceptions, such as simile (simili), chemist, the quantity varying in some words, such as pathos (peibos, pae]x>s). ACCIDENCE. NOUNS. Old English. GENDER. 266. THERE are three genders of nouns in OE mascu- line, feminine, and neuter. The genders of nouns are most clearly shown by the accompanying definite article 'the'- masculine se, feminine seo, and neuter fiat. The gender is partly natural, partly grammatical. It is to be noted that by natural gender names of children and young animals are neuter : pat cild ' child,' fiat cealf ' calf.' In the same way diminutives are neuter : fiat magd-en ' maiden,' ' girl.' Names of things and abstractions are often neuter, but as often masculine or feminine : fiat heafod ' head ' ; se A/re ' army ' ; seo wynn ' joy.' Names of living beings sometimes have a grammatical gender which contradicts the natural gender ; thus fiat wif ' woman,' ' wife ' is neuter. 267. Compound nouns follow the gender of the last element. Hence se wifmann ' woman ' is masculine, because se mann ' human being ' is masculine. STRONG AND WEAK. 268. All nouns belong to one of two classes strong and weak. Weak nouns are those which inflect mainly with -, such as se steorra 'star/ plural nominative steorran. All 273-1 NOUNS: OLD ENGLISH. 73 others are strong, such as se stan ' stone,' plural nominative stanas. CASES. 269. OE nouns have four cases, nominative, accusa- tive, dative, genitive, which are not always clearly dis- tinguished. DECLENSIONS. The following are the regular noun-declensions : Strong Masculine. Sing. Plur. 270. Nom. 1 stan stanas Dat. stane stanum Gen. stanes stana Sing. Plur. gnde l end ' gndas Sing. Plur. Sing. 271. Nom. has hits sap ' ship ' Dat. huse husum scipe Gen. huses husa scipes $nde gndum gndes gnda Strong Neuter. Plur. scipu scipum scipa 272. Some neuters have a plural ending -ru, such as aid, plural a'ldru, a'ldrum, cildra. The neuter plural ending - is dropped after a long syllable, that is, one containing a long vowel, as in hits ' houses,' or containing a vowel followed by more than one consonant, as \nfolc ' nations.' Strong Feminine. 273. Sing. Plur. Nom. caru ' care ' cara Ace. care cara Dat. care carum Gen. care carena Sing. Plur. synn ' sin ' synna synne synna synne synnum synne synna 1 Wherever the accusative is not given separately, it is the same as the nominative. 74 ACCIDENCE. [ 274. 274. The -u of the nom. sing, is, like the -u of the neuter plur. nom., kept only after a short syllable. Weak Masculine. 275. Sing. Plur. Nom. nama ' name ' naman Ace. naman naman Dat. naman namum Gen. naman namena Weak Neuter. Weak Feminine. Sing. Plur. Sing. Plur. Nom. eage ( eye ' eagan cirice ' church ' cirican Ace. eage eagan Cirican diriian Dat. eagan eagum cirican ciricum Gen. eagan eagena cirican ciricena 276. There are besides a number of irregular strong nouns. The most important of these are the mutation- nouns, such as the masculine mann ' man/ fot ' foot,' top 'tooth/ plur. mgnn, fet (fee?), tep, the feminine gos 1 goose/ mus ' mouse/ plur. ge s, mys. Masculine Mutation-nouns. 277. Sing. Plur. Nom. mann mgnn Dat. menn mannum Gen. mannes manna Sing. Plur. fot fet fet fotum fotes fota Feminine Mutation-Nouns. Sing. Plur. 278. Nom. mus mys Dat. mys musum Gen. muse musa 279. The relationship-words in -er, -or, such as fader ' father/ modor ' mother/ brofior ' brother ' are partly regular, partly indeclinable, the dat. sing, generally having mutation : 286.] NOUNS: OLD ENGLISH. 75 Sing. Plur. Nom.ftzder fcederas Dat. feeder feeder urn Gen. fcederjcederes fczdera Sing. Plur. brobor brobor, brobru briber brobrum brobor brobra 280. Some nouns are indeclinable, such as the abstract fern, nouns in -u, such as ieldu ' old age,' strgngu ' strength.' The fern, nieht 'night' is indeclinable in the sing, and in the nom. plur., the masc. monab 'month' being also inde- clinable in the nom. plur. ; we still preserve these unchanged plurals in the compounds fortnight = OE feowertiene nieht ' fourteen nights ' and twelvemonth. The inflection of nouns is attended by various modifications which fall under the general head of OE sound-changes : 281. Nouns ending in weak -el, -ol, -en, -er, etc. often drop their vowel before an inflection beginning with a vowel, thus se fugol ' bird/ pat wapen ' weapon ' have plurals (nglas,fuglas, wapnu. 282. For the change of ce into a in such nouns as se dag ' day,' gen. sing, dages, plur. nom. dagas, bat dal ' dale/ ' valley/ gen. sing, dales, plur. nom. dalu, see 127. 283. For the dropping of h in such nouns as se Wealh ' Welshman/ plur. Wealas, see 141. 284. In Late OE final h and medial g alternate in such words as se troh (earlier OE frog], plur. trogas, seo burh, gen. sing, burge (143). 285. Final -u in the nom. sing, of some nouns, such as bat meolu ' meal/ seo sceadu ' shadow/ ' shade/ seo sinu ' sinew ' is a weakening of original w, which reappears before an inflection beginning with a vowel, as in the gen. sing. meolwes, sceadwe, sinwe. This -u is dropped after a long syllable, as in seo mad ' meadow/ plur. madwa. 286. The dropping of h before vowels (141) leads to contraction, as v&bat feoh 'money/ gen. 76 ACCIDENCE. [287. Early Middle English. 287. In Early Southern the old gender-distinctions in nouns were still partially kept up. By degrees, however, the inflections of the adjectives and the definite article were dropped ; and when the Earliest Southern J>e, fieo, /// were levelled as they soon were under the uninflected /eo sunne=OTL seo synn and^>// hits became^ su'nne, fie hus, the old genders were gradually forgotten, simply because there was nothing to mark them. 288. The first great change in the old system of inflec- tions was the levelling of weak vowels under -e (174). By this change the distinctions of gender in the OE weak forms mona, sunne, cage were levelled in the Early Southern forms mone, sunne, eie as far as the endings were concerned. The distinctions of case were almost entirely effaced by this change in such words as OE caru, ace., dat., and gen. sg. care, nom. plur. cara. So also the inflections in OE stane (dat. sing.), sldna (gen. plur.), scipu (nom. plur.) were levelled under the same final -e. 289. The only endings which could withstand this level- ling were the gen. sing, -es, the nom. plur. -as, which both became -es in ME, as in stjnes=Qi}L stanes, s tanas, the weak -an, which became -en, the gen. plur. -ena, which became -ene. The dat. plur. -um became -em] but as this was the only case ending in m, the consonant was levelled under the more frequent n, so that ME -en represented OE -um as well as -an, as in iveren = Qi}L geferan, geferum. 290. The general result of these changes was not only to obscure the distinctions of the cases, but also in some classes of nouns to obscure the distinction between singular and plural. The confusion was most marked in the feminine 293-1 NOUNS: EARLY MIDDLE ENGLISH. 77 nouns, where the changes we have been considering gave the following as the endings corresponding to those of the OE nouns caru, synn, sunne respectively : Sing. Nom. -e -* -e % Ace. -e -e -en* Dat. -e -e -en* Gen. -e -e -en* Plur. Nom. -e* -e* -en Dat. -en -en -en Gen. -ene -e -ene 291. It is evident that the forms marked * in the above table are in the minority, while at the same time most of them obscure the distinction between singular and plural. They were accordingly got rid of by the analogical extension of those forms which were in the majority and more distinctive. The -e of care and sunne was extended to the OE nom. synn, which became ME siinne. The plural -en of sunnen^OE sunnan was extended to all feminine nouns ME car en, sunnen=O^ cara, synna. As -en was now the distinctive mark of the plural, it was given up in the singular of sunne, whose oblique cases took the same form as the nominative, as in the other two classes. The final result was that all feminine nouns were uniformly declined as follows : Sing. Plur. Nom. -e -en Ace. -e -en Dat. -e -en Gen. -e -ene 292. As might be expected, the gen. plur. -ene was often levelled under the other plural cases, becoming -en. 293. Weak masculines and neuters were declined in the same way sing, name, eie, plur. namen, eien. The only 78 ACCIDENCE. [ 294. distinction between masculine and neuter weak nouns namely in the ace. sing. (OE naman, eage) was thus lost. 294. -f=the OE neuter plur. ending -u was made into -en for the sake of distinctness, as in deoflen, children QfiL deoflu, cildru, sing, deovel, child. In many of these words -^=OE - was extended to the singular, as in dale 'valley/ bede ' prayer,' =OE d(zl,gebed, plur. dalu, gebedu. These OE plurals became dalen, beden in ME. 295. The remaining masculine and neuter nouns kept their original strong forms. The dat. sing in -e was kept at first, but often dropped, because such forms as zvei'e, worde OE wege, worde suggested a weak singular, and so the dat. sing, was levelled under the nom. in such words wet, word in accordance with the general ME tendency. The dat. plur. -i?=OE -um was disused for a similar reason -because it suggested a weak plural. The gen. plur. -e = OE -a was sometimes kept, but the more distinct weak ending -ene was often used instead h'ngene, as in alre kingene king 4 king of all kings,' wordene instead of kinge, worde both of these forms being gradually supplanted by the nominative. In the neuter plur. the OE undeclined forms were still kept hus, word but the strong masc. ending was often extended to the neuters, so as to distinguish the two numbers hftses, wordes. The following are then the regular Early Southern ME noun-inflections, those which are liable to be dropped being in( ):- Strong Masculine and Neuter. 296. Sing. Nom. stpn -word Dat. st(>n(e) word(e) Plur. stgnes word, wordes stgnes word, wordes Gen. stones wordes styne(n<;\ signet worde(ne), wordes 3oa.] NOUNS: EARLY MIDDLE ENGLISH. 79 297. The neuter child has plur. children, corresponding to OE cildru. Strong and Weak Feminine. 298. Sing. Plur. Nom. sunne, chirche Dat. sunne, chirche Gen. sunne, chirche sunnen, chirchen siinnen, chirchen sunnen(e), chirchen(e) Weak Masculine and Neuter. 299. Sing. Plur. Nom. mere eie iveren eien Dat. ivere eie iveren eien Gen. ivere eie iveren(e) eien(e) 300. The relationship-words vader, moder, stister gener- ally remained unchanged in the sing., having the regular plurals vaderes, modren, siistren. brober of course lost the OE mutation in the dat. sing., which became broker. But this mutation was transferred to the plur. on the analogy of fet, men, etc., so that brobre-=Qfe. brobru became brebre, and then, by the usual change of plural -e into -en, brebren. 301. Final e was dropped after a weak vowel, as in Igfdi ' lady ' OE hl&fdige. The plural ending -s without a vowel occurs only in long French words, as in parlurs ' parlours,' vestimenz 'vestments/ where z=(ts). In Old French such a word as vestiment is inflected thus Sing. Nom. vestimenz Plur. Nom. vestiment Ace vestiment Ace. vestimenz As the distinction between nom. and ace. had been lost in ME., the French -s was naturally identified with the English plur. inflection -es. 302. In Early Midland and Northern the distinctions of grammatical gender were entirely lost during the transition from OE, the distinction between strong and weak forms 80 ACCIDENCE. [ 303. being also done away with, except in a few isolated forms. The natural consequence was that the -es of the genitive was extended to weak nouns and to all feminine nouns, the plur. -es being then extended in the same way, first to strong neuters, then to weak nouns and feminine nouns generally. The final result was that the only regular inflec- tions left were gen. sing, -es, plur. nom. and gen. -es, the distinction between nom. and gen. plur. being kept up only in irregular plurals such as men, gen. mennes. Late Middle-English. 303. Standard ME follows the Early Midland dialect in its noun-inflections : it has only one case, the genitive ; the original nominative, accusative, and dative being now merged in one ' common case ' : Sing. Common word, sinne Gen. wordes, sinnes Plur. Common wordes, sinnes Gen. wdrdes, sinnes man mannes men mennes 304. The e of -es the gen. as well as the plur. ending is often dropped in English as well as French words after a weak syllable, as infdders (alsofddres), Iddys (also Iddyes), and after a strong vowel, in order to avoid hiatus, as in fgs ' foes.' Also in pens = earlier penies, of which pens was originally the weak form, the word having lost its stress in such combinations as iwg penies. 305. The whole ending -es is often dropped in French words and proper names ending in a hiss-consonant, as in the gen. sing. Troilus, Vjnus, and the plurals cds 'cases,' vers (also verses'). This is the result of French influence, for in Old French such 3io.] NOUNS: MODERN ENGLISH. 8 1 a word as vers, whose s is part of the body of the word, was necessarily indeclinable : Sing. Nom. vers Plur. Nom. vers Ace. vers Ace. vers 306. Originally feminine nouns sometimes keep their earlier j-less gen. sing., as in pe chirche dgre, his lady grace. We still preserve this form in Lady-day compared with Lord's day. 307. Many originally neuter nouns with unchanged plurals still keep these, such as folk, der, hors, shep, swm, pound. It must be observed that most of these plurals have a collective meaning ; thus the plur._/0/ is oftener used in the sense of ' people in general ' than in that of ' nations/ and in MnE swine is used exclusively in the collective plural sense, not being used in the singular at all. fot when used as a measure was made invariable in the plural on the analogy of the old neuter pound, and the invariable night in fburte-night (280), etc. 308. In its general meaning fot keeps its mutation-plural fet. So also man, wom(ni)an (OE wifmann), top, etc. have plurals men, wom(ni)en (OE wifmgnri), tep, etc. 309. The weak plural-ending -en is preserved not only in oxe plur. oxen, but also in other words which have now lost it in the spoken language, such as asche, aschen, hgse, le ' eye,' len, fg ' foe,' fgn, tp ' toe,' tgn, scho ' shoe/ schdn. In other words this ending is a ME extension, as in brepren, children, dohtren, sustren. cow has plur. /$j/#=OE cu, plur. cy, the northern dialect keeping the older form ki. Modern English. 310. By the beginning of the MnE period the j of inflectional -es had been voiced (232), (s) being kept 82 ACCIDENCE. [3ii. only in monosyllables such as geese, pence. In Early MnE the e was kept after a hiss-consonant for the sake of distinctness, as in horses (horsez), and was dropped every- where else, the (z) being necessarily unvoiced after voice- less consonant, as in beasts (twsts) from beastes (b^stez), while it was of course preserved after vowels and voiced consonants, as in days, heads (hmiz). 311. The ME dropping of -es after hiss-consonants is still kept up in a few phrases such as for old acquaintance sake, for Jesus sake; but in the spoken language the -es is generally kept, as in Si. James's Square, where it is also written. Such genitives as ^Eneas', Socrates' wife occur only in the literary language ; in the spoken language the full -es is added, or else the construction of JZneas etc. is used. One result of the contraction of inflectional -es in MnE is that radical s has been sometimes mistaken for the plural inflection, so that an original singular has been made into a plural, as in the case of alms, eaves, riches, summons : these ' apparent plurals ' correspond to the OE singulars celmesse, ejese (plur. gfesan] and the Old French singulars richesse, semonse. Most of these apparent plurals are not used in the singular ; but summons is used in the sing, without any change a summons. There are some plurals which form a curtailed sin- gular by throwing off the radical final s. Thus the collective plural pease= the OE weak plural piosan has developed a sin- gular pea, whence a new orthographic plural peas has been formed. Inflectional plurals often come to be used as singulars by change of meaning, such as news, sixpence. They may then form new plurals, such as sixpences. 312. The ME (and OE) alternation of breath and voice consonants in the inflection of such native words as wif, 314-1 NOUNS: MODERN ENGLISH. 83 gen. sing, wives, plur. wives, has been kept up only partially in MnE. It has been entirely abandoned in the gen. sing., which is now formed afresh from the common case wife's. We still keep the voice consonant in such plurals as wives, paths (paaftz), but such a plural as the earlier MnE turves has been made into turfs. We still keep the gen. sing, calves in the compounds calves- head, calves-foot expressing articles of food ; otherwise ^vz^has the regular gen. sing, calf's. The following are the main types of noun-inflection in Present English : 313. Sing. Common hos dog kaet waif guws maen Gen. hosiz dogz kasts waifs guwsiz masnz Plur. Common hosiz dogz kaets waivz gijs men Gen, hDsiz dogz kaets waivz gijsiz menz Like horse are inflected words ending in the hisses (s, z ; J, 3), such as piece, box, size, adze, fish, church (tjaatf), age (eids). Like dog are inflected nouns ending in a vowel or any voiced consonant except (z, 5), such as day, lady, neighbour (neiba), mile, dove, son, lord. Dice (for gaming) and pence, the plurals of die and penny have (s) because they were shortened to monosyllables already in ME, dies (for coining) and pennies being new-formations from the singulars on the analogy of the regular plurals days, ladies, etc. Like cat are inflected nouns ending in any breath con- sonant except (s, J), such as earth, cliff, clerk, bishop. 314. All the nouns inflected like wife 'voice-breath nouns ' show a long syllable* before the inflection in Late ME, as in staves =1^.^ ME staves (Early ME slaves), wolves G 2 #4 ACCIDENCE. [ 315. = ME waives. Hence nouns with original short i never make this change piths (pips), cliffs. The only voice- breath noun ending in (s) is house, plural houses (hauziz). The chief voice-breath nouns in (p) are bath (baap), baths (baat>z)=Late ME bab, babes (bap, baaSes), path, oath, mouth, clothes was originally the plural of cloth, which now forms a regular plural of its own cloths. The great majority of nouns in (J>) keep the breath-sound in the plural ; such nouns are moth, death, hearth, health, birth. Some, such as lath, truth, youth, have both pronunciations, that with voice consonants in the plural being, of course, the older one. Nouns in -f show the change more frequently : after long Late ME vowels, as in life, knife, wife, thief, leaf, loaf; after /, as in half, calf, elf, self, shelf, wolf. Nouns in -rf, such as dwarf, scarf, turf, wharf, made this change in Early MnE dwarves, etc. but they now generally keep they in the plural dwarfs, etc. Nouns in -oof also keep the f, as in hoofs, roofs. So also belief. But the French noun beef still keeps its plural beeves, which, however, is now iso- lated from its singular, through the latter having lost its original meaning 'ox.' staves was originally the plural of staff (Lz.te ME staf, slaves), but having diverged from it in meaning, it has now developed a new singular stave, while staff itself has developed a new plural staffs, as in army staffs. IRREGULAR PLURALS. 315. The following mutation-plurals are still in common use : man, men ; woman, women (wumgn, wimin), this plur. being Southern in spelling, though Midland in pronunciation ; foot, feet; goose, geese ; tooth, teeth ; louse > lice ; mouse, mice. 320.] NOUNS : MODERN ENGLISH. 85 316. The only n-plurals in common use are ox, oxen child, children, brother now has the regular plural brothers, the old plural brethren being used only in a metaphorical sense, cow also has a regular plural cows, the older kine occurring only in the higher literary language. 317. sheep and deer keep their unchanged plurals. 318. These are the only absolutely invariable words. In all other invariable words the unchanged plural implies either measure or collectiveness. As in Late ME, so also in MnE many nouns of measure have an unchanged plural only when preceded by a numeral, as in two dozen knives compared with dozens of knives ; and many of them keep it only in groups or compounds such as ten-pound note compared with ten pounds, the earlier MnE ten pound being now obsolete or vulgar. It is only when a noun of measure is used also as an ordinary descriptive noun that it occasionally keeps its unchanged plural under all circumstances, as in how many stone does he weigh? 319. While the use of the unchanged plural of measure has been gradually restricted in MnE, the unchanged collec- tive plural has been extended, swine has now lost its sin- gular, the sing, and separative plur. being expressed by/?^, pigs. But in most cases the collective and separative plurals are used side by side, as in to catch fish compared with the story of the three fishes. FOREIGN PLURALS. 320. Many foreign words especially Latin and Greek keep their original plurals, but some of them have also regular English plurals ; some have the two plurals in different meanings. Some are used only in the plural. Some are unchanged in the plural. 86 ACCIDENCE. [321. 321. The most important Latin endings are : -a ... -se : formula, formula. -us . . . -\: fungus, fungi. -um . . . -a : desideratum, desiderata. -is ... -es : analysis, analyses. -es .... -es : species. -ix, -yx, -ex . . . -ices : appendix, appendices ; calyx, calyces ; vortex, vortices. There are other isolated Latin plurals : genus, genera ; stamen, stamina. 322. -on . . . -a is a Greek plur. : phenomenon, phenomena. 323. We have Italian plurals in bandit, banditti ; dilettante, dilettanti', virtuoso, virtuosi. 324. The Hebrew plurals cherubim, seraphim are collec- tive, and are occasionally used as singulars in Early MnE a cherubim. 325. The French plural ending x in beaux, flambeaux, has the same sound as the regular -s. 326. The plural of Mr. is expressed by the different word Messrs., in full Messieurs. Mr. is a weak form of ME meister from old French meistre, the corresponding strong form being master. Messieurs is the French mes Sieurs 'my Lords/ the sing, of which is Monsieur. The plural of the feminine Madam = French ma Dame 'my Lady' is Mesdames French mes Dames ' my Ladies/ which, however, is not much used in English. 327. The tendency of the language now is to get rid of foreign plurals as much as possible, except where the foreign plur. marks a difference of meaning. Many words which have foreign plurals, form their plurals also regularly, some- times with a distinction of meaning, as in appendixes and appendices. 332.] ADJECTIVES. 87 ADJECTIVES. Inflections. OLD ENGLISH. 328. In OE the adjectives have the three genders of nouns, and the same inflections, though with partially different forms, together with the distinction of strong and weak. Adjectives (as also pronouns) also show traces of an instru- mental case, which is, however, generally expressed by the dative, 329. Adjectives agree with their nouns in gender, number, and case : hie comon mid langum scipum, nd mam'gum ' they came with long ships, not many.' 330. The weak form is used after the definite article and other defining words, as in se goda cyning ' the good king/ se hdlga ' the holy (man),' whence the weak masc. noun hdlga ' saint,' pas hdlgan cyningas ' these holy kings,' compared with sum god cyning ' a certain good king,' hdlge mgnn ' holy men.' 331. The following are the strong inflections of god, the forms which differ from those of the nouns being marked * : Masc. Neut. Fern. Sing. Nom. god god god Ace. godne* god gode Dat. godum* godum* godre' Instr. gode gode gode Gen. godes godes godre Plur. Nom. gode* god gode* Dat. godum Gen. godra* 332. The weak forms are identical with those of the weak ACCIDENCE. [ 333* nouns, except in the gen. plur., which, however, sometimes appears as -ena with the same ending as in the nouns, instead of taking the ending of the strong adjectives : Sing. Nom. Ace. Dat. Gen. Plur. Nom. Dat. Gen. Masc. goda godan godan godan Neut. gode gode godan godan Fern. gode godan godan godan ^Y" godan godum godra* 333. The -u of the strong fern. nom. sing, and the strong neut. nom. plur. is kept under the same circumstances as in the noun-inflections; thus sum ' some' has sumu in the above cases, as opposed to the long-syllable god. Adjectives in -el. -en, etc. drop the e as in noun-inflection ; thus hdlig, micel, dgen ' own,' have plurals halge, micle, dgne. Where final -u is a weakening of -w, the w is restored before an inflection begin- ning with a vowel, as in nearu ' narrow,' salu ' sallow,' geolu ' yellow/ plurals nearwe, salwe, geolwe. In late OE final -h alternates with medial g in such forms as genoh ' enough ' [earlier genog~\, plur. genoge. The dropping of weak h between vowels leads to contraction ; thus heah ' high,' Mercian heh, has plural hea (from heahe). MIDDLE ENGLISH. 334. The levelling of noun-inflections in ME and the loss of gender distinctions naturally led to the disregard of concord. Hence the case-endings in the singular of strong adjectives began to fall off at the beginning of the ME period. The distinction between singular and plural and between strong 338.] ADJECTIVES. 89 and weak inflection was preserved in the adjectives as well as in the nouns, god represented the strong singular, gode the strong plural and the weak singular. As the weak form of the adjective was generally followed by a noun, it was superfluous to mark the distinction of number in the adjec- tive, and consequently the weak singular ending -e was used also in the plural. The result was that in Late ME the adjective had only two inflections : Strong Sing, god Weak Sing, gode Plur. gode Plur. gode 335. The weak form is used much as in OE : pe yonge sonne ' the young sun,' pis ilke monk ' this same monk/ 336. In the Northern dialect all adjectives became in- declinable already in the Early period through loss of final weak -e. 337. The old cases were partially preserved in the Earliest ME. The gen. plur. ending -re=OE -ra, as in alre kingene &>/= OE eallra cyninga cym'ng, lingered longest, because of its distinctiveness. In Late ME alder, from earlier alre through aldre, became a sort of prefix to superlatives, as in alderbest ' best of all ' ; in Early MnE Shakespere still has alderliefest ' dearest of all.' MODERN ENGLISH. 338. In MnE the loss of final -e made the adjectives in- declinable as far as case and number are concerned. Adjec- tives thus became formally indistinguishable from adverbs, except by their syntactical relations, the only change of form that was left to them namely comparison being shared by adverbs. But Early MnE still preserved a trace of the ME 90 ACCIDENCE. [ 339. inflections in the distinction between enough sing., enow plur. = ME inoh, inowe. Comparison. OLD-ENGLISH. 339. In OE the comparative is formed by adding -ra, and is declined like a weak adjective, as in leof-ra ' dearer' masc., leofre fern, and neut., the corresponding adverbs ending in -or : leofor, heardor. The superlative is formed by adding -osi, and may be either strong or weak : leofost ' dearest/ se leofosta mann. The uninflected form of the superlative is used also as an ad- verb : leofost, heardost. Some adjectives form their comparison with mutation, the superlative ending in -est, as in lang ' long/ Igngre, longest, neah ' near/ superlative niehst, riiext (Anglian neh, nest, next}. In some comparisons the comparative and superlative are formed from a word distinct from that which constitutes the positive : god [adverb wel\ bgtera [adverb bgt], bgtst. The positive of some comparatives and superlatives is represented only by an adverb ; thus to arra ' former ' (in time) cerest ' first ' corresponds the adverb ar ' formerly.' Many of these form the superlative with -m, which is an older form than -s/. The original form of this superlative is seen in for-ma ' first/ the positive of which is represented by the adverb fore ' before.' But in most cases the meaning of this old superlative ending was forgotten, and the ending -st was added generally with mutation giving the double superla- tive -mest. Thus from forma the new superlative fyrmest ' most foremost/ ' first ' was formed. Other examples are tnnemest, norfimest from inne ' inside/ norp ' north/ 344-1 ADJECTIVES: COMPARISON. 91 MIDDLE-ENGLISH. 340. In Early ME the endings are -re, -ere [adverbial -ii he hit heo Ace. meC (me) pec (f>e) hine hit hie (hy") Dat. me pe him him hire Gen. mln pm his his hire Plur. Nom. we ge hie (hy, heo} Ace. usic (us) eowic (eow) hie (hy, heo) Dat. us eow him (heom) Gen. ure Sower hira, heora Sing. Nom. hwa hwat Ace. hivone hwcet Dat. hwcem (hwdm) Gen. hivces Instr. hivy 360. The change of the plur. him into heom is the result of the influence of the gen. plur. heora together with the H 98 ACCIDENCE. [ 361. desire to distinguish between singular and plural. The late nom. plur. heo is the result of levelling under heom and heora. 361. Many of the above inflections had weak forms, in which long vowels were shortened, such as weak J>u, heo= strong/*, heo. MIDDLE-ENGLISH. 362. In ME the genitive of the personal pronouns was gradually restricted to the function of a possessive pronoun, though it still retained something of its independence in such phrases as oure aller hfa ' the salvation of us all ' = OE ure eallra ha>lu. 363. In ME the distinction between accusative and dative was done away with, these two cases being levelled under one which we call the ' objective ' case, this objective case being really the old dative used also as an accusative. This extension of the dative began already in OE, me, fie, us, eow being the regular accusatives even in Early West-Saxon. The explanation is that as the personal pronouns generally refer to living beings, we naturally think of ' I,' ' you/ etc. not as mere passive objects of striking, calling, sending, etc., but as being to some extent actively interested in these pro- cesses ; and hence we are inclined to use the interest-case or dative to express the personal complement even of purely transitive verbs. Hence even in OE they began to say he slog me ' he struck me ' instead of he slog mec in the same sense as he slog pone stan 'he struck the stone/ but from a different point of view. In ME the change was carried out consistently, him supplanting hine and so on. But with the specifically neuter -pronouns the process was reversed : *'/ and what being mainly thought of as passive complements of verbs, not only kept their old accusatives which was 367-1 PRONOUNS: MIDDLE ENGLISH. 99 made still more easy by these accusatives having the same form as the nominatives but used them to express the much rarer relation of interest, and so the old accusative *'/ has come to represent the dative as well as the accusative in MnE, while the old dative him serves as accusative as well as dative. 364. In ME as also in OE all the third person pro- nouns had weak forms wiihout h- : ur emphatic or weak im by the side of emphatic or strong him, although in writing only the emphatic form was used, just as in MnE we write / saw htm, whether the htm is emphatic (him) or weak (im). But even in the earliest Midland we find it written every- where by the side of he, etc., showing that this originally weak form had supplanted the strong one. The reason is that it was so rarely necessary to emphasize the impersonal pronoun that the strong hit was forgotten and disused. But hit was preserved in South-Thames English up to the end of the Late ME period. 365. OE id split up into the two forms ich (North-Thames ic) and t. The latter which was, of course, originally the weak form gradually supplanted the fuller form, which became extinct in Standard ME, although it still survives in the dialects of the West of England. 366. So also the weak us (with short vowel) gradually supplanted the strong us, ous. 367. In ME weak eo often became a through intermediate ea. Already in Early Old-Anglian we find weak heara by the side of the older strong heora. In Early ME heara passed into hare, and in the same way Late OE heom ' them ' became ham. The weak OE heo ' she,' which in Late OE also expressed ' they,' passed through the same change, be- coming ha. This weak ha was then extended to the masc. H 2 100 ACCIDENCE. [ 368. sing. So in Early Southern we find the strong and weak pairs with a in the latter : he (ha) ' he/ heo (ha) ' she,' ' they,' heom (ham) ' them/ heore (hare) ' their.' ha was liable to drop its h by still further weakening, whence the Early MnE a=he in quoth' a, quotha, 'a must needs. 368. Strong heo ' she ' passed through heo into (hjoo, j^oo), which last is the Early Midland form, written -%ho parallel to wha ' who'=(w^aa). But the feminine demonstrative seo ' that one/ ' she ' gradually took the place of heo, at first in the Midland dialect, and then in the Standard ME. seo passed through seo (sjoo) into sho in some dialects with the change of (sj) into ([). This sho, being a weak form, existed side by side with the strong seo, and in some Midland dialects the two were blended together into a new form sheo, which became she by the regular change of eo into e. Strong heo was soon discarded, because this vowel-change levelled it under the masculine he. 369. eow in its weak form passed through (joow) into (juuw), written %uzv, which then became _y0ze;=(juu), the (w) first changing the o into u, and then being itself absorbed by the u. Early Southern has ou with dropping of the ow, }>e by imitation of Old French. 102 ACCIDENCE. MODERN ENGLISH. 375. In Early MnE the use of the ceremonious plural ye, you was so much extended that it became the usual polite form of address, the singular thou being used mainly to ex- press familiarity and contempt, which latter use brought about its complete disuse in the spoken language of the pre- sent century, which therefore makes no distinction of number in the personal pronoun of the second person. But we still preserve the old thou in the poetical and liturgical language. 376. In Early MnE the objective form you came to be used as a nominative, and in Present English you has com- pletely supplanted ye in the spoken language. The change is partly the result of a general confusion between nomin- ative and objective in MnE, partly of the influence of the singular pronoun thou. In Early MnE the ME fie, ye became (Sii, jii), which were shortened into (tSi, ji) when weak. So also ME bow, yow became Early MnE (Sou, jou) by the regular change of (uu) into (6u), the short (u) of the ME weak forms being necessarily preserved unchanged in the Early MnE (Su, ju). In Early MnE thou and ye were liable to lose their vowels before another word beginning with a vowel or A + vowel, so that thou art, ye are were shortened into th'art, y are, just as the earth was shortened into tH earth. This gave the following Early MnE forms of the second person pronoun : Nom. ('Sou, $u, $) (jii, ji, j) Obj. (ii, =Si) (jou,ju) 377. It will be observed that each of these pronouns has two groups of endings which have exactly opposite functions, (-6u), etc. being the nominative ending in the singular thou, 38o.J PRONOUNS: MODERN ENGLISH. 103 but the objective ending in the plural you, while (-ii) is the objective ending in the singular, the nominative ending in the plural. The natural result of this was that the associations between form and grammatical function became unsettled, and when ye, you came to be frequently used in a singular meaning, thou (Sou) and you (jou) were associated together, till at last you came to be regarded as a nominative. This confusion was increased by the shortened forms y'are, etc., in which it was impossible to know whether the y* was a contraction of ye or of you. 378. The phonetic similarity between thee and ye led to the frequent use of ye as an objective, especially in the weak form (ji), which was used indifferently as an objective or a nominative, being often further weakened by dropping the consonant, as in hark'ee, harkee, lookee, thankee. Such forms as I tell ye (ji) were still frequent a few generations ago, and (i) may still be heard in how do you do ? (hau d i duw), but such forms as (luki, J^genki) survive only as vulgarisms. 379. As (Sou) and (Su), (jou) and (ju) diverged con- siderably in sound, one member of each pair was got rid of in the course of the Early MnE period, namely the weak (Su) and the strong (jou), whose place was taken by (juu), a lengthening of the weak (ju). As this (uu) did not develop till after the change of ME (uu) into (ou) had been com- pleted, it was, of course, preserved from that change. 380. We have seen that the ending -e(e) in the second person pronouns is the mark both of the nominative (ye) and the objective (thee). The same cross-association runs through some of the other pronouns : Nom. : he, she, we, ye Obj. : me, thee 104 ACCIDENCE. [ 381. The fact that in four cases out of six -ee is the nom. ending explains ho\vj>e was able to maintain itself as a nom. in spite of the support given to you by the sing. thou. 381. Confusions between nominative and objective may occur in any language through misunderstanding gram- matical categories. Thus in the Bible we find zvhom do men say that I am f, where what ought to be the nominative is put in the objective through attraction through being regarded as the object of the verb say; and although OE is strict in its distinction between nom. and accusative, yet the OE version shows the same attraction : hwane sgcga)> mgnn pat sy mannes sunu ? But as long as a language marks the distinctions of case with clearness, such confusions are confined to isolated constructions. In MnE, however, the distinction between nominative and objective was marked only in a few words, and even there was marked in a way which inevitably led to confusion ; and even apart from this cross-association there was no uniformity : thus in the pairs /, me ; he, him ; we, us the objective cases have no formal characteristic in common. Hence in MnE the linguistic sense for the distinction between nominative and objective has been almost as much weakened as that for the distinc- tion between indicative and subjunctive. 382. In Early MnE the usage was more unsettled than it is now, the nominative being as freely substituted for the objective as vice-versa, as in such constructions as 'tween you and I. you and I were so frequently joined together as nominatives-^y0 and I will go together, etc. that the three words formed a sort of group-compound, whose last element became invariable. 383. The tendency of Later MnE is to merge the dis- tinction of nominative and objective in that of conjoint and 385.] PRONOUNS: MODERN ENGLISH. 1 05 absolute, that is, to keep the old nominative forms only when in immediate connection with a verb / am ; said he so that, as the pronouns in the nominative generally precede the verb, /, he, etc., are felt almost to be inseparable verb- forming prefixes, as in / c all, compared with to call. When a pronoun follows a verb, it generally stands in the objective relation ; hence, on the analogy of he saw me, tell me, etc., the literary it is I is made into it is me in the spoken lan- guage, so that me is felt to be the absolute form of the conjoint I, being also used as the answer to the question who is there?, etc. In the vulgar language this is carried out consistently, the slightest separation from the verb being enough to elicit the objective form, as in me and John came home yesterday =\h& polite John and I came home yesterday, them that is her e= they that . . In Standard spoken English the absolute use of the objective forms is most marked in the case of me, which is put on a level with the old nomin- atives he, etc. : it is me, it is he, it is she. But the usage varies, and in more careless speech such constructions as it is him, it is us are frequent. 384. The tendency to use the nominative forms before the verb has had the contrary effect on the pronoun who. Already in Early MnE whom do you mean ? was made into who do you mean ? on the analogy of / mean . . , you mean . . , etc. In Present spoken English whom may be said to be extinct, except in the rare construction with a preposition immediately before it, as in of whom are you speaking /=the more purely colloquial who are you speak- ing of? 385. The pronouns thou, thee and_>> are now confined to the liturgical and the higher literary language. In the singular the distinction between nom. thou and obj. thee is J06 ACCIDENCE. [ 386. strictly maintained. In the Bible ye is the nom. and you is the corresponding obj., but in the present language of poetry there is a tendency to usej 1 ^ in the obj. as well ^s the nom., in order to avoid the prosaicjw* : ye see, I see ye. 386. In Early MnE them which seems to be a weak form of ME }>eim finally got the upper hand of ME hem, which has survived only as a weak form, being written 'em from the mistaken idea that it was a shortening of them. We still use (am) as a weak form of them by the side of (tSam), but only in very familiar speech. 387. The MnE it, her are also equivalent to ME weak forms. 388. The ME weak ha occurs occasionally in Early MnE in the form of 'a, a, but only in very familiar, careless speech. 389. The following are the present forms of the personal pronouns : Sing. Nom. / Obj. me thou, you thee, you who whom what what Plur. Nom. we Obj. us ye, you ye, you Sing. Nom. he Obj. him it it she her Plur. Nom. Obj. ~^r they them ('em) 390. The shortening (-s)=us occurs only in let's. In Early MnE it was more general. 394-1 POSSESSIVE PRONOUNS. 107 Possessive Pronouns. OLD ENGLISH. 391. The OE possessive pronouns are the genitives of the corresponding personal pronouns : mm ' my/ ure ' our,' pin ' thy/ eower ' your,' his ' his, its/ hire ' her/ hira, heora ' their.' The possessives of the third person his, hire, hira together with hw&s ' whose/ are indeclinable, those of the first and second person min, }nn, ure, eoiver being declined like strong adjectives : mid his freondum ' with his , friends/ mid minum freondum. MIDDLE ENGLISH. 392. In ME his was made declinable on the analogy of mm, etc., that is, it took a plural ending -e, as in alle hise men compared with OE ealle his mgnn. This being the only inflection of the possessives in ME, those ending in -e necessarily remained or became indeclinable. The Early ME ozver, 5#r=OE eower took final -e in Late ME by the analogy of ure, becoming youre. 393. mm and>z dropped their final n before a consonant in Early ME mi fader keeping it before a vowel or h + vowel : mm arm, Jnn herie. In Late ME the n was often dropped before a vowel as well. The n was, of course, always kept when the possessives were used absolutely, or when they followed their noun : hit is min, broker mm / 394. In Late ME the possessives ending in -e generally take the genitive ending -.$ when used absolutely : to mm hous or to youres ; al J>is gold is oures=QE, to minum huse I08 ACCIDENCE. [ 395. oppe to eowrum ; eall pis gold is ure. This -s is an extension of the -s of his : his gold, pat gold is his. 395. In the weak forms long vowels were sometimes shortened min, mi and final e was dropped : hir, our, etc. 396. In North-Thames English pei brought with it the possessive peire= Icelandic peira, which gradually made its way into the London dialect, where it also appears in the weak form fore parallel \.o pem=peim. 397. The following are the possessive pronouns in Standard ME, weak forms being in ( ) : Conjoint : mm mi (min, mi) ; pin, pi (pin, pi) ; his (hes) ; hire, hir (her) ; oure, our ; youre, your ; here, her, peire, peir (per e, per). Absolute : mm ; pin ; his ; hires, hirs, heres, hers ; oures, ours; youres, yours; heres, hers. All those beginning with h were, of course, liable to lose it in their weak forms. 398. The Early ME possessive whas became whos in Late ME through the influence of who. MODERN ENGLISH. 399. In Early MnE his was still the possessive of it as well as he : it (the serpent) shall bruise thy head, and thou shalt bruise his heel (Bible). But. already in the Midland dialect of ME the want of a special possessive for // was supplied by using the uninflected it as a possessive instead of his ; and this usage appears also in Early MnE : the hedge- sparrow fed the cuckoo so long that it's had its head bitten off by it young (Shakespere). Towards the end of the Early MnE period the present genitive its came into general use 403.] PRONOUNS: SELF. 109 a form which does not occur at all in the Bible, and very rarely in Shakespere. 400. The ME distinction between conjoint mine, thine and my, thy was still kept up in Early MnE, but the shorter forms were frequently used before vowels : mine eyes, my eyes. In the higher literary language the distinction is still kept up : mine eyes, mine host. But many modern poets drop the n before sounded (h), as in my ^ar/= Early MnE mine heart, keeping it only before vowels and silent h + vowel, as in mine honour. 401. The following are the present forms of the possessive pronouns : Conjoint : my, mine ; thy, thine ; his ; its ; her ; our ; your ; their; whose. Absolute : mine ; thine ; his ; its ; hers ; ours ; yours ; theirs; whose. Self. OLD-ENGLISH. 402. In OE the emphatic self is added to nouns and personal pronouns, being generally inflected like a strong adjective in agreement with its head-word : God self hit geworhte ' God himself made it,' swd-swd hie cwddon him selfum 'as they said to themselves,' he f orgeat his selfes 'he forgot himself.' In the nominative the weak-inflected selfa is used in the same way : God selfa, ic selfa. 403. In OE the personal pronouns are used also as reflexive pronouns, as they still are in such phrases as he looked about him compared with he must take care of himself. OE self does not make a pronoun reflexive, but simply 110 ACCIDENCE. [404. emphasizes one that is already so, as in ivyscion him selfum the shorter wyscton him being enough to express the meaning ' wished for themselves.' Hence such a phrase as he ofsticode hine might mean either ' he stabbed him ' (some- one else), or ' he stabbed himself.' By degrees he ofsticode hine selfne, which at first meant both ' he stabbed that very man ' and ' he stabbed himself/ was restricted to the latter meaning, the simple hine, him, etc., being restricted more and more to the non-reflexive meaning, so that already in Early ME we find j^used very much as in MnE. 404. In OE a personal pronoun in the dative is often added reflexively to a pronoun in the nominative, but without materially affecting its meaning, as in he ondred him }>one mann ' he was afraid of the man,' literally ' feared for himself,' hie gewiton him ' they departed.' This pleonastic dative is often added to self, self a in this way : he bif> him self gehwoefrer, sunu and feeder ' he (the phoenix) is himself to-himself both (pronoun), son and father/ ic me self gewdt 'I myself de- parted/ he him self a sceaf reaf of lice ' he to-himself himself pushed the robe from the body'='he took off his robe.' MIDDLE ENGLISH. 405. In ME the meaning and function of the datives me and pe in the combinations ich me self, feu fee self, etc., were soon forgotten, so that these constructions became unmean- ing, which led to the change of me and fee into the posses- sives ml, }ri, self being regarded as a noun, as shown in such constructions as ml self havefe ' myself has ' compared with fel self haves I ' thyself hast.' On the analogy of rn'iself, feiself the plurals ureself, -gureself were formed. The dative was preserved in himself ' himself, itself/ plur. himself (Late ME 409.] PRONOUNS: SELF. Ill alsofamself). hireself 1 herself could of course be regarded either as dative or possessive. The forms -selve, -selven also occur : mtselve, mtselven, himselve, himselven. selven is prob- ably the OE dat. sing, or plur. selfum, selve being either a shortening of selven or else = OE weak self a. MODERN ENGLISH. 406. In Early MnE self came to be regarded more and more as a noun, which led to such constructions as the Shakesperian thy fair self, Tarquiris self. A new plural selves was now formed on the analogy of shelf, shelves, etc. : myself, ourselves, to your gross selves (Shakespere). 407. But the older dative was still preserved in himself, themselves, itself must also be regarded as containing the objective (= dative) case of /'/ rather than as a contraction of ifs self. In Present English we have the forms his self, their selves in vulgar speech ; and even in the Standard dialect these forms are necessary when own is added : his own self. 408. The following are the forms of the spoken lan- guage : Sing, myself; yourself; himself, itself, herself. Plur. ourselves ; yourselves ; themselves. To these may be added the indefinite oneself. 409. It will be observed that yourself, yourselves make a distinction between sing, and plur. which is lost in the simple you, the sing, thyself being, of course, preserved only in the higher literary language. So also a form ourself occurs occasionally in older writers in the sense of ' myself ' ; but in the present literary language an author speaks of himself as ourselves, if he uses the plural. 112 ACCIDENCE. [410. 410. In the literary language self is used as an inde- pendent noun : till Glory s self is twilight (Byron) ; then, all forgetful of self , she wandered into the village. 411. In the spoken language the emphatic and reflexive meanings of myself, etc., are distinguished by the stress, these forms having strong stress when emphatic, weak when re- flexive, as in I did it myself compared with he roused himself. Demonstrative . OLD ENGLISH. 412. The OE demonstrative se ' that, this, the, he/ etc., and pes ' this, this one ' are inflected as follows : Masc. Neut. Fern. Masc. Neut. Fern. Sing. Nom. se (se) poet seo pes (pes) pis peos Ace. pone pat pa pisne pis pas Dat. pam,pam pare pis sum pisse(re) Gen. pees pare pisses pisse(re] Instr. py pare pys pisse(re) Plur. Nom. pa pas Dat. pam,pam pissum Gen. para, para pissa, piss era The forms se, bes are used only as noun-pronouns in the sense of ' this one/ ' he.' MIDDLE ENGLISH. 413. In ME the s of the OE se, se, seo was made into p by the influence of the more numerous forms beginning with b, and of bes, pis, peos. 4 i8.] DEMONSTRATIVE PRONOUNS. 113 414. The resulting pe, pat, peo was at first used, as in OE, both as a demonstrative and as a definite article. But by degrees the neuter sing, pat and the plur. pa were restricted to the demonstrative meaning. In Early Southern pgt hits = OE poet hits is still used in the sense of ' the house ' as well as of ' that house ' ; but in Late ME pat is restricted to the more emphatic meaning, as in MnE. This restriction was still more marked in the plur.; already in the Earliest ME/^ men, p hus were used only in the demonstrative meanings ' those men/ ' those houses,' 415. pg was now regarded as the plur. of pat, and was completely disassociated from the definite article. Hence it became necessary to eliminate the old /a-forms ace. sing. fern, and nom. plur. from the inflection of the definite article. This was done by extending the nom. sing. fern, first to the ace. sing. fern. -peo sunne = OE pa synne as well as seo synn and then to the plur, nom.: peo sunnen=OE ba synna, 416. The old pas the ace. sing. fern, and nom. plur. of pes was now associated with the old pa, till at last ME pg and pgs were completely confused, pgs being regarded as pg with the plural -j added, so that pg men, pgs men both came to mean ' those men.' 417. The form pgs was now eliminated from the inflection of pes in the same way as pg was eliminated from the in- flection of pe, the fern. sing. nom. peos being extended first to the ace. fern. sing, and then to the nom. plur. : peos siinne=QE peos synn and pas synne, peos sunnen=O^E pas synna. 418. The following are the full inflections of the Early Southern demonstratives corresponding to OE se and/w: i ACCIDENCE. [ 4*9- Neut. Fern. Masc. Neut. Fern. p$t peo pes pis peos Pet peo pesne pis peos pen per pisse pisse pisse pes per pisses pisses pisse peo peos pen pissen per pisse 114 Masc. Sing. Nom. pe Ace. pene Dat. pen Gen. pes Plur. Nom. Dat. Gen. 419. But already in Early Southern there was a tendency to make the definite article indeclinable pe. The main causes of this were (a) the want of stress of the article, which made its endings indistinct, (<5) the general loss of the sense of gender- and case-distinctions, and (c) the confusion which arose from using/// both as an article and a demonstrative. 420. The new demonstrative pat was in like manner extended to the masc. and fern. sing, and then to the oblique cases of the sing., so that pat hus, pat man were sharply dis- tinguished from pe man, pe hus. 421. The neuter pis was extended in the same way : pis man, pis hus, pis cu = OE pes mann, pis hus, peos cu. 422. At first the indeclinable pat was not always restricted to its demonstrative meaning, but was used also as an article in all three genders. This usage survived in Late ME in a few combinations : pat gn ' the one/ pat oper ' the other/ pat ilke ' the same '=OE se ilca, pat ilce, etc. The final / of the pat was often regarded as the beginning of the next word, and the a was weakened to e so as to make the curtailed pat into the definite article pe, the first two of the above combina- tions being written pe tjn, pe toper. The tother has been preserved to the present day in vulgar English. In Early MnE the tother and the other were blended into t'other, which was still used in the literary language of the last century. 4 28.] PRONOUNS: ONE, A; NONE, NO. 115 423. In the plural, where there was no distinction of gender, fig, figs and fieos became indeclinable even sooner than the singulars fiat, fits. 424. The plural fieos ' these ' was discarded in Late ME, and a new plural was formed direct from fits by adding the regular adjective plural ending , giving fiise, which also appears in the weak form fiese, like hese=.hise. fiese may, however, be the result of the influence of the older fieos, fieos, which in Late ME would become fies, fies. MODERN ENGLISH. 425. Standard MnE finally settled down to the demon- strative forms Sing. that this Plur. those these 426. In Early MnE the article the is often shortened to tK before vowels and h + vowels, as in tH enemy, th'hilt, and even before other consonants, as in tKworld, where the w was probably dropped. one, a ; none, no. 427. In OE the numeral an 'one/ which was inflected like a strong adjective (but with ace. sing. masc. es there was in OE a third demon- strative pronoun geon, which however became obsolete already in Early West-Saxon. It was preserved in North-Thames English, being still in. common use in the north of England and Scotland in the form of yon. In MnE yon has been confused with the adverb yond, yonder yond cloud, yonder hill of which it was supposed to be a shortened form, and was consequently written yon . yond is now completely obsolete, and yonder is more frequent than_jwz in the literary language, both being obsolete in the spoken language. 441. The OE demonstrative of quality swglc, swilc, Late West-Saxon swylc=.*swalic, *swilic ' so-like,' *swa being the older form of swd ' so,' dropped its / in ME in the same way as hwglc did, Southern swiich becoming swuch by the influence of the w, which was then absorbed by the u, giving such. The tradition of the Midland form sivich is still preserved in the vulgar sick. Indefinite. 442. The particle a ' always ' was in OE prefixed to pro- nouns and adverbs especially interrogative ones to give them an indefinite sense, as in dhwdr ' anywhere,' ahwcefier ' either of two.' Interrogative pronouns and adverbs were also used in an indefinite sense without any prefix, as in gif hwdpds boc dwritan wile 'if anyone wants to make a copy of this book.' The indefinite meaning grows naturally out of the interrogative, such a question as ' who ? ' being necessarily indefinite, for if we knew who the person was, we should not ask the question. The indefinite meaning was made more prominent by putting the interrogative word between swd . . 120 ACCIDENCE. [443- swa 'as . . as ' : swd-hwd-swd ' whoever,' swd-hwcet-swd ' whatever/ swd-hwglc-swd ' whichever.' In ME the first swd was dropped in these groups : whg-s$, what-sg. In Late OE cefre ' ever/ ' always ' is sometimes added like the older a though more loosely to express indefiniteness, as in eall fiat cefre bglst wees ' whatever was best ' ; and in ME this usage was much extended, whence the MnE whosoever, whatsoever, and, with dropping of the now superfluous so, whoever, whatever, whichever, etc. 443. In OE the noun wiht ' creature/ ' thing/ came to be regarded almost as a pronoun, and when the indefinite a- was prefixed to it, the origin of the resulting noun-pronoun dwiht was forgotten, and it was contracted to duht, dht, aht. The prefix d- also appears in the form of o-, whence the parallel forms owiht, oht. Hence ME has both auht, aht, and ouht, oht. In OE negative forms were obtained by prefixing n- : ndwiht, nduht, naht, nowiht, noht, whence ME nauht, naht, and nouht, noht. The fluctuation between au and ou in these words still continued in MnE, even when the two spellings had come to represent the same sound (o). We now write only au in aught, making an arbitrary distinction between naught and nought. In OE nauht, etc. were used as ad verbs =' not at all/ ' by no means/ and in ME they became less and less emphatic, especially in the weak forms, which dropped the h, becoming nat, not, which at last became equivalent to the older ne ' not/ 444. some^OE is still used as a plural noun-pronoun, the singular being represented by the compounds someone, somebody, something. In ME the two indefinite pronouns sum and what were combined in sumwhat to express the same meaning as something; somewhat is now used only as an adverb. 448.] INDEFINITE PRONOUNS. 121 445. any = the OE noun and adjective a 'both those,' 'both the/ bd being the fern, and neut. form corresponding to the masc. I eg en ' both,' just as the fern, and neut. two. ' two ' cor- respond to the masc. twegen. 122 ACCIDENCE. [449. 449. each=OE die from *agilic, literally 'ever each/ where the ge- has the same collective meaning as in gefera 'companion,' etc. ale in ME became glch, and with the same dropping of the / as in which gch, the Northern form being ilk, which was thus confounded with ilk 'same'=OE ilca. each is still a noun as well as an adjective, though there is also a compound noun-form each one. 450. every is a ME compound of afre (442) and ate, the earliest ME form being jvrjlch, then gvrich, which in Late ME was shortened to jveri. every is now used only as an adjective, the noun being represented by the compounds everyone, everybody, everything. 451. either = OE &gj>er, czghwcefier from *dgihwce/>er. OE cegper has the meaning of Latin uterque 'each of two/ ' both of two/ the meaning ' one of two/ Latin alteruter, being expressed by ahwceper without the collective ge-, which often shortened to aufrer, aper. The difference of meaning is seen in such sentences as on izgpere healfe eas ' on both sides of the river ' and gif he duper pissa forl&tt ' if he gives up either of these two things/ In ME the pronoun ^/for=OE duper was gradually disused, and /$w=OE cegper was used to express both meanings. * In MnE either is now generally restricted to the alternative meaning alteruler. In ME both jiper and guper continued to be used as con- junctions, weak guper being contracted to gper, Qr, or. gper . . or ' either . . or ' in which the first member kept its fuller form because it kept the strong stress was in Late ME made into jtfer . . or, as in MnE. 452. In OE there was a negative form corresponding to duper: ndhw&fier, nduper, ndper, nohwceper, noper. In ME it was preserved as a conjunction, the weak form being shortened to nor. The strong form njufier was, on the other 455-1 NUMERALS: CARDINAL. 123 hand, made into a new- formation njifier on the analogy of fiper t being used both as a pronoun and as the first member of the correlative conjunction- group rigiper . . . nor, as in MnE. 453. In MnE either and neither are used both as adjec- tives and as nouns. 454. There are a few quantitative pronouns remaining, whose etymology and history deserve notice : several has the same form in ME and old French; it comes from the Late Latin separate, corresponding to Old Latin separdbilis ' separable.' few=OE^ea,/eawe plur. many=OE manig, Late West-Saxon miznig by the ana- logy of cenig. ME mani, mem' with back-shortening. Early MnE (mani, meni). NUMERALS. Cardinal. 455. The cardinal numerals 1-12 are expressed by the following isolated words : one. OE an. two. OE masc. twegen, neut. and fem. twd. Already in the Earliest ME twd was extended to the masc. : twd men= OE twegen mgnn. But tweien, tweie=QE, twegen was pre- served, and, indeed, survives in the present literary English in the form of twain, but was used indiscriminately in all three genders. In Late ME /z^=OE twd became two by the same influence of the w as in who (373). In -Early MnE the (w) of (twuu) was soon absorbed, giving (tuu). three. OE fine, neut. and fem. fireo. In ME the latter form was extended to the masc., becoming fire in Late ME. 124 ACCIDENCE. [ 456. four. OE feower, which in ME became fower, four, the e being absorbed by the two lip -consonants between which it stood. five. OE fif, absolute fife, flf, like the other isolated numerals above three, though uninflected when joined to a noun, is generally inflected when used absolutely : flf m/nn, heora zvczron fife ' there were five of them,' In ME both forms were kept, the conjoint flf and the absolute five, the latter being by degrees extended to the conjoint use, whence the MnEjive. six. OE stex, six, Anglian sex. seven. OE seofon. eight, OE eahta, Anglian cehta, whence ME eighh. nine. OE m'gon. ME nigen, nln, absolute nine. ten. OE fieri, Anglian ten. ME fen with shortening, eleven. OE gndleofan. ME enleven, etteven, absolute elevtne. twelve. OE twglf, absolute tivglfe. ME iwelf, twelve. 456. The teen-numerals 13-19 are compounds of the units with -trene, Anglian -tene : thirteen. OE pritlene, frritfiene, fireofiene. ME }>ritten-e. The MnE form shows the same consonant-transposition as in third (466). fourteen, QRfeowerfiene. fifteen. OE flf tune. In ME fif tene the I was shortened before the consonant-group, sixteen. OE sixfiene. seventeen. OE seofontiene. eighteen. OE eahtafiene, Anglian cehtatene. ME tightetene, contracted eightene. nineteen. OE m'gontiene. ME ntgentene, ntntene. 457. The ty-numerals 20-90 are formed in OE by com- 460.] NUMERALS: CARDINAL. bining the units with -tig, which was originally a noun meaning ' a lot of ten/ ' half a score,' so that twmly originally meant ' two tens/ The numerals 70-90 also prefix hicnd- : twenty. OE twentig from *twegen-tig, twentig. thirty. OE brltig, brittig. ME. britti, Late East-Midland birii, with the same transposition as in third. forty. Q}Lfeozuertig. fifty. OE/i/tig. TftE.fi/li. sixty. OE sixttg. seventy. OE hundseofontig. ME seovenlig, seventi. eighty. OE hundeahtatig , Anglian hundczhtatig. ninety. OE hundnigontig. 458. In OE the ty-numerals are sometimes declined as adjectives, as in after brltigra daga face ' after the space of thirty days/ When undeclined they are used in their original function of nouns governing the genitive : six tig mila brad ' sixty miles broad/ 459. The high numerals hundred and thousand are in OE neuter nouns, hund. hundred and busend, governing the genitive : twa hund wintra ' two hundred winters (years)/ busend manna ' a thousand men/ 460. In OE there was no numeral higher than thousand. million, ME millioun, is the French form of Late Latin mlllio, aec. millidnem formed from Latin mille 'thousand/ billion, trillion, etc. are much later formations, in which the Latin prefixes bi- and tri- (as in biennial^ triennial] were substituted for the initial syllable of million, so that billion was regarded as a sort of contraction of *bimillion. milliard is a Modern French formation from Latin mille, or rather from million, by substituting the augmentative ending -ard for -on, so that the word means ' big million/ million itself originally meaning ' group of thousands/ 126 ACCIDENCE. [461. 461. Numeral-groups are either cumulative, as in twenty-five 20 + 5, or multiple, as in two hundred-= 2x1 oo. In such cumulative groups as twenty-five the units always came first in OE -flf and twentig manna and we still say five-and-lwenty as well as twenty-five, but only with the lower ty-numerals; thus we hardly ever say he is five and fifty. 462. In speaking we generally count by hundreds up to 1900, especially in dates. Thus 1066, 1891 are called ten hundred and sixty-six, eighteen hundred and ninety-one, or, more briefly ten sixty-six, eighteen ninety-one. 463. The high numerals are not used alone, but require a or one before them, the latter when emphatic, as in one hundred, not two hundred, a in a hundred, a thousand, etc. may be the indefinite article, but is more probably the weak one, as in a day or two (433. i). 464. In MnE all the numerals are treated as adjectives followed by nouns in the plural, a-hundred, etc. being a kind of group-adjective : ten men, twenty men, a hundred men, two thousand men. 465. But all the numerals can also be used as nouns with plurals in -s. They necessarily become nouns when their head-word is suppressed, as in units, tens, and hundreds, to go on all fours, there were ten of us; but even when the head- word is expressed, the numeral may be made into a noun whenever it has any independence of meaning, as in thousands of people. Ordinal. 466. Most of the ordinal numerals are derivatives of the cardinal ones, but the first two ordinals are expressed by distinct words : 468.] NUMERALS: ORDINAL. first is the Qfcfyrest, which originally meant 'foremost' (349) ; but this meaning was sometimes so much weakened that fyrest became practically equivalent to forma, which is the regular OE ordinal corresponding to an, second was introduced in ME, being the French form of Latin secundus. The OE word was oper, which was discarded because of the ambiguity resulting from it having also the meaning ' other.' third =O~Efiridda, Late Northumbrian pirda, ME pirde. The other OE ordinals below 20 are formed from the cardinals by adding -pa, the p becoming / after s or f, and final n of the cardinals being dropped : fourth = (JEfeowerpa, feorpa, ME fourpe. [Cp. fourteen, forty.-] fifth=OE ft/fa, ME fifte, Early MnE fift. In later MnE the th was restored by the influence of the other ordinals, as also in Early MnE sixt. sixth =OE six fa. seventh^OE seofopa is a Late ME new-formation direct from the cardinal. So also ninth, tenth, eleventh. eighth =OE eahtopa, Anglian ahtopa, where the o is the older form of the a in the cardinal eahta, cehla. ninth. =OE nigopa. tenth =OE teopa with the unmutated vowel of the cardinal tien. eleventh = OE e^ndleofta. twelfth=OE twelfta. 467. The OE teen-ordinals end in -teopa, which in ME was made in -tenpe, a new-formation from the cardinal ending -tene, as in fiftenpe=. QfiLfifteopa. 468. The OE ty-ordinals end in -tigopa, -twgopa, which in ME became -teope, and then -tipe by the influence of the ACCIDENCE. [ 469. cardinals : OE twentigofia, Late ME twentifre. In Early MnE e was introduced by the analogy of the verb-inflection -eth, but these ordinals were still pronounced (twentib, J?irti), etc., although the spelling has now altered the pronunciation into (twenti-i)?), etc. 469. In Early MnE the ordinal ending -th was extended to the high numerals, which before had no ordinal forms : hundredth, which was pronounced (hundrej>), thousandth, millionth. 470. The OE ordinals were inflected as weak adjectives. 471. In ordinal groups only the last member of the group takes the ordinal form, the others being left in the shorter cardinal form : twenty-fifth or five-and-twentieth, hundred and second. This usage prevailed already in OE, as in on fram twd-and-twentigopan d&ge, where two. is kept in the neuter, although dag is masculine, because it forms a sort of group compound with the ordinal. 472. The ordinals are used as nouns in MnE in the combination of two ordinals to express fractional numbers, as in two thirds of an inch. VERBS. Old-English. INFLECTIONS. 473, There are two main conjugations of verbs in OE, strong and weajc, distinguished mainly by the formation of their preterites and preterite participles. If we compare these parts of the verb with its infinitive, we find that strong verbs, such as bindan ' to bind/ form their preterite by vowel- change 476.] VERBS: OLD ENGLISH. 139 band 'he bound' and add -en in the preterite participle with or without vowel-change, ge- being often prefixed, in weak as well as strong verbs gebunden ' bound '; while weak verbs, such as hieran ' hear,' form their preterite and preterite participle with the help of d or / : hierde, gehiered. 474. The following are the chief verb-endings of the active voice, including the preterite participle passive. Where two endings are given, the second is that of the weak verbs. Observe that all three persons have the same ending in the plural, and that the imperative exists only in the second person. Indicative. Subjunctive. Present Singular i -e -e 2 -st -e 3 -J> ~e Plural -ab -en Preterite Singular I -, -de -e, -de 2 -e, -dest -e, -de 3 -, -de -e, -de Plural -on, -don -en, -den Imperative Singular -, -, (- Gerund -enne Participle Present -ende Preterite -en, -ed 475. Verbs whose root ends in a vowel generally con- tract ; thus seon ' to see,' gdn ' to go/ conjugate ic seo, ic gd, we seob, we gap compared with ic binde, we bindab. 476. For the plural ending -ab, both indie, and imper., -e is substituted when the pronoun comes immediately after the verb : ge bindab, but binde ge. So also gd ge ! compared with ge gab. These forms were originally subjunctives, binde ge being a shortening of binden ge. So also in gd we ' let us go.' This change was often extended by analogy to K 130 ACCIDENCE. [ 477. the ending -on, as in mote we ' may we/ sohte ge ' ye sought ' compared with we moton, ge sohton. 477. The passive voice, and many forms of the active voice as well, are expressed by the combination of auxiliary verbs with the pret. partic. and, more rarely, the pres. partic. The chief auxiliary verbs are wesan ' be,' weorban ' become,' and habban ' have,' as in he wees gefunden, he wearb gefunden ' he was found,' he is gecumen ' he has come,' he hcefb gefunden ' he has found.' 478. But besides the pret. partic., there is a trace of the old Germanic passive in the form hdtte from hdtan, which is both pres. ' is named, called,' and pret. ' was called.' 479. The infinitive was originally an indeclinable abstract noun formed from the corresponding verb, so that bindan originally meant ' binding/ ' act of binding.' The gerund is a similarly formed noun in the dative case governed by the preposition to, which always precedes it, as in he is to cumenne ' he is to come ' = Latin venturus est. It often takes the a of the infin. to cumanne. 480. The pret. partic., as already stated, generally takes ge- before it ; but not if the verb already has ge- or a similar inseparable prefix, as in forgiefen ' forgiven/ aliesed ' re- deemed/ In West-Saxon hieran generally takes ge- through- out : gehieran, gehiered. 481. Both participles are declined like adjectives : we sindon gecumene, he hcefb hine gefundenne ' he has found him/ literally ' he possesses him found.' But in the later language the pret. partic. in combination with auxiliary habban became indeclinable through the original meaning having been forgotten : he hczfb hine gefunden. 482. In the older language the second person sing, ends in -s: bu lufas 'thou lovest/ bu lufades. But already 487.] VERBS: OLD ENGLISH. 131 in Early West-Saxon the regular forms are lufast, lufa- dest. 483. In Late Northumbrian inflectional p became s : he bindes, we bindas. 484. In Late OE the subj. plur. ending -en was made into -on by the influence of the indie., as in gyf hy waron 'if they were/ compared with Early West-Saxon gif hie war en. 485. In Late OE the -st of the 2nd pers. sing. pret. indie, of weak verbs is extended to the subj. : gyf pu lufodest ' if you loved '= Early West-Saxon gif bu lufode. STRONG VERBS. 486. In the strong verbs the plur. of the pret. indie, often has a vowel different from that of the sing. : ic band, we bundon. The 2nd sing. pret. indie, and the whole pret. subj. always have the vowel of the pret. plur. indie. : bu bunde, gif ic bunde, gif we bunden. The following are the Early West-Saxon inflections of the strong verb bindan : Indie. Subj. Pres. Sing. I binde binde 2 bindest, bintst binde 3 bindeb, bint binde Plur. bindab binden Pret. Sing. I band bunde 2 bunde bunde 3 band bunde Plur. bundon bunden Imper. Sing. bind Infin. bindan Plur. bindab Gerund to bindenne Partic. Pres. bindende Pret. gebunden. 487. Some strong verbs are inflected like weak verbs every- K 2 132 ACCIDENCE. [ 488. where except in the preterite forms. Thus swgrtan ' swear,' pret. swor, is inflected \ikefgrian (504) : pres. indie, swgrige, swgresi, szugrefi, swgriab ; subj. pres. swgrjge, swgrigen ; irnper. swgre, swgriab ; pres. partic. sw^rigende. Many strong verbs with double consonants, such as biddan ' pray,' ' ask ' pret. bad, are inflected like s$tian (5O3) : pres. indie. bidde, bitst (bides /), bitt (bideb\ biddab ; subj. pres. bidde } bidden ; imper. bide, biddab ; pres. partic. biddende. All of these verbs, both strong and weak, had a / before their endings in Germanic (148) *swarjan, *farjan, *bidjan, *saljan ; and hence all of them mutate their root-vowels. The strong verb wepan 'weep* is also a 'j-verb,' as shown by its mutation, the Anglian form being wcepan, and is declined like the weak verb hieran, which however has the same endings as a strong verb in the infinitive and present tenses, and so there is nothing to distinguish the inflections of wepan from those of the ordinary j-less strong verbs : pres. -wepe, uiepst (wepesf), wepb (wepefi), rwepab ; .imper. wep, etc. 488. The Germanic forms of the endings -si, -b were -is, -ib, which are still preserved in the oldest English : bindis, bindib. In West-Saxon these endings mutated a preceding vowel and then dropped their own vowels, as in bu lycst, hit grewb from lucan ' close,' ' lock,' growan ' grow.' The re- sulting consonant-combinations were modified in various ways (147) : tb, dj>, ddb were made into //, /, as in latt ' lets,' bitt ' waits,' bitt ' asks,' sfynt ' stands ' from Itztan ' let/ bidan, biddan, standan ; and sb became st, as in clest ' chooses ' from ceosan. Similar changes took place in the and pers. sing. : bu bitst 'you ask,'/ clest. In Anglian the full endings -es (-/), -eb were restored, the unmutated vowels being at the same time restored : leteb, bideb, bideb, biddeb, styndeb ; biddes, ceoses. VERBS: OLD ENGLISH. 133 489. The vowel-changes in the strong verbs are gener- ally due to gradation (150), which is often accompanied by consonant-change, as in weorpan, geworden (146). But in some verbs the vowel of the pret. is the result of contraction of Germanic and Arian reduplication; thus heold 'held' (infin. healdari) is a contraction of *hehold, *hehald. Traces of this reduplication are preserved in a few OE preterites, such as he-ht, later het (infin. hdian 'call/ ' command ')= Germanic *hehait (Gothic haihatf). 490. The following are the classes under which the strong verbs fall according to their vowel-changes, each class being named after a characteristic verb. A few exam- ples only are given of each class. The special Anglian forms are given in ( ). The forms are given in the order infin., pret. sing., pret. plur., pret. partic. I. Keduplicative or fall-class. 491. The pret. sing, and plur. has eo or e, the pret. partic. keeping the vowel of the infin. : feallan (fallan) 'fall' healdan (haldan) 'hold' heold cnawan ' know ' gro-wan ' grow ' beatan ' beat ' hatan ' command ' l&tan 'let' II. Shake-class. 492. These verbs have in the infin. a, ea, or, in j-verbs the mutations /, /' beak bugon bogen WEAK VERBS. 498. The weak verbs fall under two main groups, ac- cording as the vowel of the infin. is mutated or not. The mutation-group comprises two classes, the hear-class (huran) and the wean-class (zu/m'an), the unmutated verbs consti- tuting the third or love-class (lufian). I. Hear-class. 499. The following are the Early West-Saxon forms : Indie. Subj. Pres. Sing. I ktere htere 2 hterst htere 3 hlerb hure Plur. hierab hieren Pret. Sing. I hlerde hlerde 2 hierdest hierde 3 hlerde hierde Plur. hierdon hierden 136 ACCIDENCE. [ 500. Imper. Sing. filer Infin. hieran Plur. hieraj) Gerund to hierenne Partic. Pres. hierende Pret. hiered 500. This class adds -de in the pret. and -ed in the pret. partic., where the e is liable to be dropped when an inflec- tional vowel is added, as in the nom. plur. gehierde. Verbs ending in /, d, c drop the e in the uninflected form also, as in asgnd ' sent ' (infin. asgndari), where d is a shortening of dd. After the breath-consonants /, c the inflectional d is unvoiced, and c becomes h : metan ' find/ { meet ' gemelt, tcecan ' show ' getaht. But the full forms as^nded, gemeted also occur, especially in Anglian. Similar changes take place in the pret. -Ide, -p(p}de become -//e, -pie, as in gemeite ' found/ dypte ' dipped ' (infin. dyppan}. The inflectional d is also unvoiced after ss and the other breath-consonants, as in missan ' miss ' misie, compared with r&sde ' rushed ' from raisan, where the J=(z). In dypte the / is, of course, a shortening of pp. There are similar shortenings in sgndan, sgnde,fyllan,fylde, etc. 501. I b. Seek-class. In this subdivision of the hear- class the vowel of the infin. is unmutated in the pret. and pret. partic., the inflections being the same as in the other verbs of the hear-class : sgllan ' give ' sealde (salde) geseald (gesdld) secan (soecari) ' seek ' sohte, sohte gesoht, gesoht 502. Those with n followed by c or g -pgncan ' think/ bringan ' bring ' drop the nasal and lengthen the preceding vowel and modify it in other ways : fyncan, fiohte, gefioht = Germanic *j>ankjan, *J>anhta, an before h having been regularly changed to nasal a, which in OE as regularly 504-1 VERBS: OLD ENGLISH. 137 became o. Long vowels were shortened in OE before hi, so that pohte, etc. became fiohte. Seek-verbs in -gcc carry the mutated vowel / into the pret. and pret. partic. in Late West- Saxon : strgccan, ' stretch/ streahie, streaht (strahte, strahf) later sirghte, strght. 503. It will be observed that all verbs of the hear-class have long syllables in the infin. either a long vowel, as in hleran, or a vowel followed by two consonants, as in sgndan, fyllan. In the latter verb the // is Germanic [cp. the adjec- tive _/"#//], and is therefore kept through all the inflections of the verb, except where / is written for // before a consonant in contracted forms : pres. indie, fylle, fyllest {fylsf}, fyllep (fylfi), fylla}> ; imper. sing, fyll, etc. But most of the verbs of this class with double consonants in the infin., such as sgttan ' set/ are inflected like strong j-verbs such as biddan (487), the double consonant being also shortened in the pret. and pret. partic. : pres. indie, sgtte, sgtst (s$tes\ sgtt (sgte})\ sttap ; subj. sgtte(ri) ; imper. sgte, sgttap ; pres. part, sgttende ; pret. sgile=.*sgtede, pret. partic. // (sgep), sgcgaj) ; imper. s g e > 5 gtS a J> ') pres. partic. sgcgende ; pret. scegde, pret. partic. gescpgd. So also sgllan has pres. indie, sglle, sglp (sglefi), sgllafi, imper. sgle, sgllap, &c. II. Wean-class. 504. All of these verbs have infin. -ian and a short root- syllable with a mutated vowel. They form their pret. in -ede, and their pret. partic. in -ed, which is never contracted. The following are the Early West-Saxon forms of wgnian ' accus- tom ' : 138 ACCIDENCE. [ 505. Indie. Subj. Pres. Sing, i wgnige wgnige 2 wgnest wgnige 3 wgnej) wgnige Plur. weniap wgnigen Pret. Sing. I wenede wgnede 2 wgnedest wgnede 3 wenede ivgnede Plur. wenedon wgneden Imper. Sing. wene Infin. wgnian Plur. wgnia}> Gerund to wgnigenne Partic. Pres. wenigende Pret. gewgned So alsofgrian ' carry ' [faran ' go '] styrian ' stir.' III. Love-class. 505. In Germanic these verbs had infinitives -an, -on, of which -ian is a later development and therefore does not cause mutation like the -ian of the wean-class, which is of Germanic origin. The following are the Early West-Saxon forms : Indie. Subj. Pres. Sing, i lufige lufige 2 lufast lufige 3 lufap lufige Plur. lufiap lufigen Pret. Sing. I lufode lufode 2 lufodest lufode 3 lufode lufode Plur. lufodon hifoden Imper. Sing. lufa Infin. lufian Plur. lufiap Gerund to lufigenne Partic. Pres. lufigende Pret. gelufod. So also ascian ' ask,' macian ' make,' and many others. [ 509. VERBS: OLD ENGLISH. 139 Irregular Weak Verbs. 506. Some weak verbs, such as libban 'live/ show a mixture of the inflections of the hear- and the love-class : pres. indie, libbe, leofast, leofab, libbab ; subj. h'bbe(n) ; imper. leofa, libbab ; pres. partic. libbende ; pret. Kfde, pret. partic. gelifd. PRETERITE-PRESENT VERBS. 507. These verbs have for their presents old strong pre- terites ; thus the preterite-present verb wat ' I know ' was originally a strong preterite of the shine-class. The present of these verbs differs however from the strong preterites in the 2nd sing, indie., which ends in /or st, a /before the inflectional / also becoming s : ic sceal ' I shall/^ scealt ; id cann ' I know,' bu canst ; ic wdt ' I know,' bu wast. 508. From these presents new weak preterites are formed with various irregular changes : sceolde, cube, wiste. 509. Many of these verbs are defective, the infin., imper., and participles being often wanting. The subj. is often substituted for the imper. sing. The following are the inflec- tions of witan ' know ' : Indie. Subj. Pres. Sing. I wat wite 2 wast wite 3 wat wite Plur. witon witen Pret. Sing, i wiste wiste 2 wistest wiste 3 wiste wiste Plur. wiston wisten 140 ACCIDENCE. [ 510. Indie. Subj. Imper. Sing, ivite Infin. tvitan Plur. witap Gerund to witenne Partic. Pres. witende Pret. ivtten. Middle-English. EARLY MIDDLE ENGLISH. 510. The ME levelling of weak vowels under e had a comparatively slight effect on the verb inflections, especially in Early Southern, where the OE verb-inflections were pre- served very faithfully. But the inevitable change of -a, -as/, -ap, -ode into -e, -est, -ej>, -ede, as in luve, Invest, luvep, luvede = OE lufa, lu/ast, lufap, lufode, necessarily led to a complete levelling of the old wean- and love-classes of weak verbs, the ME love-class including all the OE ian-verbs whether accom- panied by mutation or not. 511. The Southern tendency to drop final n first affected the infin. and pret. partic. : Early Southern blnden, btnde ; ibunden, tbunde. 512. The tendency to shorten double consonants in weak syllables made the OE gerund to bindenne into ME to bmdene. 513. The tendency to drop final weak e after another weak syllable (174) led to the shortening of to blndene into to bmden, which made it liable to be confused with the infin. So also /'/ luvep Plur. bindep hire]) luvief> Pres. Subj. Sing. binde here luvie Plur. blnden heren luvien Pret. Indie. Sing. I bnd herde luvede 2 bunde herdest luvedest 3 bgnd herde luvede Plur. biinden herden luveden 526.] VERBS: EARLY MIDDLE ENGLISH. 143 Pret. Subj. Sing. I bunde herde luvede 2 bunde herdest luvedest 3 bunde herde luvede Plur. biinden herden luveden Imper. Sing. bind here lu-uie Plur. bindeb hereb luvieb Infin. binden heren luvien Gerund bindene herene luviene Partic. Pres. blndinde herinde luviinde Pret. ibimden iher(e)d ilu-ved. 523. In the forms binde ge, bunde ge, -e is substituted for -eb (476). 524. It will be observed that the distinction between the two classes of weak verbs is very slight, the i of the love-class being often dropped I luve, we luveb, &c. while the imper. sing, here has taken the e of luvie, luve. Midland. 525. In Early Midland many levellings which are only just beginning in Early Southern are fully carried out. The love-class lost their i entirely, and as the hear-class generally had the full Anglian endings -esf, -ej>, there is only one set of inflections for the two classes : heren, lufen Southern heren, luvien. On the other hand, the contracted forms of the hear-class are extended to the love-class, as in birfi ' befits/ ' becomes ' pret. birde = OE gebyrep, gebyrede, infin. gebyrian (wean-class). 526. The characteristic feature of the Midland verb is its extension of the plur. ending -en of the subj. pres. and of the pret. indie, and subj. gif ' }>ei lufen, bei comen ' came/ gif ' bei comen, bei brohten to the present indie, plur. : we lufen, bei cumen= Southern we luvieb, heo cumeb. But the older -(e)b 144 ACCIDENCE. [527. is kept in the imper. plur. : cumej) !, bep I ' be ye ' = Southern cumep, beop. 527. In Early Midland the gerund was completely levelled under the infin. : to binden, to heren. 528. In Midland the pres. partic. keeps the old ending : bmdende, herende, lufende. The n of the infin. and strong pret. partic. is never dropped as in Southern. The pret. partic. loses its prefix ge-. 529. The distinction between single and double conso- nant forms in the old j-verbs, such as hgbban, hfep, hof, hafen and libban, leofap, lifde, which was still kept up in Early Southern hebben, hevep\ libben, levep, livep began to break down in Early Midland through the extension of the single consonant forms ; thus in Early Midland we find pres. plur. indie. lifen-=. Early Southern libbep, although the older infin. libben is still kept in Early Midland ; but hefen is used not only as a pres. plur., but also as an infin. Northern. 530. In the Northern dialect inflectional / had been changed to s, and final n had begun to drop off already in the OE period : Old Northumbrian bindes, bindas, binda= Mercian bt'ndeb, bt'ndap, bindan. In the Early Middle period weak final e was dropped, so that the infin. dmde=O\d Northum- brian binda became monosyllabic bind, under which the gerund to bind was levelled. The subj. dmde=O\d North, sing, and plur. binde was reduced to the same monosyllable. Hence also the pret. plur. herden was reduced to the same form as the sing. herd. The effect of these changes on a strong pret. such as that of bind was to leave only two forms band ist and 3rd pers. sing, indie., and band 2nd pers. 534-1 VERBS: EARLY MIDDLE ENGLISH. 145 sing, and plur. and subj. generally and the vowel-change was soon got rid of by extension of the vowel of the ist and 3rd person sing, indie. : i band, bii band, we band. 531. In Late Old Northumbrian the old ending of the 2nd person pres. -es, -as, etc. was preserved by the influence of the new 3rd person -es, -as = -eb, -ab. Hence in Early Northern -es became the common ending of the 2nd and 3rd persons indie, pres. sing. In the pres. indie, plur -es= older -as, -ias was dropped when the verb was immediately pre- ceded or followed by its pronoun : we bat bindes, men bindes ; we bind, bat bind. The ' absolute ' form was afterwards ex- tended to the ist pers. sing, as well ; i bat bindes. 532. The n of the strong pret. partic. was not lost in Old Northumbrian because of the inflected forms gebundene, etc., by whose influence the n was restored in the uninflected form ; hence it was always kept in the ME Northern dialect as well, 533. The Northern form of the pres. partic. is -and: bindand, herand= Midland and OE bindende, herende, Southern bindinde, herinde. This a is the result of Scandinavian influ- ence : Icel. bindandi, heyrandi. 534. The following are then the most distinctive verb- inflections of the three dialects in their Early Middle periods : Southern. Midland. Northern. Indie. Pres. Sing. I binde binde bind 2 blndest, bintst blndest bindes 3 bindeb, bint bindeb bindes Plur. bindeb blndtn blnd(es) Imper. Sing, bind bind bind Plur. bindeb bindeb blnd(es) Pres. Partic. bindinde bindende bindand 146 ACCIDENCE. [ 535. LATE MIDDLE ENGLISH. 535. The most important change in Standard ME and in Late South-Thames English generally is the further assimi- lation of the pres. partic. to the verbal nouns in -inge by which the earlier blndinde became bindinge, a change of which we see traces already in Early Southern, as in heo riden sing- inge ' they rode singing ' OE hie ridon singende. But as the verbal nouns also occur without final -e, the distinction be- tween lerninge partic. and lerning noun was not entirely lost. 536. Early ME d was changed to / in the weak pret. and pret. partic. of verbs in rd, Id, nd\ girte, girt, infin. girden; bilte, bill infin. bilden ; wente, went infin. wenden = Early Southern gurde, giird; billde, biild ; wende, wend. This change served to distinguish such forms zshe sende pres. subj. and he senie pret., which in Early ME were both expressed by the first form. But it is also carried out in some words with /, //, n, nn : felen ' feel ' felte ; dwellen, dwelie ; mjnen, mente ; brennen 'burn,' brente; and after J=(z) and v, where it un- voices these consonants: losien=.Q}L losian, loste; ljven=Qfe lafan, lefte, lafte. 537. In Standard ME we see the same levelling and simplifying tendencies at work as in Early Midland and Northern. The old vowel-change in such preterites as bgnd is still kept up, but the short form bgnd is often extended throughout the pret. : bu bgnd, we bpnd as well as bu bonnde, we bounde(n). 538. In some verbs of the bear- and give-class the e of the plural is sometimes extended to the sing, as in ber, set by the side of bar, sat= OE beer, sat plur. bdron, s&ton, Anglian beron, seton. 539. Influence of the strong plur. pret. on the sing, is 544-1 VERBS : LATE MIDDLE ENGLISH. 147 also seen in such sing, preterites as slow, jaw>= Early Southern sloh plur. slowen, Late OE sloh, slogan, OE seah, sawon. 540. In Late ME the pret. partic. begins to influence the pret. plur. As a general rule the old pret. plurals were preserved in Late ME only when they had the same vowel as the pret. partic., as in pet bounden, bei drbnken, pet wonnen (class 3), riden, writen (class 6) ; otherwise the plur. pret. took the vowel of the pret. partic. : bei holpm^foghten, chgsen. 541. The sing, of the imper. began to be extended to the plur. : bind ' bind ye ' by the side of bmdeb. 542. In the love-class of weak verbs the i was dropped entirely, and the pret. ending -ede was often shortened to -ed in accordance with the general principle of dropping weak e after a weak syllable : he loveb, he loved. 543. Some of the above changes may be the result of Midland influence, of which we have an undoubted example in the substitution of -en (-e) for -eb in the plur. indie, pres. -eb was, of course, kept in the plur. imper., although here also the Midland ending seems to occur in its shortened form -e : binde. 544. The following are the Standard ME inflections of the three verbs whose Early ME inflections have been given already : Pres. Indie. Sing, i binde 2 bindest here Ifoe Ifcest 3 bindej>, bint Plur. binde(n) her(e)J> Aere(n) 5> Pres. Subj. Sing. binde Plur. binde(n) here here(n) lltve l$ve(ri) Pret. Indie. Sing. I bgnd 2 bounde, bpnd 3 bgnd Plur. bounde(n\ bynd herde herdest herde herde'ji) I8ved(e) l&vedest l&ved(e) lfrvede(n), ISved L 2 148 ACCIDENCE. [ 545- Pret. Subj. Sing. I bounde herde I8ved(e) 2 bounde 3 bounde Plur. bounde(ti) herde(sf) herde herde(ii) I8vcde(sf), loved lSved(e) ldvede(n), ISvcd Imper. Sing. bind her(e) love Plur. bmde(]>) ) bind here(J>), her Ifoe'J} Tnfin. brndeji) here(ii) love(ri) Gerund bmden(e) t bmde heren(e), here l$ven(e}, Idve Partic. Pres. bindinge heringe ISvinge Pret. (i}bounde(n) (i}herd (i)lov\e)d. The following examples will show the regular development of the different classes of strong verbs : I. Fall-class. 545. fallen hglden growen kngwen 546. shaken waken laughen drawen fell held grew knew fellen helden grewen knewen II. Shake-class. shok shaken wok woken laugh, low low en dro^^gh) drow drowen fallen hglden growen kngwen shaken waken laughen drawen Observe that the preterites of this class have split up into two groups, one with arjkt), blessed. -cd being thus used to express (t), this spelling was often extended to such preterites as burnt, smelt, which were written burned, smelled, although they come from ME brenie, smelte. But the phonetic spellings thank't, thankt (thankd], dropt, crosl (cross' d), accurst also came into partial use, and some of them have become fixed, such as past in half past one compared with the time has passed quickly. The above are organic changes. We have now to consider the internal changes in the verb-inflections, beginning with those of a levelling character. 560. The change of strong to weak verbs which we observe in ME went on in the transition from ME to MnE, and, in some cases, in MnE itself. Thus the Early MnE preterite clomb and the pret. partic. molten have now become climbed, melted. But some of the weak forms that arose in Early MnE have now been discarded, such as the Shakesperian pret. participles corned, becomed. 561. On the other hand, several weak verbs have been made strong by the analogy of strong verbs, such as stick, stuck (OE slician, sticode] by the analogy of sting, stung; wear, wore, worn (OE wgrian, wgrede) by the analogy of swear, swore, sworn. So also several weak verbs in -ow have taken pret. participles in -own by the analogy of know, known, etc., keeping the original weak pret. : show, pret. showed, pret. partic. shown (OE sceawian, sceawode). 562. The levelling of the short quantity of the vowels in the sing, of strong preterites under the long quantity of the 565-] VERBS: MODERN ENGLISH. pret. partic. and infin. seen in Late ME bar = Early ME fyr, bar is carried much further in MnE, as in brake, spake = Late ME brak, spak, pret. partic. broken, infin. brgken etc. When a certain number of preterites in a had been thus lengthened, others were lengthened without regard to the length of the other parts of the verb, such as came, &&/ 'cloth']. Scandinavian klcepa, klizfrdi, whence ME cljfien, cladde Northern cledde, as well as c!jjj>(i)en, cljpede. Vowel-change (ae . . o). 609. catch; caught. ME cacchen, caughte from Old French cachier [Low Latin captidre = Latin captdre, a frequentative of capere ' seize ']. cachier is probably a North- East French (Picard) form ; the Parisian form being chacier (Modern French chasser], whence the MnE chace, chase. ME cacchen having the same meaning and the same termination as lacchen, laughte from OE Iceccan, gelahte ' seize,' ' catch ' [compare MnE latch'], naturally formed its preterite in the same way. 610. distract; tdistraught, distracted. OE strgccan ' stretch,' pret. streahte, sirghte, appears in ME in the form of strecchen, straughte, streighte, the pret. partic. streight being still kept in MnE as an adjective straight literally ' stretched out.' In Late ME the Latin distrdctus was imported as an adj. distract (French distrait], which was made into distraught by the influence of straught. When distract was made into a verb in Early MnE, distraught was naturally regarded as its participle. Through further confusion straught itself was 6i3.] VERBS: IRREGULAR. used in the sense of ' distracted,' and a new partic. Mestraught was formed on the analogy of beset. Vowel-change (99 . . o). 611. work ; * wrought, worked (wgak, rot). OE zvyrcan, Anglian wi'rcan, the corresponding noun being weore, Late West-Saxon wore, Anglian were, which in ME influenced the verb. The ME forms are : Southern wiirchen, worchen with the usual change of wii- to wu-, Midland werken, Northern wirk. The OE pret. worhte underwent the usual r-transposition in ME, becoming wrohte, MnE wrought, which in ordinary speech survives only as an adjective, as in wrought iron. Vowel-change (i . . o). 612. bring; brought (brirj, brot). OE bringan, brohle. 613. think ; thought. In OE there were two weak I a verbs of -allied form and meaning : bgncan, bohte ' think ' ; byncan, buhte ' seem,' which was impersonal, me byncb ' it seems to me ' having much the same meaning as ic fence. In ME figncan became regularly benchen in South-Thames English, benken in North-Thames English; -xti&byncan became biinchen, binchen in South-Thames English, binken in North- Thames English. The pret. buhte was soon disused, bo(u)hte taking its place : he bohte ' he thought,' him bohte 'it seemed to him.' In Standard ME the two verbs were still kept apart in the infin. and present tenses, which had the Midland forms benken, I penke ; binken, me binkeb, etc. ; but in the compound bibinken ' consider ' = OE bebgncan, the latter had already begun to encroach. In Northern bink completely supplanted penk, as in MnE. Hence MnE think is histor- ically = OJZ j>j>ncan, and its pret. thought= OE bohte, the pret. of the lost bgncan. 1 68 ACCIDENCE. [ 614. Vowel-change (ij . . o). 614. seek ; sought ; beseech ; besought. OE secan (scecan\ sohte. ME South-Thames sechen, bisechen, North- Thames seken, biseken. The MnE seek and beseech are there- fore from different dialects of ME. Shakespere has the Midland form not only in seek, but also in beseek. 615. reach ; fraught, reached. OE rdcan, rcehie. ME rjchen, ra(u]ghte, Northern reghle. So also teach, taught from OE weak t&can ' show.' INVARIABLE VERBS. (aa). 616. cast. ME casten from Scandinavian weak kasta, kastapi. In Early MnE there is also a regular pret. casted. (ai). 617. *dight 'adorn' as in storied window richly dight (Milton). OE dihtan ' arrange,' ' appoint ' from the Latin dictdre. (B). 618. cut. ME kutten. 619. shut. OE styttan 'lock/ 'bolt' [gestot 'shot,' 'dart'; sceolan strong VII 'shoot']. ME schiitten, schutten. 620. thrust. ME priisten, brusten from Scandinavian brysta. (e). 621. let. OE strong I M/an, let, Idten. ME leten, pret. strong let, and weak lette from *lette. In MnE the short vowel of this weak pret. was extended to the infin., etc. The obsolete verb let ' hinder,' still preserved in the phrase let or 629.] VERBS: IRREGULAR. 169 hindrance, is the OE weak Igttan, Igtte, connected with lest ' slow/ late adv. ' late.' 622. set. OE sgttan, sgtte, connected with the strong verb V sittan, pret. sat. 623. shed. OE strong I scddan, sceadan, seed ' separate,' a meaning still preserved in the noun watershed. ME schgden formed a weak pret. schadde, schedde, and developed the new meaning ' separate into drops/ ' shed.' In MnE the short vowel of the pret. was extended to the pres., etc., as in let. 624. shred. OE screadian, screadode. ME schrjden, schredde, the short vowel being afterwards extended to the pres., etc. So also spread (spred) from weak OE sproedan. (99). 625. burst. OE strong III berstan, barst, burslon, geborsten. The u of burst is the result of the influence of the lip-consonant b on the eo of ME beorsten, as in burn (592), the u being afterwards extended to the pret. partic. bursten, which survived in Early MnE. 626. hurt. ME hiirten, hurten. (i)- 627. hit. ME kitten from Scandinavian hitta ' find.' 628. knit. OE cnyttan 'tie' \cnotta 'knot']. The invariable pret.-form is now preserved only as an adjective in well-knit, etc. Otherwise the pret.-form is regular knitted. 629. quit. ME quiten pret. quitte from Old French quiter from Latin qvietus. In MnE the shortened vowel of the pret. was extended to the rest of the verb. The deriva- tive requite keeps its original length, having a pret. partic. 170 ACCIDENCE. [ 630. requit in Early MnE. acquit is invariable in Early MnE. All these verbs are now regular. 630. rid. ME redden, riidden, ridden ' rescue/ ' separate fighters ' is apparently a blending of OE hrgddan ' rescue ' and Scandinavian ryfy'a pret. rudda ' clear away.' 631. slit. OE strong VI slitan, slat, sliten. ME has both strong sliten, pret. partic. sliien, and a weak verb slitten, which may have existed in OE. 632. split. ME splatten, of which Early MnE splette is probably a Northern form, spiel seems to have been made into split by the influence of slit. (o). 633. cost. ME costen from Old French coster (Modern French couier) from Latin constdre. (u). 634. put. ME /#//?. VOCALIC VERBS. Vowel-change (ai . . au). 635. bind; bound. OE strong III bindan, band, bunden. The older pret. partic. is still preserved in bounden duty. So also grind, ground; wind, wound from OE strong III grindan, zvindan. 636. find; found. OE strong III findan, fand more generally weak funde -funden. ME Vowel-change (ai . . B). V 637. strike ; struck. OE strong VI stncan, strac, stricen ' move about/ ' touch lightly.' ME striken, strgk (Northern 639.] VERBS: IRREGULAR (VOCALIC). Ijl strdk), striken. Early MnE strike, pret. stroke, stroke, struck, pret. partic. stricken, strucken, struck. Vowel-change (se . . B). * 638. hang; hung, hanged. OE strong I hdn (from earlier *hohan), heng, hangen, the g being a weakening of the h of the infin., where o = Germanic an (502), so that hon= Germanic *hanhan. There was also a weak intransitive hangian, hangode, hon itself being used transitively. In Early ME the consonantal variation in the strong verb was soon levelled : sometimes the infinitive form was extended to the pret. partic. which was made into (a)hon ; but afterwards the ng-forms got the upper hand, being supported by the weak verb hangien, and a new strong infin. hangen was formed, pret. heng, pret. partic. hangen. In some dialects the pret. was shortened to heng with short close (e), which being an unfamiliar sound in ME was made into i. This new pret. king, which is frequent in some Midland dialects, was made into an infin. in Northern by the analogy of the bind-class, with pret. hang, which afterwards made its way into the Standard dialect, in the form of hgng parallel to sgng ' sang.' A pret. partic. hung was further developed on the analogy of sing, sang, sung, and hung was then extended to the pret. sing, in the same way as clung, etc. (565), the older infin. hang being preserved in the Standard dialect. In MnE the strong form hung is both transitive and intransitive, hanged being used only transitively, contrary to the OE usage. Vowel-change (i . . B). 639. dig; dug, t digged. ME diggen, diggede, equiva- lent to OE dlcian [die ' ditch '], of which it seems to be a 1 72 ACCIDENCE. [ 640. modification by some analogical influence. The vocalic pret. dug developed itself towards the end of the Early MnE period ; it is not found in the Bible. 640. cling; clung. OE strong III clingan, clang, clungen 'wither.' ME clingen, clpng, clungen 'shrivel,' ' adhere,' ' hang.' So also slink, slunk ; spin, t span, spun ; sting, stung ; swing, swung ; win, won ; wring, wrung from OE strong III slincan, spinnan, stingan, swingan. gewinnan, wringan. 641. fling; flung. ME strong III flingen from weak Scandinavian flgngja [compare ME wing from Scandinavian v l n S r \ flingen was, of course, made strong on the analogy of sting and the other strong verbs in -ing. 642. sling; slung. ME strong III slingen from Scandinavian slongva, which passed through slengen into slingen, and then became strong in the same way as fling. The pret. slang occurs in the Bible. 643. stick ; stuck, tsticked ' pierce,' ' adhere.' OE stician (stiociari), sticode ' pierce/ ' adhere.' ME strong V steken, slak, steken and stoken [like spoken = OE speceri\ ' pierce/ ' imprison/ which may represent an OE strong verb, stuck may owe its u to the influence of stung. 644. string; strung, stringed. This verb is a MnE formation from the ME noun string from Scandinavian s/rgngr, with the usual change of Scandinavian -eng into -ing. We keep the older consonantal inflexion in stringed instruments. Vowel-change (e . . ss . . u). ijfc 645. run; ran; run. OE strong III irnan, iernan (eornari), Late West-Saxon yrnan, pret. grn, am, pret. partic. urnen, with the same transposition of the r as in burn, the older 648.] VERBS: IRREGULAR (VOCALIC\ 173 forms being preserved in gerinnan ' coagulate/ literally ' run together,' gerann, gerunnen. The ME verb was influenced by the two Scandinavian verbs, the strong rinna, rann, runninn and the weak rgnna, rgndi, the Standard ME forms being indeed entirely Scandinavian: rennen, ran, irunnen. The Early Southern forms of the infin. are irnen, eornen, urnen probably =># from Late West-Saxon yrnan. The infin. run appears in Northern by the side of the Scandi- navian rin. The u c f the infin. seems to have been origin- ally a Southern development out of urnen, perhaps by the influence of burn. Vowel-change (i . . se). 646. sit; sset. OE strong V j-verb sitlan, s&J, seten. ME si/ten, sat, selen and also siten with the vowel of the infin. From the ME partic. siten is derived the obsolete MnE pret. and pret. partic. sit, which made the verb invari- able. The obsolete MnE pret. sale is due to the analogy of came, spake, etc., the short sat being kept up at the same time by the short vowel of the infin. sit. 647. spit ; spat. There were in OE two weak verbs of the same meaning spittan, spitte and sp&tan, sp&tte, both of which were kept in ME, where the pret. spcelte became regu- larly spalte. The MnE spit, spat is, therefore, a mixture of two distinct verbs. Vowel-change (i . . SB . . B). 648. begin ; began ; begun. OE strong beginnan. So also drink, drank, drunk(en) ; shrink, shrank, shrunk ; sing, sang, sung ; sink, sank, sunk(en) ; spring, sprang, sprung; stink, stank, stunk; swim, swam, swum 174 ACCIDENCE. [ 649. from OE strong III drincan, scrincan, singan, sincan, springan, stincan, swimman. 649. ring ; rang ; rung. OE (ti)ringan, which is ap- parently weak. Vowel-change (i . . ae . . i-n). 650. (for)bid ; -bad ; -bidden. OE strong V j-verb biddan, freed, beden ' pray,' ' ask ' ; strong VII beodan, bead, boden ' offer/ ' command/ The corresponding ME forms are bidden, bad, bjden and by the analogy of the infin. bidden ; beden, bed, bgden. But already in Early ME the two verbs began to be confused, bidden in the special sense of ' ask to one's house,' ' invite ' soon got confused with beden, which developed the meaning ' offer an invitation,' the con- fusion being aided by the weak verb bpd(i}enOE bodian ' announce ' itself connected with beodan. Hence even in Early ME we find iboden used in the sense of ' invited.' It was still more natural to soften down the command expressed by beden by the substitution of the milder bidden. The pret. bad soon supplanted bgd by taking to itself the meaning ' commanded,' except in the emphatic forbeden, which in Standard ME only rarely has the pret. forbad instead of forbad. The following are the Standard ME forms bidden, beden ; bad; bjjden, bgden. forbeden ; forbgd (forbad] ; for bgden. In the transition to MnE the bid-forms were gradually extended till they entirely supplanted the others. The relation between the two forms bad and bade is the same as that between sat and sate (562). In Early MnE the pret. partic. was often shortened to bid, which was used also as a pret., so that the verb became invariable. 655.1 VERBS: IRREGULAR (VOCALIC). 175 Vowel-change (ij . . e). 651. bleed; bled. OE weak bledan (blcedari), bledde. \blod ' blood ']. So also breed, bred ; feed, fed ; lead, led ; meet, met ; read, read (rijd, red) ; speed, sped from the OE weak bredan,fedan, Icedan, metan, rddan, spedan. Vowel-change (ij . . e . . ij-n). + 652. eat; ate; eaten. OE strong V, with exceptional (Germanic) lengthening in the pret. sing., etan, cet, pret. plur. ceton, pret. partic. eten. ME jien, et, at, //en, the pret. at being of course due to the influence of the other verbs of the same class. Vowel-change (ou . . e). 653. hold; held. OE strong I healdan, haldan; heold; gehealden, gehdlden ME hglden ; held, held, hild; ihglden. We still preserve the fuller form of the pret. partic. in beholden. Vowel-change (o . . e . . o-n). 654. fall ; fell ; fallen. OE strong I feaUan, fallan ; feoll; feallen, fallen. ME/allen fel,fel, fil ; fallen. Vowel-change (ai . . ei . . ei-n). 655. lie ; lay ; lain. OE strong V j-verb licgan, lag, gelegen, imper. sing, lige, etc. The ME development of this verb is analogous to that of the other cg-verbs (586). In Early Southern the infin. liggen was preserved by the side of the imper. lie; but in the North-Thames dialects it was levelled under the g-forms, becoming //, lien. The Standard ME forms are lien, lat, pret. partic. kien, kin. 176 ACCIDENCE. [ 656. Vowel-change (v . . ei . . B). K 656. come ; came ; come. OE strong IV, with anom- alous weak vowel in the pres. and infin. and exceptional extension of the vowel of the pret. plur. to the pret. sing. : cuman; cwom, com; c(w)omon; cumen. The pret. com was preserved in Standard ME, but was partially supplanted by the new formation cam on the analogy of the strong verb IV nimen ' take,' nam, nomen. cam underwent the usual length- ening into came in MnE. Vowel-change (i . . ei . . i-n). 657. give ; gave ; given. OE giefan (gefari) ; geaf Vowel-change (ai . . i). 658. light; lit, lighted. OE weak lihtan, lihte 'illu- minate ' and ' make light,' ' alleviate ' \leohl adj. ' light of colour ' and ' light of weight ']. There was a third OE weak verb lihtan, alihtan ' alight from a horse.' The MnE verb light in light on must be referred to this last. The conson- antal preterite-form lit does not, of course, appear till light had become (bit), that is, in the MnE period, when it arose from imitation of bite, bit, etc. The verb alight still keeps the older consonantal inflexion, which is also used in the other verbs. Vowel-change (ai . . i . . i-n). 659. bite; bit; bitten. OE strong VI bitan. The shortened pret. partic. is still kept in the phrase the biter bit. 660. chicle ; chid ; chidden. OE weak cidan, cidde. ME chlden, chidde. In Early MnE the verb was made strong 666.] VERBS: IRREGULAR (VOCALIC]. 177 chide, chode, chidden on the analogy of ride, rode, ridden. The pret. partic. was then shortened to chid, and extended to the pret. The verb is nearly obsolete in the present spoken English, hide, hid, hidden is a strong verb of similar recent formation, except that it does not seem to have developed any pret. analogous to Early MnE chode : OE hydan, hydde, ME hiden, hidde. Vowel-change (ij . . ij . . ij-n). 661. beat; beat; beaten. OE strong I beatan, beat, beaten. Vowel-change (ai . . o). 662. shine ; shone. OE strong VI sclnan, scan, scinen. Vowel-change (e . . o . . o-n). ^ 663. (for)get; forgot; forgotten, got. In OE the strong V verb gietan, gytan (getari) ; geat (gcef) ; gielen, gyten (gelen) occurs only in the compounds begietan ' get,' ongietan ' understand/ forgietan ' forget ' and a few others. In ME begiten, begeten was shortened to giten, geten through the in- fluence of the Scandinavian geta, gat, getinn ' get/ or rather the Scandinavian word was substituted for it. 664. tread; trod; trodden. OE strong V tredan, trad, treden. ME trjden, trad, trjden and by the analogy of broken, etc. trjden, troden. Vowel-change (ij . . o . . o-n). 665. seethe; tsod, seethed; sodden, tsod, seethed. OE strong VII seopan, seap, soden. Vowel-change (uw . . o). r 666. shoot; shot. OE strong VII sceotan, sceaf, scoten. N 178 ACCIDENCE. [ 667. Standard ME scheten, schjt, schoten. There is also an infin. schuten in ME, whose u probably =ti from OE eo, as in choose (680), which afterwards became (uu) and was written oo in Early MnE. Vowel-change (ai . . ou). 667. climb ; t clomb, climbed. OE strong III dim- man, clamm, clummen and also dimban, damb, dumben, although the latter is found only in late texts. ME climmen, dam, dommen and dimben, cljmb (clamb\ domben. Vowel-change (ai . . ou . . i-n). 668. (a)bide; fbode, tbid, bided; tbiden, tbid, bided. OE strong VI bidan 'wait,' dbidan 'endure/ ME (a}biden, bpd, biden, there being also a weak pret. abidde. 669. drive; drove, tdrave; driven. OE strong VI drlfan. So also ride, rode, ridden ; rise, rose, risen ; shrive, t shrove, shrived, shriven; smite, smote, smitten; stride, strode, t stridden, strode; write, wrote, written from OE strong VI ridan, drisan, scrlfan, smitan ' smear/ stridan, writan. 670. strive; strove; striven. ME strong VI striven, strgf, striven, which is the Old French estriver [from Old Low-German strip ' strife '] made into a strong verb on the analogy of driven. 671. thrive ; throve ; thriven. ME Driven from the Scandinavian strong reflexive verb brlfask. Vowel-change (ei . . ou). ^ 672. wake ; woke, waked. OE strong II wacan, woe, wacen, generally compounded with on- : onwacan, awacan. (on)wacan and the weak d(wczcman), wacian ' keep awake ' 678.] VERBS: IRREGULAR (VOCALIC). 179 are intransitive. The corresponding transitive verb is wgccan, weahte, wghte. ME has (a)waken, wok, waken and wakien, wakede; wakenen, wak(e)nede. The (ou) instead of (uw) in the MnE woke is probably due to the influence of the numerous preterites of the shine-class rose, etc. 673. stave ; stove, staved. This verb was first formed in MnE from the noun stave ' piece of a cask/ itself a late formation from staves, plur. of staff. Its vocalic inflexion is of course the result of analogy. Vowel-change (ei . . ou . . ou-n). 674. break; broke, t brake ; broken, t broke. OE strong IV brecan, brae, brocen. Vowel-change (ij . . ou . . ou-n). 675. freeze; froze; frozen, tfrore OE strong VII freosan,freas, froren. 676. heave; hove, heaved; thoven, hove, heaved. OE strong j-verb II hgbban, hof, hafen. ME hebben, hgven ; hof, haf; hjven, hjven, the last form being due to the in- fluence of the infin., while haf, hgven are due to the influence of U'jven, zvaf, wgven (679). There was also a weak ME pret. hefde, hevede. The MnE hove probably points to a ME pret. hpf\\\t\\ the vowel of the pret. partic. 677. speak; spoke, t spake; spoken, t spoke. OE strong V sprecan, sprcec, sprecen. In Late OE this verb began to drop its r especially in the Kentish dialect. In ME the r disappeared entirely, and the pret. panic, took o on the analogy of broken, etc. : spjken, spak, spgken, spgken. 678. steal; stole; stolen. OE strong IV stelan, steel, stolen. N 3 l8o ACCIDENCE. [ 679. 679. weave; wove, weaved; woven, weaved. OE strong V we/an, waf, wefen. ME wjven, waf, wjven, wgven. Vowel-change (uw . . ou . . ou-n). 680. choose ; chose ; chosen. OE strong VII ceosan, ceas, coren. ME chesen, chjs, chosen. There was also a West- Midland infin. chtisen with the regular West-Midland change of OE eo into H. In Early MnE (tfiuz) became (t/uuz), which was written phonetically choose, although the older spelling chuse survived till the end of the last century, chese also occurs in Early MnE. Vowel-change (ai . . o). 681. fight; fought. OE strong III feohtan (fehtan); feaht (fahf) ; fohlen. ME fighten, faught, foughten. In the pret. Early MnE fluctuates between au and ou. Vowel-change (es . . o . . o-n). 682. bear; bore, tbare; born(e). OE strong IV beran, beer, boren. MnE makes a distinction between born in the sense of French ne and borne=. 1 carried' which did not exist in OE or ME. 683. swear; swore, tsware ; sworn. OE strong j- verb II swgrian, swor, swaren, sworen, the o of the last form being due to the influence of the preceding w. ME swen'en, swjren ; swor, swar ; swgren. swar is, of course, due to the analogy of bjren, bar. 684. tear; tore, ttare; torn. OE strong IV teran. 685. wear; wore, tware; worn. OE weak wgrian, wgrede 'wear clothes.' The vocalic forms were first developed in Early MnE by the analogy of bear. 689.] VERBS: IRREGULAR (VOCALIC). l8l Vowel-change (ei . . o). 686. freight; * fraught, freighted. The Late ME weak verb fraughten [imported from Dutch ?] was made into freight in Early MnE by the influence of the synonymous fret, and fraught itself came to be regarded as the pret. of this new verb freight by a vague association with work, wrought, etc. But fraught was still used as a pres. in Early MnE : the good ship . . . and the fraughting souls within her (Shakespere). Vowel-change (ia . . o . . o-n). 687. shear; t shore, t share, sheared; shorn, t sheared. OE strong IV scieran (sceran); sc'ear (sccer)', scoren. Vowel-change (ij . o . . ij-n). 688. see; saw; seen. OE strong V seon; seah (seek}; sawon (segon) ; sezven (segen). In Late Northumbrian the adjective gesene= West-Saxon gesiene 'visible' was used as the pret. partic. Early ME seon, sen ; seih (Southern), sah, sauh pret. plur. sgwen, seien ; pret. partic. seien, sein. In Late ME the pret. sing, forms dropped the h by the influence of the pret. plur. and pret. partic., giving sei, sat and saw, the last being the usual North-Thames form, especially in Northumbrian, which also kept the Old-Northumbrian pret. partic. in the form of sen. The Standard ME inflections are se(n) ; seigh, sat; (t')sem. In MnE the Northern pret. saw and pret. partic. seen were introduced into the Standard dialect. Vowel-change (33 ... u). 689. stand; stood. OE strong II with* inserted in the pres. etc. : slandan, stod, standen. l8a ACCIDENCE. [ 690. Vowel-change (ei . . . u . . . ei-n). 690. forsake; forsook; forsaken. OE strong II forsacan ' renounce,' ' deny.' So also shake, shook, shaken from OE strong II scacan. 691. take; took; taken, *ta'en. ME strong II taken, tok, taken from Scandinavian faka, tok, tgkinn. In Northern this verb was contracted like make, and the pret. partic. id en passed into Standard MnE. Vowel-change (ai . . . uw . . . ou-n). 692. fly; flew; flown. OE strong VII fleogan (flegan, fligari) ; fleag, fleah (fleti) ; flugon ; flogen. ME flen, flien ; fleigh,fley with the same dropping of final h as in sei = Q~E gescehfly; pret. plur.flowen,flpwen (influence of pret. partic.); pret. partic. flgwen. The Early MnE pret. flew (fliu) probably arose in the same way as drew, etc. (573). Vowel-change (ei . . . uw . . . ei-n). . 693. slay; slew; slain. OE strong II slean (from sleahari); slog, sloh; slagen, slcegen, slggen. ME Southern sljn, Midland sign, Northern sld ; sloh, Late ME slough^ slow =;(sluu) ; pret. partic. slawen, sleien, slain. In MnE, the ai of the pret. partic. was extended to the infin., and the ow of the pret. underwent the usual analogical change into ew. The archaic forms she = slea, pret. slue still lingered in Early MnE. Vowel-change (ou . . . uw . . . ou-n). 694. blow ; blew ; blown, blowed. OE strong I blawan ' blow ' (of wind), bleow, blawen and blowan ' bloom/ blemv, blowen.' ME blgwen, blew, blgwen and blowen, bliw, blowen. 7oi.] VERBS: MIXED. 183 695. crow; crew, crowed; fcrown, crowed. OE strong I crawan, creow, crdwen. grow, grew, grown; know, knew, known from OE strong I growan, cndwan. Vowel-change (D . . . uw . . . o-n). 606. draw; drew; drawn. OE strong II dragan; drog, droh; dragen. MIXED VERBS. 697. There are several verbs which have a strong pret. partic. in -en with a regular consonantal pret. Some of these are old strong verbs which have become partially consonantal; but others are weak verbs which have taken the partic. ending -en by the influence of old strong verbs which they happen to resemble. In the following list the latter class are marked J. 698. go; went; gone. OE strong I gdn, gangan; geong, eode (weak) ; gegdn, gegangen. ME gg(n\ gange(n) ; yode, wente; gg(n), gangen. In ME the longer form gang was gradually restricted to the Northern dialect. The cur- tailed Southern pret. partic. gp is still preserved in the adverb ago=Q}L agan ' passed ' (of time). 699. grave, graved; graven, graved. OE strong II gra/an, grof, grafen. 700. hew; hewed; hewn, hewed. OE strong I heawan, heow, heawen. 701. tlade, load; tladed, loaded; laden, tladed, tloaden, loaded. OE strong II hladan, hlod, hidden, hladen. The MnE change of lade into load is through the influence of the noun load, ME Ijjde = OE lad (fern.) ' leading,' ' way,' connected with l&dan 'lead/ which had also the meaning 184' ACCIDENCE. [ 702. ' carry,' so that in ME lyde came to mean ' load/ and was at last confused with the verb laden. 702. melt; melted; molten, melted. OE strong III mellan. molten is now used only as an adjective. 703. mow ; mowed ; mown, mowed. OE strong I mawan, meow, mawen. 704. rive ; rived ; riven, rived. ME strong VI riven, T 2fi riven from the Scandinavian rifa. 705. Jsaw ; sawed ; sawn, sawed. ME weak saw(i)en. MnE sawn by the analogy of drawn. 706. shape ; shaped ; shapen, shaped. OE strong II j-verb scieppan, scyppan (scgppari) ; scop ; scapen, sc&pen. In ME this verb was influenced by the Scandinavian verb skapa, skop. 707. shave ; shaved ; shaven, shaved. OE strong II scafan, scof, sea fen. 708. Jshow; showed; shown, showed. OE weak sceawian, sceawode ' survey/ ' look at.' ME schfw(t)en, schgwien, Northern ^haw. Early MnE shew and show, shown by the analogy of known, etc. 709. sow; sowed; sown, sowed. OE Strong I sdwan, seow, sazven. 710. ^strew ; strewed ; strewn, strewed. OE weak strewian, streowian. ME strewen, strgwen, strawen. strewn by the analogy of hewn. 711. swell; swelled; swollen, swelled. OE strong III swellan. ISOLATED FORMS. 712. Some obsolete verbs occur only in isolated forms, namely quoth, hight, idept, wont. 713. quoth. OE strong V cwefian, cwcep, cw&don, 7I7-] VERBS: ANOMALOUS. 185 gecweden 'say.' In ME the strong consonant of the infin. was kept throughout : cweben, cwab, icweben ; so also bicweben 'bequeath/ which in MnE is consonantal be- queathed. In Late ME the simple cweben was gradually dis- used except in the pret. sing. As cwab was often unstressed in such combinations as cwab -he, it developed a weak form cwod, quod through the regular rounding of unstressed a into o after a lip-consonant, as in OE Oswold= earlier Oswald. The explanation of the d is that cwab he etc. were made into (kwajree) which became (kwafree, kwotJ-ee) ; and when (kwocS) was detached and received strong stress as it natu- rally would the final (<5), being an unfamiliar sound in strong syllables, was changed into (d). The form quoth is a blend- ing of strong quath and weak quod. 714. Light ' is named, called,' ' was called/ ME highte is a blending of the OE passive form hatle (478) and hehl, the active pret. of the same verb hatan. 715. iclept=ME icljped, OE gecleopod 'called' the pret. partic. of the weak verb cleopian, clipian. 716. wont 'accustomed ' = OE gewunod, pret. partic. of the weak verb gewunian [gewuna 'custom/ 'habit.'] ANOMALOUS VERBS. 717. Most of the MnE verbs that we class as anomalous are old preterite-present verbs. Two of these preterite- present verbs dare and owe = OE dearr, dg have been made regular in certain meanings. The original inflections of these verbs have been much curtailed in MnE, most of them having only the inflections of the finite present and preterite. The only one which has an infin. is dare, which seems to have taken it from the regularly inflected verb dare. Two of 1 86 ACCIDENCE. [ 718. the old preterite-present verbs must and ought occur now only in the OE preterite forms, which have taken the place of the OE present mot and ag, so that these verbs are incap- able of marking the distinction between pres. and pret. 718. can, canst ; could, couldst. OE cann, canst, plur. cunnon ; pret. c ube ; infin. cunnan ' know.' ME can, canst, plur. connen, can ; coube, coude ; infin. connen. coude probably owes its d to the influence of wolde and scholde (723, 724). In Early MnE coud(e) it was made into could on the analogy of should and would OE scolde, wolde. 719. dare, darest, (he) dare, tdares ; durst; infin. dare. OE dearr, dearst, durron; dorste; ME dar, dar (as in the pret. bar), dar si; dorste, durste with the u of OE durron ; infin. durren, ddren, of which the former represents the probable OE infin. durran, the latter being a new-forma- tion from dar. In MnE dare in the transitive sense of ' challenge ' has become quite regular : he dared him to do it. The intransitive pres. partic. daring is used only as an adjec- tive. 720. may, mayst; might, mightst. OE mag, bu meaht (rncehf), miht, plur. magon ; pret. meahte (rncehte) mihte ' be able/ [Compare matgen, meaht, miht ' power,' ' force.'] The ME forms seem to have been influenced by another OE preterite-present verb of similar meaning, namely deag, deah ' avail ' plur. dugon ; pret. dohte ; infin. dugan. The ME forms are : mai, miht, and, very late, mayst, plur. mawen, muwcn, moun ; pret. mahie, mihte, mohte. 721. tmote (muut); must. OE mot, most, mdton; moste ' may.' ME mot, most, moten ; moste. The pres. sur- vived only as an archaism in Early MnE : as fair as fair mote be (Spenser). Already in ME the pret. was used in the sense of the pres., and in Early MnE this usage became 7 24-J VERBS: ANOMALOUS. 187 fixed. It began with the use of the pret. subj. which was practically indistinguishable from the pret. indie. to express mild command, so ihatfiou mos/e=' you would be able/ 'you might ' was understood to mean ' you will have to/ ' you must.' The vowel of moste passed through (uu) into (u) in Early MnE, the shortening having probably begun in the weak form. 722. (owe) ; ought. OE ag, dh,bu aht, aht, plur. agon ; pret. dhte, ahte ; infin. agan ' possess.' The adjective agen ' own ' is an old pret. partic. of this verb. From agen is formed the weak verb dgnian, ' appropriate/ ' possess.' In Early ME ahte developed regularly into a(u)hte, but after- wards j was introduced from the infin. etc., giving p(u}hte. In ME gwen in the sense of ' possess ' soon took regular weak inflection i j>we, we gweb, etc. still keeping the older puhte as its pret. The meaning 'possess' gradually de- veloped into that of ' have a debt/ ' owe/ which, again, de- veloped the abstract meaning ' ought/ especially in the pret., which by degrees took the function of a pres. in the same way as must (721). 723. shall, shalt ; should, shouldst. OE steal (sccel), scealt (sccelt), sculon ; scolde, Northumbrian scalde by the analogy of walde (724)=zw/u nylt, he nyle, wenylla}) ; 188 ACCIDENCE. [ 725. noJde, etc. One of these negative forms is still preserved in the phrase willy nitty, Early MnE will he, nill fo=OE wile he, nyle he. The ME forms are : wile, wole, will, wolt, willep, willen, wollen ; wolde, walde, wolde, whose (u) is the result of the influence of the pres. forms wole, etc., which were prob- ably at first weak forms, in which the w rounded the follow- ing vowel and gradually assimilated it to itself. 725. twot; twist. OE wdt, wast,witon; wiste; witan; witende. The adjective gewiss ' certain ' is an old pret. partic. of this verb. ME wjjt, wgst, witen ; infin. witen ; pres. partic. wilinge. In Early MnE wot was sometimes made the base of a regular verb : he wotteth, wots, pret. wotted, pres. partic. wotting. The old pres. partic. still survives in the adverb unwittingly, and the infin. in the adverb phrase to wif= viz. The ME adjective zw/.y=OE gewiss has in MnE been often wrongly divided i ivis, as if it were the pronoun / with a verb equivalent to wot, a view which has been further supported in recent times by the chance resemblance of the Modern German equivalent of wot, namely weiss, plur. wissen. 723. need. This verb agrees with the preterite-present verbs in having no j-inflection. The loss of the s which seems to have begun in the transition from ME to MnE is apparently partly the result of similarity of meaning to that of the preterite-present verbs ; but the absence of the in- flectional j is partly due to the verb need ' require ' being formed directly from the noun need through the am- biguity of such sentences as Early MnE what need all this waste P We now come to the anomalous auxiliary verbs be, have, do. 729-1 VERBS: ANOMALOUS. 189 727. The verb be in OE is made up of three distinct roots ; that seen in (a) is, are, () was, and (c ) be : Indie. Subj. Pres. Sing. I eom (earn) j beo sle, sy ; beo 2 eart (earp)j bist ste, sy ; beo 3 is; bib sle, sy ; beo Plur. sind, sindon (ear on) ; beo}> slen, syn; beon Pret. Sing. I wees ivcere 2 ivcere ware 3 was ivcere Plur. ivceron ivoeren Imper. Sing. ives; beo Infin. wesanj beon Plur. we sap; beob Gerund to wesenne; to beonne Partic. Pres. wesende 728. The ea in eart and the Anglian earn, earon is a weakening of eo (368), preserved in the West-Saxon eom and the occasional eort, eorun. In Late Northumbrian this ea undergoes the usual further weakening into a: am, arf>, aron. 729. The Standard ME forms are: am, art, is, be(ri); subj. be, be(n); pret. was, wjr(e), was, wjre(ri); pret. subj., wer(e\ wgre(ti); imper. be, beb; infin. be(n); participles beinge, be(ri). The ME pret. partic. is, of course, an analogical new-formation. The North-Thames plur. ar(n) is still' rare in Standard ME, but is firmly established in Early MnE, which inflects: am, art, is, are; subj. be; pret. was, wast, wert, plur. were; subj. pret. were, wert, were; infin. be; partic. being, been. The use of be in the pres. indie, is still kept up in Early MnE : / be, thou beest, they be, etc. ; the iform he bes is, however, very rare. There is in MnE a tendency to get rid of the distinctively subjunctive inflections 190 ACCIDENCE. [ 730. of this verb not only by using thou beest as if it were a subjunctive if ihou beest = if thou be but also by substi- tuting if I was for if I were, etc. was = were was frequent in the last century not only as a subjunctive, but also in the indie. you was. In the present Spoken English the distinc- tion between was and were is strictly maintained, the substitution of was for were being a vulgarism. The subj. pres. is, on the other hand, extinct in the spoken language, except in a few phrases. 730. have. The OE inflections resemble those of libban (506) : h&bbe, hafast, h&fst, hafab, hafb, plur. habbab ; subj. habbe, hcebben ; pret. hcefde ; imper. hafa, habbab ; infin. habban ; partic. h&bbende, gehafd. In ME the old bb was gradually supplanted by the z> = OE/"of the other forms, the v itself being often dropped by contraction. The Standard ME forms are : have, weak hav, hast, hab, plur. hdve(ri), han, han; pret. hadde; pret. partic. had. In ME the weak short-vowel forms gradually supplanted the long- vowel ones; but we keep the long-vowel forms in the derivative behave, pret. behaved ^SE, behaven. The MnE literary forms are : have, hast, hath, has plur. have ; subj. pres. have; pret. indie, had, hadst; pret. subj. had; imper. and infin. have; partic. having, had. Early MnE still kept the shortened infin. ha, a=ME han: she might a been (Shakespere). 731. do. OE do, dest (d&sf), deb (doej>\ plur. dob; pret. weak dyde; imper. do, dob; infin. don; partic. donde, gedon. The mutation in dest, deb is common to all the dialects. In Standard ME the 5 of the other parts of the verb supplanted the older e: do, dost, dob, plur. don; dide; imper. do, dob; partic. ddinge do(n). In MnE (uu) = ME o. 734-1 PARTICLES: ADVERB-ENDINGS. 19! PARTICLES. 732. All the OE particles are either primary or secondary. The secondary particles are formed from other (declinable) parts of speech ; thus ham in he eode ham 'he went home' is formed from the masc. noun ham ' home,' ' homestead.' Primary particles, such as be ' by,' swd ' so ' are not formed from other parts of speech. There is no strict division between the three classes of particles, most of the prepositions being used also as adverbs, some adverbs being used also as conjunctions. Thus cer is a preposition in cer dcege ' before day(break),' an adverb in he eft wees papa swd he cer was ' he was pope again as he was before,' and a conjunction in cer beet flod com ' before the flood came.' 733. Some of the particles are simple, some derivative, such as uf-an ' above,' some compound (group-compounds), such as be-neoban ' beneath,' which is compounded with the preposition be. The above are primary adverbs. Secondary particles also admit of the same divisions, such as hdm^ sobUce ' truly,' ealne-weg ' always,' literally ' all (the) way.' Adverb-endings. 734. In OE, adverbs are regularly formed from adjectives by adding -e, a preceding & being generally changed to a : deope 'deeply,' hearde ' strongly,' 'severely,' nearwe 'narrowly,' late 'slowly,' 'with delay' from deop, heard 'hard,' 'strong,' ' severe,' nearu, l, east, west ; call ' entirely,' neah ' nearly,' genog ' sufficiently.' The most important inflectional endings are -um and -es\ 741. -um : hwilum ' sometimes/ stundum ' at intervals ' \stund strong fem. 'period'], -malum from the neut. noun m&l ' mark/ ' point of time ' is a frequent adverb-ending, as in styccemcelum ' piecemeal/ floccm&lum ' in troops.' From 743.1 PARTICLES: ADVERBS. 195 adjectives are formed miclum 'greatly,' lytlum and lytlum 'by little and little,' ' by degrees.' The isolated ME whilom is still preserved in the higher language, -m&lum in ME passed through -melen into -mile, as in dropmele, pecemele, where stycce was replaced by its French equivalent. 742. -es in OE was extended to fem. nouns as an adverb- ending : dccges and nihtes ' by day and by night,' sumeres and wintra \wintra masc. gen. like suna], riiedes ' of necessity ' \nied fem.] ; ealles ' entirely,' /lies ' otherwise ' from a lost adjective. The adverb-ending -weardes interchanges with the uninflected -zveard, as in hdmweard(es] ' homewards.' In ME and MnE this ending was dropped in some words, as in Late ME day and night; but it was more often extended, especially to adverbs which in OE ended in a vowel or n, in order to make them more distinct, as in alwqys= Early MnE alwai, OE ealneweg, gnes ' once ' = OE czne, the mutated a> being supplanted by ^=the OE a in an, twies 'twice,' Juries ' thrice ' = OE hviwa, firiwa. OE heonone ' hence ' (747) became he(o]nne in Early MnE, and by the addition of -s, hennes, OE J>anon ' thence/ kwanon ' whence ' becoming Late ME pennes, whennes by the influence of hennes. So also OE sipfran (749) ' since ' passed through sifrpen, sin into sifiens, sins. This extension of -s went on in MnE also, as in sometimes = earlier MnE sometime, which is still preserved in the higher language. Some adverbs in -es took final / in Early MnE or Late ME, as in amidst, betwixt, whilst, amongst=W$L amiddes a blend- ing of OE onmiddan and tomiddesbetwixt(), whiles, amgng. 743. The following are examples of OE group-adverbs : ealneweg, ealneg ' always,' georstandceg ' yesterday/ on weg 1 away/ on bac ' backwards/ ' back/ of dune ' down/ literally O 2 196 ACCIDENCE. [ 744. ' off the hill/ todceg ' today/ where to governs an exceptional form of the dative. All the above show isolation either of form or meaning, and therefore approximate to compounds. Such collocations, on the other hand, as on life ' alive ' literally ' in life/ on sl&pe ' in sleep, asleep/ on eornost ' in earnest ' show no isolation either of meaning or form. But in ME there was a tendency to shorten weak of and on to a whenever they were closely associated with the following word. Hence the ME forms adune, adun ' down ' adv., awai, abak, alive, aslepe, the a having been dropped in the MnE adverbs down, back. The same weakening took place in ME and Early MnE combinations, as in aclock, now written oclock=of (the] clock, and also in freer combinations, as in go a fishing OE gdn on fiscnob, twice a day = OE twiwa on dcege. In MnE this a was taken for the indef. article, so that in jackanapes=jack-of-apes it was made into an before a vowel. Some French group-adverbs formed with the preposition a were introduced in ME, where they were o'f course put on a level with the similar native combinations : apart, aflds=MnE. apace. 744. In ME the OE preposition be became bl (770), but the old be was kept in compounds such as beforen = OE be- foran, and also in some traditional collocations such as OE be sldan ' by the side/ ME beside, which was now completely isolated from bl J>e(re) side, just as alive was isolated from in al his lif etc. But the new preposition bl was sometimes introduced into these groups, being however shortened to bi\ bifore, bislde. On the analogy of the older compounds the new-formation bl cause ' by the cause ' was made into bicause, because. 745. In ME and MnE the place of a lost or obscured 74.] PARTICLES: ADVERBS. 197 ending was sometimes supplied by a preposition, giving rise to new group-adverbs, such as of a trutk = OE sopes, of right = OE ryhtes, bi pecemeleQJL styccemdlum, by little and little = OE lytlum and lytlum. Sometimes a preposition was added even when the ending was clear, as in at unawares. Pronominal Adverbs. 746. Among the OE primary adverbs there is a sym- metrical group of adverbs of place, connected with the pronouns he, fiat, hwan ' from the north,' supan ' from the south.' -on, -an often takes final - ' towards ' : wibinnan ' within/ wibutan ' without/ ymb ' around ' : ymb-utan ' around.' 765. Other compound prepositions are formed of prepo- sitions + nouns or adjectives in the four cases governed by OE prepositions the ace., dat., instr., gen. : ongemang 1 among,' literally ' into the crowd ' ; ongean, Anglian ongggn, ongen, ' against/ and togeanes, Anglian togggnes, togenes ' towards/ ' against ' contain an obsolete noun of uncertain meaning; tomiddes 'amidst' is formed from the adj. midd ' middle ' ; betweonum, betwix are formed from an obsolete adjective connected with twiwa ' twice.' 766. Those OE prepositions which govern both ace. and 769.] PREPOSITIONS. 205 dat., generally take the ace. to express motion, the dat. (or instr.) to express rest : he eode on fiat hits ' he went into the house ' ; he wunode on bam hu.se ' he remained in the house.' 767. As we see in the last examples the preposition on does duty for in, which became extinct in Later OE. 768. The OE prepositions are closely allied to the adverbs. Most of them can be used as adverbs without any change of form. Thus on is an adverb in he dyde on his byrnan ' he put on his corslet,' the MnE don and doff being contractions of OE do(n) on, do of. So also in he him tocivceb ' he said to him ' compared with he cwceb to him. Some pre- positions however, such as/or, are not used as adverbs, while others undergo change of form. Thus the adverbs corre- sponding to be and in (on) are bi and inn : he stod bi, he stod him bi, he eode inn compared with he stod be him ' he stood by him,' he eode in (on) beet hits. The preposition be is, of course, the weak form due to want of stress, bi being the original strong form. 769. In such combinations as par on, par to, which in OE are regularly used to express on it, to it, etc. on and to must, of course, be regarded as adverbs, therein, herein are, indeed, often expressed by bdrinne, herinne with the pure adverb inne=innan. It is to be observed that the prepositions were origin- ally all adverbs, which could modify either verbs (he stood by) or nouns. Adverbs were originally added to inflected nouns to express more definitely the meanings already indi- cated by the inflection. Thus ( motion to ' was originally expressed by the ace. alone, as we see in the Latin domum venit ' he came home ' and also in the adverb home itself, and the prepositions on, in, through, etc. were put before the ace. 2C6 ACCIDENCE. [ 776. of motion to define it more exactly. So also in on bam huse the idea of ' rest in a place ' was primarily expressed by the dative, which here represents the Arian locative. MIDDLE AND MODERN ENGLISH. 770. In ME the adverb bi was extended to the function of a preposition a change which had already begun in OE so that be was preserved only in compounds and traditional groups such as beforen, beside. By the change of -an, -on into -e the OE adverb foran and the preposition fore ' before ' were levelled under the latter, and by the analogy of the adverbs inne, ute=znnan, titan, the preposition mid 'with' when used as an adverb was made into mide, as in bjrmide= OE b&rmid. So also for, which had no corresponding adverb-form in OE, developed a ME adverb fore, as infi/rfore, whgrfore. The confusion that thus arose between OE for and fore was avoided by an extended use of the compound before(n). 771. In ME innan 'inside' came into general use as a preposition so as to avoid the ambiguity of OE on = ' on,' ' in.' Being generally unstressed, it was shortened first to me and then to in, the original distinction being thus restored. 772. In ME fry from Scandinavian frd andyhwz=OE fram were used both as adverbs and prepositions. We now use fro only as an adverb in the phrase to and fro. 773. In ME the preposition mid ' with ' got confused with wib ' against ' a confusion which would easily arise in such phrases as fight with (OE feohtan wib}, deal with, where the relation between the parties might be considered either from its original point of view as ' towards,' ' against,' or from that of ' participation,' ' having in common.' By degrees the more 777-1 INTERJECTIONS. 207 marked meaning of OE wip was expressed by against, and ME wifr took the meanings of mid, which then became extinct. For the differentiation of OE of, wtj> into MnE of, off, (\vrf>, wij>), see 232. 774. In ME the rare construction of prepositions with the genitive was soon given up except of course in iso- lated groups such as tomiddes and when the distinction between the other oblique cases and the nom. was lost in the nouns, and nothing was left but the distinction of nom. and objective in some of the personal pronouns, the only trace left of case-government by prepositions was that they were sometimes followed by a personal pronoun in the objective case. 775. In OE the adverbial ending -weard is sometimes used detached in connection with the preposition to in such constructions as wip hire (dat.) weard ' towards her.' In ME this is often carried further, as in to wode ward, to Troie war des = toward fie wode, towardes Troie, frg B or deux ward compared with framward Teukesburi, where framward is a new formation on the analogy of toward. In Early MnE the Bible still has to God ward. 776. In ME the a of amiddes was restored to its full form on, for which in was afterwards substituted. The body of the word was then regarded as an independent noun, so that at last inmidst developed into in the midst (of). INTERJECTIONS. 777. Interjections are primary and secondary. Primary interjections are mostly imitations of sounds that accom- pany emotions : ah, o, oh, pah, pooh, hush. From them 208 ACCIDENCE. [ 778. other parts of speech may be formed ; thus hush is used as a verb to hush. Such interjections as what ! dear me I are secondary. There are also mixed interjections, made up of primary interjections combined with other parts of speech, such as alas from Old French halas, alas [Modern French he'las], made up of the interjection a and las = Latin lassum ' weary/ 778. The OE la I eald ! ' oh ! ' seem to be primary. wd I ' woe ! ' is the same word as the noun wdwd, wd ' misfortune.' wdld ! wdldwd ! ' alas ! ' are therefore mixed interjections. 779. Interjections may stand in various grammatical relations to other words. Hence in OE, wd sometimes governs a dat, as in wd pam mgnn I ' woe to the man ! ' wdld governs a gen. in such phrases as wdld pare iermpe ! 1 alas for the misery/ on the analogy of the gen. after verbs of repenting, &c. As we see from the above examples, interjections are frequently connected with prepositions in MnE. COMPOSITION. Old-English. 780. The normal way of forming compounds in OE is by joining together two words which may be themselves compound or derivative words the former word being unin- flected, the latter, if declinable, keeping its power of inflec- tion, and, if a noun, determining the gender of the whole compound. Thus the neuter noun gold and the masculine noun smip can be combined to form the compound mascu- line noun goldsmip ' goldsmith.' So also dfen-fid ' evening time ' is feminine because its last element is a feminine noun. 783.] COMPOSITION. 309 These compounds of noun + noun are the most frequent. There are also compounds of adjective + noun, such as halig-dceg ' church festival ' literally * holy-day/ cwic-seolfor ' mercury ' ; of noun + adjective, such as wln-sced * sati- ated with wine,' and of adjective + adjective, such as wid- cuj> ' widely known.' 781. In the above examples the part of speech of the whole compound is determined by that of the last element. But there is a class of adjective + noun compounds having the function of adjectives, such as glcsd-mod 'having a glad mood/ Uip-heort ' blithe of heart,' ' cheerful/ formed from the adjectives glad, bfifre and the nouns mod, heorte. As we see from the last example, the noun is sometimes shortened in such compounds. We call these compounds conversion- compounds, because they involve the conversion of a noun into an adjective. They are very old formations, such con- version-compounds as the Greek dus-menes ' having an evil mind/ having apparently been formed in Parent Arian. The OE ending -lie is really the obscured second element of old conversion-compounds (844). 782. The form-isolation of compounds in OE consists in the indeclinability of the first element. It is only by this criterion that we can distinguish such compounds as godd&d ' benefit ' from the word-group god ddd ' good action,' as in the dative plural goddddum compared with godum dddum. 783. Normal OE compounds take the stress on the first element; but as word-groups beginning with the genitive of a noun or an inflected adjective do the same, stress is in OE no criterion of composition as opposed to mere grouping. Hence there is in OE no formal distinction between such a word-group as -cyninges sunu ' king's son/ in which the mean- p 210 ACCIDENCE. [ 784. ing of the whole follows from that of its elements, and one in which there is isolation of meaning, such as the plant- name peaces-sure ' sorrel,' literally ' cuckoo's-sour.' But as most of the latter class developed into true compounds in MnE through keeping their uneven stress (786), it is con- venient to regard them as ' genitive-compounds ' in OE as well. The following are examples of such OE genitive compounds, many of which, it will be observed, have been obscured in MnE : Tlwes-d&g ' Tuesday/ literally ' day of the war-god' (Tiw), the name being a translation of the Latin dies Martis (French Mardi), Jgngla-land ' England,' literally ' land of the An- glians' [Jungle plur. 'Anglians/ 'English']. These com- binations are especially frequent as place-names, such as Seoles-ieg ' Selsey,' literally ' seal's island,' Oxena-ford ' Ox- ford,' literally ' ford of oxen.' 784. Verbs are very rarely compounded directly with nouns or adjectives in OE. But the frequent combinations of verbs with prefixes, such as mis-don ' act amiss,' ' do wrong,' led to combinations with certain adjectives in similar adverbial meanings, such as full in fullfyllan 'fully fill,' ' fulfil/ full-wyrcan 'fully work/ 'complete/ and efen 'even/ ' equal/ which in composition expresses the idea of com- munity or association, as in efen-prowian ' sympathize/ liter- ally ' suffer in common with.' The want of stress in the first elements of these compounds shows that they are felt as mere prefixes. Modern English. 785. In MnE some compounds are formed by adding to the first element the Latin and Greek connecting-vowel o, but only when the first element is in a Latin or latinized 787.] COMPOSITION. 211 form, as in Anglosaxon, Anglo-Indian, Franco-German, a concavo-convex lens. The connecting vowel o is very frequent in Greek compounds, such as hippo-ddmos ' horse-taming,' philo-sophia ' philosophy,' literally ' loving wisdom.' In such forms as hippo-, philo- is preserved one of the most frequent forms of uninflected nouns and adjectives in a primitive stage of Parent Arian. When in- flections were fully developed, these old uninflected forms sur- vived only as the first elements of compounds. The ng in the MnE nightingale may be due to the influence of evening. In MnE handiwork the i is the OE prefix ge-, preserved in en0ugh = OE genog, the OE form of the compound being hand-geweorc. The /- was preserved in MnE probably through association with the adjective handy, handicraft =QJL handcrceft probably owes its i to the influence of handiwork and handy. 786. One of the formal tests of composition in MnE as well as in OE is the inseparability and indeclinability of the first element. But owing to the scantiness of the inflections in MnE and its more rigid word-order, these tests are not so decisive in it as in OE, especially when an adjective is the first element. The only certain test by which we can dis- tinguish between compounds and mere word-groups in MnE is stress, the former throwing the stress on to one of the elements, while in the latter the stress is equal. This is how we distinguish between the compound -blackbird and the group "black -bird. 787. One result of this further development of stress- distinctions in MnE is that we are able to recognize a special class of MnE genitive-compounds, distinguished from mere genitive-groups in the same way as compounds beginning with an adjective are distinguished from the corresponding word-groups, namely by having uneven instead of even P 2 212 ACCIDENCE. [ 788. stress, as in the compound crow's-foot (a plant) compared with the group a crow s fool. 788. Hence also the OE compounds goldfczt, godddd have in MnE been separated into the groups -gold 'vessel, good deed, such OE compounds as goldsmib, cwicseolfor being preserved as compounds in the form of goldsmith, quicksilver by their uneven stress ; while the OE groups domes dag, blcec bgrige have been made into the compounds doomsday, black- berry. 789. Some compounds of MnE formation have a noun in the plural as their first element, but only when this noun in the plural has developed a meaning of its own different from that of the singular, so that it is isolated from its singular, the connection between them being sometimes forgotten. Such compounds are cloihesbrush, clothes-basket, etc., newsboy, newspaper. 790. As regards the use of the different parts of speech in composition, the most noticeable difference between OE and MnE is the greater freedom with which in MnE verbs enter into composition with nouns and adjectives, the result of the combination being sometimes a noun, as in break-water, clasp-knife, sometimes a verb, as in browbeat, whitewash, according as the last element is a noun or a verb. But such compounds are still comparatively rare, the main combina- tion of verbs being with particles, as in OE. Meaning of Compounds. 791. The general rule of English as also of Parent Arian composition is to put the adjunct-word before the head-word, on the same principle of putting the modifier before the modified word as we follow in the group adjective 794-1 COMPOSITION. 213 -f- noun. Hence the order in the compound blackbird is the same as in the group black bird. In such groups as man-of-war, bread-and-butter, on the con- trary, the modifying element follows, instead of preceding, and accordingly the stress is thrown on to the second element. 792. In many cases the logical relation between the elements of a compound may be defined with certainty and accuracy. Thus it is perfectly clear that in goldfish the first element defines the second one by stating something that the second element resembles, the compound being equivalent to 'gold-resembling fish,' or more definitely 'gold- coloured fish.' So also it is evident that sight in sightseer stands in the same relation to seer as it does to the verb see in he saw the sights, and that the elements of churchgoer stand to one another in the same relation as church and go do in he goes to church. 793. But in many cases these logical relations are less definite. Thus a water-plant might mean a plant growing in the water, or a plant growing near the water, or, on the analogy of water-melon, we might suppose it to mean a plant containing a great deal of moisture, and perhaps growing in a comparatively dry place. The logical relations between the elements of causal and phenomenon-compounds are often difficult to define accurately, even when the meaning of the compound itself is definite, as in sundial, which might be explained either as a ' dial for showing the position of the sun,' or as a ' dial worked as it were by the sun instead of by clockwork, etc.' 794. It must, indeed, be borne in mind that this very vagueness is the chief reason why composition is resorted to : it is only by leaving open the logical relations between the elements of compounds that we are able to form them 214 ACCIDENCE. [ 795. as we want them without stopping to analyze exactly the logical or grammatical relations between the words we join together, as we might have to do if we connected them together by more definite means, such as prepositions or inflections. 795. An important general distinction between compounds as regards their meaning is the closeness of the logical connection between them. We may from this point of view distinguish between coordination- and subordination- compounds. Thus in a causal compound the relation between the two elements is an intimate one, like that between the clauses of a causal complex sentence. There are hardly any pure coordination-compounds in English, such a combination as deaf-mu/e=' a. person who is deaf and dumb ' being an even-stress group-compound and not a pure compound. DERIVATION. Native Elements. PREFIXES. 796. Some of the OE prefixes are strong (strong-stressed), some weak (weak-stressed). Noun- and adjective-prefixes that is, prefixes added to nouns and adjectives respectively are generally strong, as in 'mis-deed 'misdeed,' un-cuj> ' unknown ' ; while verb-prefixes are generally weak, as in forgiefan 'forgive.' When the same prefix is used both with nouns (and adjectives) and with verbs, it generally takes a shortened and weakened form in the latter combination, which is the natural result of its weak stress. The follow- 8oi.] NATIVE PREFIXES. 215 ing are examples of such pairs of originally identical prefixes : and-giet ' intelligence ' on'gietan ' understand ' cef-fiunca ' grudge ' ofbyncan ' to grudge ' 'or-banc ' device ' a'p^ncan 'devise ' 'bl-gang ' circuit ' be'gan ' practise ' 797. When a verb is formed direct from a noun or adjective, the strong form of the prefix is preserved un- changed, as in 'andswarian (andswgrian) ' to answer ' from the noun -and-swaru ' answer.' 798. Conversely, in a noun formed from a verb the verb- prefix is preserved unchanged, as in a'liesednes ' redemption,' literally ' loosenedness.' from d'liesan ' release,' ' redeem/ It sometimes happens that a noun which originally had a strong prefix takes the corresponding weak one by the influence of a verb of similar meaning. Thus bigang is often made into be'gang by the influence of be'gan. 799. In some cases older distinctions between the strong and weak forms of prefixes have been levelled. Thus the weak for- in forddn ' destroy,' forweorban ' perish ' appears in the earliest OE as fer-, for- being then used only as the corresponding strong form in such nouns as -forwyrd 1 destruction,' from which it was gradually extended to verb- forms. So also to- in td'brecan ' break to pieces ' is repre- sented by te- in earlier OE. 800. In the case of these two prefixes the weak stress was kept in the originally weak forms in spite of the adoption of the strong forms. But in some cases the prefix not only kept its strong form when transferred to a verb, but also its strong stress ; thus the prefix mis- has strong stress in 'misdon as well as in misdad. 801. The prefix ge-, on the other hand, always has weak 2l6 ACCIDENCE. [ 802. stress, not only before verbs, as in geseon ' to see,' but also in nouns, such as gesihp ' sight,' where it has supplanted an older strong form. 802. Prefixes to pronouns and particles are sometimes weak, sometimes strong. The following are the most important of the OE prefixes, the strong being marked (). 803. a- (strong form or-). The original meaning of this prefix was ' out,' ' from,' ' forth/ which may still be traced in such verbs as arisan 'arise,' while in many cases it is practically unmeaning or, at most, emphatic as in aberan 'carry,' 'endure.' 804. - a- is a shorter form of dwa ' always.' It is a strong prefix which is used only with pronouns and particles to give them an indefinite meaning, as in -ahwceper ' either of two,' ahwdr 'anywhere,' from hwaper 'which of the two?' and hivar ' where ? ' 805. 'ieg- was originally a ' always ' followed by the prefix ge- in its older form^-z'- (808), whose i mutated the preceding a into }u ' anger,' ME wrabbe, MnE wrath, from the adjectives lang, brad, hat, eald, wrap. 830. -TiJ>, -]> fem. with and without mutation : treowb 1 fidelity,' biefb 'theft,' W& befbe,befte, from treowe ' faithful,' beof ' thief/ sl&wb ' sloth/ which in ME became slgube by the influence of the adjective from which it was formed, namely OE slaw, ME slgw ' indolent.' Togeogub ' youth ' corresponds the adjective geong 'young.' In ME the ending -/tf=OE -b was substituted for the equivalent -=OE -u, as being more distinct, whence the MnE length, breadth=W lengbe, OE Igngu, etc. Similarly OE dupe from deop ' deep ' has become depth. So also ME wele, MnE weal=GE wela ' prosperity/ ' wealth/ has developed a secondary form wealth on the analogy of health from OE halu. In ME and MnE some new derivatives in -th have been formed, not only from adjectives, as in warmth, dearth, but also directly from verbs, 224 ACCIDENCE. [831. as in growth. Stealth, the latter on the analogy of ME freffrc ' theft/ 831. -ung, -ing, fern, forms abstract nouns from verbs : bletsung ' blessing/ leornung, learning ' learning,' rdding 1 reading,' from the verbs bletsian, leornian, rddan. In OE this ending is restricted in its use, and is very rarely used to form derivatives from strong verbs because these are generally provided with other derivatives, such as cyme ' coming,' gang ' going ' corresponding to the strong verbs cuman, gdn. In ME the use of -tnge, -ing was so much extended that at last abstract nouns could be formed with it from any verb, till it finally developed into a purely grammatical form the gerund (553). In MnE many words in -ing have assumed con- crete meanings, such as being=. ' creature/ In most cases these concrete words in -ing express either the result of the action expressed by the verb, as in building ' what is built/ ' edifice,' dripping, leavings, or the instrument of the action of the verb, as in clothing, covering, footing 'ground to put the foot on/ In some words -ing has a collective meaning, as in paling, shipping. Some of these words, such as shipping, seem to be formed directly from nouns. The following endings were originally independent words in OE itself: 832. -dom masc. is from the noun dom ' judgment,' ' authority,' and expresses first ' rank,' and then condition generally : cynedom ' royal authority,' ' kingdom,' king having been substituted for the less familiar cyne (825) in ME, martyrdom^ crisiendom, wisdom. In MnE there are a few new-formations, such as dukedom. Christendom and heathendom have now become concrete. In OE itself Icecedom ' medicine ' from Idee ' leech/ ' physician ' had a concrete meaning. 834-1 NATIVE SUFFIXES. 2,2$ 833. -had masc. from the noun had ' rank,' ' condition,' ' character/ ' nature ': biscophdd ' rank of bishop,' 'episcopacy,' preosthdd ' priesthood,' cildhdd, ma-gbhdd ' virginity,' the more familiar magden being substituted for mcegb 'virgin,' 'maid' in the ME maidenhod. widwan-hdd 'widowhood' is really a group-compound of had and the genitive of the weak noun widwe. In ME this ending became -hod with close o instead of g, whence the MnE -hood. The frequent ME form -hjde, -hgd is the result of the influence of another ending of similar meaning, namely -rjde from OE -r&den (834), the form -hode being another result of these blendings. In OE -had is used only with nouns, but its ME and MnE representatives form derivatives from adjectives also, such as hardihood, likelihood, falsehood. Many of the derivatives from nouns have taken concrete mostly collective meanings, as in priesthood and the new-formations brotherhood, neighbourhood. The ME form -hjde is now almost extinct, surviving only in maiden- head and Godhead. 834. -rseden fern., gen. -rizdenne, from the noun r&den ' regulation/ ' agreement ' [connected with ger&dan ' put in order/ ' arrange ' and the MnE ready] : geferr&den ' fellow- ship/ ' agreement/ freondrceden ' relationship/ ' friendship/ In OE this ending was applied only to nouns. ME keeps many of the OE derivatives, frendrgde, sibrjde 'relation- ship ' = OE sibbrizden, and on the analogy of these forms the new derivative hater 'jde, hatrgde ' hatred ' [ME hate is a blending of the OE noun hgte ' violence/ ' hostility ' and the corresponding verb hatian\. The analogy of sibrgde, etc. also led to the ME change of OE cynren line of descendants/ 'family' \-ren=ryne, 'course/ connected with iernan 'run'] into kinrjde, whence, by the usual insertion of /(200), the MnE kindred. 226 ACCIDENCE. [ 83-. 835. -scipe masc. '-ship,' from a lost noun connected with the verb scieppan ' shape,' ' create ' : hlafordscipe ( lordship/ ' authority,' freondscipe, weorpscipe ' honour' \weorj) noun and adjective 'worth/ 'worthy']. This ending is frequently used in MnE to form new derivatives, especially from personal words, as in ownership, consulship, relationship. In OE it is used to form derivatives almost exclusively from nouns, but in MnE we have such derivatives as hardship, courtship from the adjective hard and the verb to court. Adjective-forming. 836. -ede forms compound adjectives from names of parts of the body preceded by a modifying word : sureagede 1 blear-eyed/ literally ' sour -eyed/ frriheafdede ' three-headed.' In MnE this ending has been necessarily shortened to -ed, and so has become indistinguishable from the preterite parti- ciple inflection. 837. -en with mutation generally denotes material, being also used in the more general sense of ' belonging to ' : cecen ' of oak ' \ac ' oak '], gylden ' golden/ wyllen ' woollen ' \wulle /'wool'], h&pen 'heathen' \hce/> 'heath']. In beren from bera ' bear/ as in beren fell ' a bear's skin/ earlier biren, the e has been brought in from the noun ; so also in leaden ' leaden ' [lead ' lead ']. In MnE these adjectives restore the unmutated vowel everywhere, as in golden, woollen, on the analogy of which new derivatives had been formed, such as wooden, hempen. The similarity of meaning between material nouns and adjectives has in some cases led to the conversion of adjectives in -en into nouns, as in //Hf=OE linen ' flaxen' from lin ' flax/ and the tree names aspen = OH cespe, linden = OE lind fem. 839.] NATIVE SUFFIXES. 22? Some adjectives in -en with mutation were originally pre- terite participles of strong verbs : druncen ' intoxicated,' agen ' own,' fcegen 'glad,' whence MnE fain, from drincan, agan ' possess,' gefeon ' rejoice.' 838. -ig ' -y ' corresponds sometimes to Germanic -t'g, -tg, sometimes to Germanic -ag, etc., causing mutation in the former case, but not in the latter : halig ' holy ' \hdl ' entire/ ' sound '], modig ' proud,' isjg ' icy ' ; hejig ' heavy ' [con- nected with hgbban, preterite participle ha/en, ' lift '], bysig ' busy,' dysig ' foolish,' whence MnE dizzy. In MnE this ending has been widely extended, and in many words it has taken the place of the material -en, as in fiery =QE fyren \_fyr ' fire '], clayey, gluey, where the Early MnE spelling -ey is preserved, as it regularly is after vowels. -ig is also a noun-ending, as in bodig ' body,' tfig ( ivy,' hunig ( honey.' 839. -isc ' -ish ' with mutation which is sometimes got rid of by the influence of the underived word is most fre- quently used to form names of nations, but also in derivatives from common nouns : fnglisc, Frgncisc ' French ' \Frandand ' land of the Franks/ ' France '], Scyttisc ' Scotch ' \Scottland, ' Ireland/ afterwards ' Scotland '], Wgtisc ' Welsh ' ; cierlisce mgnn 'serfs' \ceorl ' serf '], folcisc 'popular/ 'vulgar.' In ME some of the names of nations were contracted by omis- sion of the vowel of the ending, whence the MnE French, Scotch by the side of the fuller Scottish, in both of which the unmutated vowel has been restored by the influence of Scot, Scotland. In the other words formed from nouns -ish generally expresses contempt, as in mannish, womanish compared with manly, womanly, childish compared with childlike, brutish, -ish added to an adjective expresses simple Q 2 228 ACCIDENCE. [ 840. diminution, as in oldish, longish, especially with names of colours, such as reddish, yellowish. 840. -sum ' -some ' forms adjectives from nouns, ad- jectives, and verbs; wynsum 'pleasant' \wynn 'joy']; langsum ' tedious ' ; hlersum ' obedient ' \huran ' hear/ ' obey ']. There are many ME and MnE new formations : handsome, troublesome; wholesome, wearisome. The following endings were independent words in Ger- manic : 841. -feald ' rfold ' [compare the verb fealdan ' fold '] forms adjectives from adjectiverwords, especially numerals : manigfeald ' manifold/ ' various/ seofonfeald^ hundfeald, 1 hundredfold/ 842. -full ' -ful/ from the adjective full ' full/ forms adjectives from abstract nouns : carfull ' careful/ sorgfull ' sorrowful/ synnfull ' sinful.' There are numerous new- formations in ME and MnE some from concrete nouns : artful, powerful, masterful. 843. -leas '-less ' from the adjective leas ' deprived of/ ' without ' [compare forleosan ' lose '] forms adjectives from nouns and verbs : geleafleas ' unbelieving/ sl&pleas ' sleep- less ' ; rgcceleas ' careless/ from rgccan ' reck.' In ME this ending appears both as -Ijs and as -les with the vowel shortened, which may be due to the influence of lesse ' less/ It is frequently used in new-formations, such as fearless, useless. 844. -lie '-ly': eorplic 'earthly/ freondlic 'friendly/ gearlic ' annual.' These derivatives were originally con- version-compounds with lie ' body/ the weak vowel being afterwards shortened, so that w'iflic ' feminine/ for instance, meant originally ' having the body or form of a woman ' 848.] NATIVE SUFFIXES. 329 (781). Derivatives in -lie from adjectives and adverbs are less frequent : lablic ' hateful/ uplic, upplic ' sublime/ This ending is freely used in new-formations in ME and MnE, as in princely, quarterly, sickly. 845. -weard, '-ward,' from an obsolete adjective con- nected with weorpan-=. Latin verier e (807), forms adjectives from nouns, adjectives, and adverbs : hamweard, inneweard from ham ' home,' inne ' within.' Verb-forming. 846. -na is a Scandinavian suffix forming weak intransi^ tive verbs, mostly inchoative and from adjectives, as in Ice- landic hvitna ' become white,' harpna ' become hard.' Many of these verbs were imported in ME, such as harbna, which became hardnen by the influence of the ME adjective hard. There have been many new-formations in ME and MnE, some from adjectives, such as gladden, redden, some from nouns, such as lengthen. The native verbs awaken, fasten are not formed direct from wake and fast, but the OE weak verbs awcecnian, fcestnian were formed from the nouns wcecen ' watching,' fasten ' fastness,' ' fort,' which are, of course, derivatives of wacan ' wake ' and fast 'fast,' 'firm.' 847. -sian with mutation : cliensian ( cleanse/ blcedsi'an, bletsian ' bless/ from blod ' blood/ with shortening of the d>, the original meaning being ' to sprinkle (the altar) with blood.' In Scandinavian this ending appears as -sa, as in hreinsa ' purify ' [hreinn ' pure '], whence our rinse. 848. -Isecan from the noun-ending -lac, preserved in MnE only in lOtdlocizsQE wgddldc \wgdd ' pledge/ ' con- tract '] : genealcEcan ' approach' from neah ' near/ geryhtlacan 230 ACCIDENCE. [ 849. ' correct.' In ME a new verb cnjjuljchen was formed with this ending from OE cnawan ' know,' whence in Late ME a noun ctijjuljche was formed which, by the change of weak ch into (ds) gave MnE knowledge. Foreign elements. 849. The foreign derivative elements in English are mainly of French, Latin, and Greek origin. Many which were at first introduced into English in their popular French forms were afterwards latinized, at first in spelling only, but afterwards, in many cases, in pronunciation also. In some cases they were wholly or partially latinized in French itself, though sometimes in Late Old French in spelling only. In some cases false etymological spellings of derivative elements of Latin origin were introduced either in French or English, some of which have corrupted the pronuncia- tion. 850. Although foreign derivatives are often so disguised as no longer to be recognizable as derivatives, yet many foreign derivative elements have remained as distinct as the native ones. Many of them are freely used to form new derivatives from words of native as well as foreign origin. Some of them are even detached and used as independent words, such as extra. PREFIXES. 851. In Latin many of the prefixes are liable to various changes according to the nature of the initial consonants of the word they modify, the full form of a prefix ending in consonants being generally preserved before a vowel, while before consonants the final consonants of the prefix are 854.] FOREIGN PREFIXES. 231 liable to assimilation and loss ; and these variations have generally been preserved when the words containing them were imported into French and English. 852. The foreign prefixes will now be treated of in their alphabetical order. Specially French prefixes are marked *, Greek prefixes are marked f> Latin prefixes being left unmarked. 853. ab-, abs-, a-, ' from, away ' : ab-erration, ab-rupt ; abs-tinent ; a-vert. The above are formed from verb-roots. abnormis, which in English was made into abnormal on the analogy of the Latin adj. normalis, is an example of an ab- derivative from a noun Latin norma ' pattern.' All the above words were taken directly from Latin or from learned French. In popular French ab- became av-, but the Latin form was generally restored, as in abus from abusum, whence MnE abuse. 854. ad-, a-, also in the assimilated forms ag-, af- etc., according to the consonant that follows, ' to.' In Old French this prefix was shortened to a-, as in aventure 'adventure' from Latin res adventura ' a thing about to happen.' The double consonants in such Latin words as aggravdre, as- sentare=adgravare, adsentare were shortened both in pro- nunciation and writing in Old French agrever, asenter double s being, however, often kept (assenter) to show that the s was pronounced (s) and not (z). But in Late Old French the d was often introduced again by the influence of the Latin orthography, whence the spellings adventure, etc. Hence many of these words appear in ME in a variety of forms, one, of early introduction, pure Old French, the other or others more or less latinized, while in some cases the latinized form does not appear till after the ME period. Thus in ME we have aventure and a contracted form aunter, ACCIDENCE. [ 855. in Early MnE adventer, which in the Present English has been further latinized into adventure. The double con- sonants were restored in the same way, sometimes in ME, but generally not till later; thus we have ME agreven, asenlen (also assenteri) = MnE aggrieve, assent. Sometimes the prefix a- was made into ad- from a mistaken etymology, as in advance, advantage =^,1^ avancen, avantage, Old French avancer being a verb formed from the particle avant = Latin *ab-ante. 855. amb-, -am, an-, 'around': amb-ilion; per -am- bulate ; an-cipilal ' two-headed/ ' doubtful/ 856. tamphi- ' around ' : amphi-bious, amphi-theatre. 857. tan- before vowels and h + vowel, a- before other consonants, ' un-' : an-archy [compare mon-archy], an- hydrous ' without water ' ; a-theist. 858. tana- 'up/ 'again/ 'apart/ 'according to/ 're- versal/ etc. : ana-thema, originally ' thing put up or dedi- cated/ ana-baptist, ' re-baptist/ ana-logy ' according to pro- portion/ ana-gram ' transposition of letters/ ana-chronism. 859. ante-, anti- 'before': ante-cedent ' ; anti-cipate. Freely used in new-formations, such as anteroom, antedate. 860. tanti- ' against ': Anti-christ, anti-pathy, anti-podes. Freely used in new-formations, such as anti-radical, anti- spasmodic. 861. tapo-, before vowels ap-, before h aph-, the h itself being dropped : ' from/ ' away/ ' forth/ etc. : apo-cope literally ' cutting away/ apo-stasy ; aph-orism. 862. bi- ' half/ ' twice ' : bi-ennial [compare annual], bi-sect. bi-cycle is a newly formed hybrid from Greek kuklos ' circle.' 863. tcata-, cat-, cath-, ' down/ ' through/ etc. : cata- ract> catastrophe, catalogue ; cat-echize ; catholic. 86?.] FOREIGN PREFIXES. 233 864. circum-, circu- ' round ' : circum-navigate, circum- stance, circumspect, circumvent', circu-itous. 865. cis- 'on this side of: Cisalpine. 866. com-, con-, co- ' with,' ' together/ being another form of the preposition cum ' with ' ; often merely intensitive. In Old French the vowel of this prefix was made into o through the influence of cum, which was often lengthened in ME words taken from French, whence the MnE (B, au) in comfort, council, counsel etc., the (o) in such words as conduit, earlier MnE (k^ndit) being due to the spelling. In Old French as also occasionally in Latin itself the final con- sonant of this prefix was often dropped before consonants, whence the MnE coven/ = convent in Covent Garden [Latin conventio], covenant. The following are further examples of this prefix : commit, comprehend, comfort [Old French com- forter, conforter\ ; convince, conclude, concern, conduct ; coin- cide, cohere; col-league, connect, corrupt. This prefix is used in new-formations, such as corn-mingle, compatriot, especially in the form of co- : co-exist, co-tenant. The predominance of the latter ending has led to the change of contemporary into cotemporary; but the former is now preferred, as being nearer the Latin form. 867. contra-, contro-, *counter- 'against/ originally used only to form verbs. The Old-French form is cuntre- t centre- with the o made into o on the analogy of Old French com-, con-, out of which English counter- has developed in the same way as in counsel etc. But in Old French contre- was often made into contre- by the influence of the Latin spelling. The Latin forms are less frequent than the French : contradict, contravene ; controversy, controvert. The form counter- is used not only in French words, such as counter- feit, counterpart, counterpoise, but also in new-formations, such 234 ACCIDENCE. [368. as cotmler- attraction, counterbalance, counter-revolution, counter is also used as an independent adverb, as in to run counter to, being partly the Old French adverb and preposition contre, partly the detached prefix. 868. de- is partly the Latin (and French) de ' from,' ' away,' also expressing ' difference,' ' negation/ ' completion,' being often only intensitive, which is both a preposition and a prefix ; partly French des-, de- from Latin dis- ' asunder,' 'apart/ which often develops the same negative meaning. de-= Latin de-: degrade, devious, literally 'out of the path/ dethrone. de= Latin dis- (also di-, and assimilated dif-) : defeat [Latin *disfacere, disfactum\, depart, detach. 869. *demi- ' half from Latin dimidium : demigod, demi- semibreve. demy (drmai) is used as an independent word. 870. fdi- ' twice ' : digraph, di-phthong, diploma. The double ss in dissyllable was introduced in French through confusion with the Latin prefix dis-. 871. fdia-, di- 'through': diadem, diameter; diocese, diorama. 872. dis-, di-, assimilated dif- 'asunder/ 'apart/ 'pri- vation/ 'negation.' The Old French form des- [Modern French des-, dt-~\ is still preserved in descant ' tune with modulations/ In the other derivatives taken from Old French the Latin dis- has been restored, as in di'sarm = O\d French desarmer, disdain [compare Modern French de- daigner\, dishonest, disease. The following are of direct Latin origin : discreet, dissolve, distant; different, difficult. The form di- is rare in words of French introduction, such as diminish, and not very frequent in words of Latin form, such as divide, dilate, divert, dis- is freely used in new-formations, such as disconnect, disburden, disheartened. In 879-] FOREIGN PREFIXES. 235 misllken it has been substituted for a similar-sounding native prefix ; so also perhaps in disbelieve, distrust. *en-, *em- ' in ' : see in-, im-. 873. fen-, em-, assimilated el-, ' in ' : encyclopedia, energy ; emblem, emphasis ; ellipse. 874. fendo- ' within ' : endogamous ' marrying within the tribe,' endogenous ' growing from within.' Center- ' between ' ; see inter-. 875. fepi-, ep-j eph- 'upon': epigram, epidemic ; ephemeral. 876. ex-, e-, assimilated ef- ' out of.' The Old French form is es-, Modern French /-. es- has been preserved in English only in a few obscured words, such as essay, escape. Wherever the meaning of the prefix has been kept clear it has been restored to its Latin form in English : exchange from Old French eschangier, extend, extinguish. The other Latin forms are seen in elegant, evade; effect. As ^r = (ks), an initial s is often dropped after ex-, as in ex-pect [compare re~speci\, exude, extirpate [from Latin suddre ' sweat/ stirps 1 stem ']. ex- is frequently used in new-formations to express 'one out of office' etc., as in ex-president, ex-secretary ; so also in the adjective ex-official. 877. fex-, ec- 'out of : exodus] ec-stasy. 878. fexo- ' outside ': exogamous 'marrying outside the tribe,' exoteric 'suitable for outer world, for people in general.' 879. extra- 'beyond' is used in Latin chiefly with ad- jectives : extraordinary, extravagant; so also in the new- formations extra-official, extra-parochial, extra by itself is used in English as an adjective and adverb, being either the Latin adverb and preposition extra ' beyond,' or else the detached prefix. Hence such combinations as extra work, ACCIDENCE. [ 880. extra careful, extra-superfine are not derivatives, but word- groups or compounds. 8 8O. fhyper- ' over/ ' beyond ' : hyperbole, hyperborean, hypercritical. 881. fhypo-, hyp-, hyph- 'under': hypodermic 'be- longing to the parts under the skin/ hypothesis ; hyphen. 882. in-, im-, in-, i-, assimilated il- etc. 'un-' with which it is cognate as well as with Greek an is joined to adjectives and occasionally to nouns. The following are examples of words which had this prefix in Latin itself: insane, insipid, injury] impious, ignorant; illiberal, immortal. In English this prefix is applied only to foreign words of some length, as in inequality, injustice compared with unequal, unjust. If new words are formed from foreign words by means of English endings, un- is prefixed, as in ungrateful, undecided, compared with ingratitude, indecisive. But un- is also prefixed to some words with exceptionally familiar Latin endings such as -able, as in uneatable, uncon- querable compared with intolerable, invincible. 883. in-, im-, il-, etc. 'in/ 'into' is mainly a verb- former. The French form of this prefix is en-, em-, pre- served in English in such words as endure, engage, envoy ; embellish, employ. But in many words of French introduc- tion the Latin form has been restored, as in indite, ME endlten, inquire, imprint. As the spelling makes no differ- ence in the present pronunciation en-, m-=(-in, -im) it fluctuates in some words between the Latin and French forms, the latter being now preferred in such cases of doubt, as in encage, enjoin, entitle, embark, formerly written also incage etc., although impeach now follows the Latin spelling. The following are examples of purely Latin words with this prefix: inaugurate, invade; impel; illuminate, immerse, ir- 889.] FOREIGN PREFIXES. ruption ' breaking in/ There are many new-formations with the French form of the prefix : enlarge ; embody, enliven. In impoverish and improve the Latin form of the prefix has taken the place of the less distinct a- from Latin ad-, the former word being the Old French apovrir [Latin *appauperire\, while the latter is a variation of approve-=^aprove. In a few words, such as inborn, income the prefix is of English origin. 884. inter-, *enter- ' between.' The French form is preserved only in enterprise, entertain, the Latin form having been substituted in all other words of French introduction : intercede, interpret, interval. In intellect, intelligent and their derivatives the Latin assimilation before / is kept, which is disregarded in other words, such as interlude. This prefix is frequently used in new-formations, such as international, intertwist. t 885. intro- 'within,' 'into': introduce, introspection. 886. tmeta-, met-, meth- ' with/ ' after,' ' change ' : metaphysics ' the study that comes after physics/ metamor- phosis; method. 887. ne- ' not ' : nefarious, ne-uter, neutral, 888. non- ' not/ The adverb non 'not' is not used as a derivative in Latin, occurring only as the first element of a few group-compounds such as ndn-nulli ' some/ literally ' not-none.' In Modern French and English it is used as a prefix in such words as nonsense, nonentity, as it already was in Late Latin in non^eniitds, It is freely used in new- formations, such as non-conductor, non-intervention. 889. ob-, o(b)s-, o-, assimilated occ- etc., 'towards/ 'against': ob-stacle, obviate; os-tensible\ omit; occur, offend, opposite. In some cases the full ob- has taken the place of an assimilated form, as in obfuscate. 238 ACCIDENCE. [ 890. *par- see per-. 890. fpara-, par-, parh- 'beside,' 'against': paradox, paragraph ; parenthesis, parody ; parhelion. 891. per- ' through,' occurring also as a preposition. The assimilated pel- is preserved in English only in pellucid. The French form both of the prefix and of the preposition is par, preserved in English only in pardon [Late Latin per- dondre~\ and parboil. In other words the Latin form has been restored, as in perfidy, permit, pervade. W&parfil from Latin perfectus through French par/ait was latinized in WE first into perfit, then into perfec t. *por-, see pro-. 892. post- 'after' : posthumous, postpone, postscript. 893. pre- Latin prae- ' before/ French pre-\ precept, pre- cede, prefer, prescribe, present, pretend. It is freely used in new-formations in the sense of ' before in time,' as in precon- ceive, pre-engage, presuppose. 894. preter- Latin praeter- 'beyond': preter-ile, preter- mit, preternatural. 895. pro-, prod- 'before,' 'forth/ 'away from/ 'depriva- tion/ as in profane literally ' away from the temple/ pro- hibit ' acting as substitute/ as in proconsul, ' relation/ as in proportion, pro 'before/ 'for' etc. is also an independent preposition. There was in Latin an allied prefix por-, as in portendere 'portend.' The popular Old French form of pro- and the preposition pro was pur, por [modern French po ur\, which was probably a blending of pro and per. This form is preserved in such words as purchase [Latin *procaptiare\, pursue', portrait; poursuivant, pursuivant 'state messenger or attendant.' The following are examples of the Latin form : pro-duce, progress, provide ; prod-igy, prodigal. 896. tpro- ' before' : problem, programme, prologue. 9 o5-] FOREIGN PREFIXES. 239 897. fpros- ' towards ' : pros-elyte literally ' coming towards,' prosody. *pur-, see pro-. 898. re-, red-, 'back,' 'repetition,' as in repeat, 'oppo- sition,' as in resist, having often only an intensitive force, as in rejoice. In French re- often became r- before a vowel, but the full form was restored in English, as in reenter from French rentrer. The fuller form red- is preserved in redeem, redolent, redintegrate, re- is freely used in new-formations, such as reconsider, reintroduce, recover an umbrella distinct from the traditional recover. 899. retro- 'backwards': retrograde, retrospection. 900. Be-, Bed- ' apart/ ' away ' : secede, select, separate ; sedition. 901. semi- 'half: semicircle, semicolon. Also in new- formations, such as semi-detached. 902. sine- 'without' : sinecure. 903. sub-, assimilated sue- etc. ' under,' whence a great variety of secondary meanings ' near,' ' behind,' ' following,' ' inferiority,' ' diminution,' ' approaching,' 'help,' ' completion,' the primary meaning also developing into that of ' stealth/ ' secrecy ' : subscribe, suburb, subsequent, subordinate, subdivide, subvention, suborn ; succumb, support, suffice, supply, succour, surreptitious, sub- is freely used in new-formations, such as subcutaneous, subway, especially to express subordination etc., as in sub-committee, sub-editor, sublet, and diminution, as in the adjectives sub-transparent, sub-tropical. 904. subter- 'under': subterfuge. 905. super- ' above/ ' beyond ' became sur- in Old French, which is frequently preserved in English, as in surmount, surpass, surface by the side of its Latin original superficies. It expresses ' beyond in time ' in survive, super- 240 ACCIDENCE. [ 906. annuated. Its most frequent metaphorical meanings are ' addition/ ' excess/ ' superiority/ as in surname [which is an Anglicised form of French surnoni\ ; surfeit, supernatural, superfluous', surpass, super eminent. The Latin form of the prefix is freely used in new-formations, generally to express ' excess ' : super-sensual ' beyond the reach of the senses/ super-phosphate. 906. supra- 'above/ 'beyond': supramundane. *sur-, see super-. 907. sus- has the same meaning as sub-, being a con- traction of subtus '. susceptible, suspend, sustain [Latin sus- tinere through French]. Shortened to su- in su-spect, sus- picion. 908. tsyn-, sy-, assimilated syl- etc. ' with/ ' together ' : synagogue, syntax ; sy-stem ; syl-lable, symmetry. 909. trans-, tra- 'across/ 'through/ ' beyond/ The Old French form is tres-, preserved in English only in trespass, compared with the Latin form of the prefix in transgress, transs- is shortened to trans- as in transcend. Various shades of the primary meaning are seen in such words as transient, transitory, transpire, tra-duce. trans- often ex- presses ' change/ both of place as in transplant, transpose, and of quality as in transform, translate, travesty. It is used in new-formations, as in Transatlantic, tranship. *tres-, see trans-. 910. ultra- ' beyond/ both of place and of quantity and superiority : ultramontane ' beyond the mountains/ that is, ' belonging to the Italian party in the Church of Rome/ ultramarine ' a colour brought from beyond the sea/ ultra- mundane. Freely used in new-formations to express excess : ultra-radical, ultra-clerical, whence the detached ultra has come to be used as an independent adjective in the sense 9 1 3-1 FOREIGN SUFFIXES. 241 of ' extreme,' as in ultra measures, whence the derivatives . ultraist, uliraism, SUFFIXES. 911. The foreign suffixes will now be treated of under the general heads of ' noun-forming ' etc., and the sub- divisions ' personal/ ' abstract,' the suffixes under each section being arranged so that those which consist entirely of vowels come first, and are followed by those that contain consonants in the alphabetic order of those consonants. Noun-forming. Personal. 912. *-ee is the strong form of French -/ from Latin -dlus, and denotes the person who. takes a passive share in an action or agreement, the corresponding active agent being denoted by -or, -er. Thus lessee is the person to whom a house is let on lease, as opposed to the lessor ; so also grantee, legatee, mortgagee. Some of these derivatives have no special active word corresponding to them, such as patentee, referee, trustee. In these words the passive meaning is less prominent, and patentee, for instance, may be taken to mean either ' one to whom a patent is granted/ or ' one who takes out a patent ' ; and in some cases -ee is a purely active suffix as in absentee, devotee, refugee. The weak form of this suffix is -y, -ey, as in attorney =Q\& French atdrne" 924). -iff, see -ive under ' Adjective-forming.' -an, -ean, -ian, -ine, -nt, see under 'Adjective-form- ing.' 913. -ar, -er, -eer, -ier from Latin -drius, -art's, Low Latin R 242 ACCIDENCE. [ 914. -erius, whence the Old French -ier, which in ME became -er. In ME -er was shortened to -er when weak, whence such MnE derivatives as officer, prisoner, stranger. In ME it was often levelled under the English suffix -ere, as in scolere, templere. Many words took the ending -ar through the influence of the original Latin forms, some already in ME, such as vicar, others later, such as scholar, Templar. The MnE -eer, -ier comes from the strong form of the French suffix, both forms being freely used in new-formations, especially -eer : cavalier, cuirassier, muleteer, pamphleteer, volunteer. 914. -or from Latin -or, *-our from Latin -orem, through Old French -or. In Latin this ending is preceded by derivative /, which under certain conditions becomes s : imperdtor, professor. In Old French the / was weakened and then dropped, leaving a hiatus, as in emperetir, sauveor (Latin salvdtorem]. The / was of course kept in learned words of later importation into French, and was reintro- duced into popular words when they were latinized, whence the MnE forms autour, author (210), creditor, orator. In Early MnE the spelling -our was still preserved, but we now write the Latin -or even in words that have not been other- wise latinized, such as emperor, tailor, conqueror = earlier emperour etc., though we still write saviour. 915. -or has in many words taken the place of French -er (as also in some English words, 823) : bachelor [Early MnE bacheler], chancellor, warrior = OE bacheler, Modern French bachelier etc. This is partly the result of -or and -er having the same sound (ar) even in Early MnE (230). In some words the opposite change has taken place, as in miner, robber =1&JL mlnour, robbour. -ary, see under ' Adjective-forming.' 9 1 7-1 FOREIGN SUFFIXES. 243 916. -ard, -art. Although introduced into English from French, this suffix is of Germanic origin. In the Germanic languages -hard ' hard ' in the sense of ' strong,' ' brave/ was a frequent termination of proper names of men, many of which were introduced into Old French, whence they passed into English, such as Richard, Reynard ; Renard was origin- ally a man's name Old High German Reginhart which was given to the fox in the story of ' Renard the fox,' which was introduced into France in the twelfth century from Flanders. In Flemish the name of the fox is Reinaert, which in French became Renart; and the story became so popular in France that renard is now the only French word for fox, the Old French goupil ' fox ' surviving only as a proper name. The name-suffix -ard, -art was soon used in Old French and the other Romance languages to form per- sonal nouns, which were at first nicknames, and had a depreciatory sense. Thus from the Romance forms of Latin cauda ' tail ' was formed Italian codardo, Old French coart ' coward.' Other examples are bastard, wizard, which were imported from French, and English formations such as braggart, drunkard, sluggard. This suffix is used to express nationality in Spaniard, Savoyard, probably at first with an idea of ridicule. It was also used to form names of animals, as in buzzard, mallard ' wild drake ' [formed in French from the adjective male] ; rarely to form names of things, as in petard, poniard [Old French poing ' fist']. -ese, see under ' Adjective-forming.' 917. *-ess, French -esse from Latin -issa denotes female persons and more rarely female animals : goddess, priestess, countess, shepherdess, patroness lioness, tigress. Exceptional formations in point of meaning are Jewess, negress ; mayoress = ' wife of mayor.' Final weak and silent vowels are omitted R 2 244 ACCIDENCE. [918- before this suffix, as in princess, negress, votaress from prince, negro, votary. Nouns in -er, -or often throw out the vowel when -ess is added, as in tigress, actress from tiger, actor. Nouns in -erer, -eror, and some in -urer drop the second of these two weak syllables before -ess, as in sorceress, con- queress from sorcerer, conqueror. Similarly in governess from governor. Some words show further changes : abbess, an- choress from abbot, anchorite; duchess (duke), marchioness (marquis}, mistress (master). 918. t-ist, Latin -ista from Greek -istes, generally ex- presses 'trade/ 'pursuit,' or adherence to a party, dogma etc. : artist, florist, chemist, communist, royalist, deist. It is used in a more general sense in such derivatives as bigamist, copyist, provincialist. In tobacconist from tobacco an n is inserted on the analogy of botanist, mechanist etc., in egotist by the side of egoist a / on that of dramatist, both insertions being prompted by the desire to avoid hiatus. The parallel t-ast in phantast, enthusiast. 919. t-ite, Latin -ita from Greek -ties, is used to form names of nations, sects etc: Canaanite, Israelite, Carmelite] Jacobite. 920. -trix is the Latin fern, of -tor: executrix, testatrix from executor, testator. Diminutive. 921. -ule, -cule : capsule, globule ; animalcule also in the fuller Latin form animalculum corpuscule. The latter ending was shortened to -de in French in most words where the diminutive meaning was not prominent, whence the English article, miracle, spectacle etc. But several of them retain the diminutive meaning, especially where i precedes : cutich l outer thin skin,' particle. 9 26.] FOREIGN SUFFIXES. 245 022. -et, -let. -et forms diminutive nouns and adjectives : circlet, islet, cygnet; dulcet, russet. On the analogy of circlet from circle etc., where the / came to be regarded as part of the suffix, a new diminutive -let has developed itself, which is freely used in new-formations, such as leaflet, streamlet, Iroutlet. In many words these suffixes have lost their diminutive meaning. A bstract. 923. *-y, -ey. -y represents Early MnE, ME and Old French -ie from Latin -ia, and is chiefly used to form abstract nouns, as in fury, modesty, perfidy, and in more popular French words, such as company, courtesy, fancy. Some of these words have more special and concrete meanings, such as comedy, family, navy. -y = Latin -ia is frequent in names of countries, as in Italy, Germany, Normandy, although in most cases the full Latin ending has been restored, as in Asia, India, Austria, -y also corresponds to the Latin neuter ending -turn, as in monastery, remedy, study, forming concrete as well as abstract words. 924. -y is also the MnE representative of weak ME -t^ (Bratntnar, Sixth Edition, extra fcap. 8vo, price Two Shillings & Sixpence Sixth Edition, Revised and Enlarged AN ANGLO-SAXON READER IN PROSE AND VERSE Extra fcap. 8vo, cloth, price Eight Shillings and Sixpence A SECOND ANGLO-SAXON READER ARCHAIC AND DIALECTAL Extra fcap. 8vo, price Four Shillings and Sixpence Old-English Reading Primers, No. i SELECTED HOMILIES OF CLERIC Extra fcap. 8vo, stiff covers, price One Shilling & Sixpence WORKS BY HENRY SWEET, M.A., PH.D. LL.D. Old-English Reading Primers, No. 2 EXTRACTS FROM ALFRED'S OROSIUS Extra fcap. 8vo, stiff covers, price One Shilling & Sixpence FIRST MIDDLE ENGLISH PRIMER EXTRACTS FROM THE ANCREN RIWLE AND ORMULUM J)t'$ (Bratntnar an& Extra fcap. 8vo, cloth, price Two Shillings SECOND MIDDLE ENGLISH PRIMER EXTRACTS FROM CHAUCER (H)t^ (Brammar anb (Bfoeearg SECOND EDITION Extra fcap. 8vo, price Two Shillings and Sixpence ELEMENTARBUCH DBS GESPROCHENEN ENGLISCH ((Bramtnatift, ^jiit unfc Zweite verbesserte aufage Extra fcap. 8vo, stiff covers, price Two Shillings & Sixpence WORKS BY HENRY SWEET, M.A., PH.D. LL.D. A PRIMER OF SPOKEN ENGLISH Extra fcap. 8vo, price Three Shillings and Sixpence AN ICELANDIC PRIMER (EDtflJ (Bratnmar, (ttofee, anb (Bfoeerarg Extra fcap. 8vo, price Three Shillings and Sixpence A PRIMER OF PHONETICS Extra fcap. 8vo, cloth, price Three Shillings and Sixpence A MANUAL OF CURRENT SHORTHAND ORTHOGRAPHIC AND PHONETIC IMMEDIATELY Ojeforfc AT THE CLARENDON PRESS LONDON : HENRY FROWDE OXFORD UNIVERSITY PRESS WAREHOUSE, AMEN CORNER, B.C. CLARENDON PRESS SCHOOL BOOKS (All books are in extra f cap 8vo unless otherwise described} ENGLISH School Dictionaries Concise Etymological Dictionary, by w. w. SKEAT. A new edition (1901), rewritten throughout and arranged alpha- betically. Crown 8vo, 676 pp. 5s. 6d. Saturday Review: ' Mr. SkeatY larger dictionary has estab- lished his title to the gratitude of all scholars ; and of his smaller dictionary we can only say that it is not less useful and valuable.' Student's Dictionary of Anglo-Saxon, by H. SWEET. Small 4to. 233 pp., printed in 3 columns. 8s. 6d. net. Notes and Queries : ' For the purpose of the student, no work so trustworthy, so convenient, and so valuable has seen the light.' Concise Dictionary of Middle English, from A.D. 1150 to A.n. 1580 ; intended to be used as a glossary to the Clarendon Press Specimens of English Literature, etc. ; by A. L. MAYHEW and W. W. SKEAT. Crown 8vo. 7s. 6d. Dr. Sweet's Grammars New English Grammar, logical and historical, in two parts, sold separately : Part I, Introduction, Phonology and Accidence, crown 8vo, second edition, 523 pp., 10s. 6d. Part II, Syntax, crown 8vo, second edition, 146 pp., 3s. 6d. School World: ' As an English grammar the book is of high value ; as an historical study it is of the deepest interest, while its clearness and careful style make it as readable to the literary man as to the grammatical student.' Short Historical English Grammar. 272 PP . 4s. ed, Guardian: ' In the best sense of the word a scholarly book one that, we hope, will for a long time exercise its influence on the teaching of English.' Educational Times : ' Excellent in every way.' Primer of Historical English Grammar, including History of English, Phonology, Accidence, Composition, and Derivation, with Specimens of Old, Middle, and Modern English added. 120 pp. 2s. L. IO.OOO CLARENDON PRESS SCHOOL BOOKS Dr. Sweet's Primers and Readers First Steps in Anglo-Saxon, containing 25 pages of grammar, 43 of text, and 40 of explanatory notes. 2s. 6d. Anglo-Saxon Primer. With grammar and glossary. Eighth edition revised. 126 pp. 2s. 6d. Anglo-Saxon Reader, in prose and verse. With grammar, metre, notes, and glossary. Seventh edition, revised and enlarged (1898). Crown 8vo. 414 pp. 9s. 6d. A Second Anglo-Saxon Reader, archaic and dialectal. 220 pp. 4s. 6d. Old English Reading Primers, being supplements to the Anglo-Saxon Readers. I : Selected Homilies of ^Elfric. Second edition. 2s. II : Extracts from Alfred's Orosius. Second edition. 2s. First Middle English Primer, with grammar and glossary. Second edition. 2s. 6d. Second Middle English Primer : extracts from Chaucer, with grammar and glossary. Second edition. 2s. 6d. Primer of Phonetics. Second edition (1903). 3s. 6d. Educational Times : ' A concise, definite and practical primer, eminently the book for a beginner.' Primer of Spoken English. Second ed. revised. 3s. 6d. A Book for the Beginner in Anglo-Saxon. By J. EARLE. Fourth edition (1903). 2s. 6d. A Primer of English Etymology. By w. w. SKEAT. Fourth and revised edition (1904). Stiff covers. 120pp. Is. 6d. A Primer of Classical and English Philology. By W. W. SKEAT. Cloth, 2s. 2 ENGLISH Annotated Texts Old and Middle English Laurence Minot's Poems, edited by J. HALL. Second edition. 4s. 6d. Gospel of St. Luke in Anglo-Saxon, edited by J. W. BRIGHT. 5s. Selections from Gower's Confessio Amantis, edited by G. C. MACAULAY (1903). 302 pp. 4s. 6d. Miracle Plays, Moralities and Interludes, being specimens of the pre-Elizabethan drama. Edited, with intro- duction, notes, and glossary, by A. W. POLT.ARD. Fourth edition (1903), with ten illustrations. Crown 8vo. 7s. fid. Specimens of Early English : with introductions, notes, and glossarial index. Part I : From Old English Homilies to King Horn (A.D. 1150 to A.D. 1300): by R. MORRIS. Second edition. 572pp. 9s. Part II : From Robert of Gloucester to Gower(\.r>. 1298 to A.D. 1393): by R. MORRIS and W. W. SKEAT. Fourth edition revised. 530 pp. 7s. 6d. Part III : From the Ploughman's Crede to the Shepheards Calendar (A.D. 1394 to A.D. 1579): by W. W. SKEAT. Sixth edition. 582 pp. 7s. 6d. Prof. Skeat's editions The Oxford Chaucer, containing in one volume the com- plete text of Chaucer's works; with introduction and glossarial index. Crown 8vo. 906 pp. 3s. 6d. On India paper, from 5s. The Minor Poems of Chaucer. With notes, etc. Crown 8vo. Second edition. 586 pp. 10s. 6d. The Hous of Fame. Crown 8vo. ise pp. 2s. The Legend of Good Women. Crown svo. 286pp. 6s. The Prologue, the Knightes Tale, the Nonne Prestes Tale, from the Canterbury Tales. R. MORRIS'S edition, re-edited. 324 pp. 2s. 6d. The Prologue. School edition. 96 pp. Is. 3 CLARENDON PRESS SCHOOL BOOKS The Prioresses Tale, Sir Thopas, the Monkes Tale, the Clerkes Tale, the Squieres Tale, etc. Seventh edition. 412 pp. 4s. 6d. The Tale of the Man of Lawe, the Pardoneres Tale, the Second Nonnes Tale, the Chanouns Yemannes Tale, from the Canterbury Tales. New edition revised (1904). 4s. 6d. Langland's Piers the Plowman. Sixth edition. 264 pp. 4s. 6d. The Tale of Gamelyn. Second edition. 104 pp. is. ed. AVycliffe's Bible I Job, Psalms, Proverbs, Ecclesiastes, and the Song of Solomon. 3s. 6d. The New Testament. 6s. The Lay of Havelok the Dane (1003). with two illustrations. 4s. 6d. The Dream of the Rood. An Old English poem attri- buted to Cynewulf. Edited by AI.HKRT S. COOK. :?s. 6d. Elizabethan M Ore's Utopia, edited, with introduction, notes, and full glossary (by Miss MURRAY), by J. CHURTOX Coi.i.ixs (1904). Crown 8vo. 3s. 6d. Elizabethan Critical Essays, selected and edited by GREGORY SMITH : with introduction on the value of Elizabethan criticism and notes. Crown 8vo, 2 vols. 12s. net. Specimens of the Elizabethan Drama. From Lyi\ to Shirley, A.D. 1580 to A.D. 1642. Edited, with introductions and notes, by W. H. WILLIAMS. Crown 8vo. 7s. (id. The Oxford Shakespeare, containing the complete text of Shakespeare's works, edited, with glossary, by W. J. CRAIC. 3s. 6d. 1264 pp. Crown 8vo. On India paper, from 5s. Select Plays of Shakespeare, stiff covers. Edited by W. G. CLARK and W. ALDIS WRIGHT. Hamlet. 2s. Merchant of Venice. Is. Macbeth. Is. 6d. Richard the Second. Is. (id. Edited by W. ALDIS WRIGHT. As You Like It. Is. 6d. King John. Is. 6d. Coriolanus. 2s. 6d. King Lear. Is. 6d. Henry the Eighth. 2s. Midsummer Night's Dream. Is. 6d. Henry the Fifth. 2s. Much Ado about Nothing. Is. 6d. Henry the Fourth, Part I. 2s. Richard the Third. 2s. 6d. Julius Caesar. 2s. Tempest. Is. 6d. Twelfth Night. Is. 6d. 4 ENGLISH Marlowe's Edward 1 1 , edited, with introduction and notes, by O. W. TANCOCK. Third edition. 2s. and 3s. Marlowe's Dr. Faustus and Greene's Friar Bacon and Friar Blingay, edited by A. W. WARD. Fourth edition (1901). Crown 8vo. 448 pp. 6s. 6d. Spenser's Faery Queene, Books i and 11, with intro- duction and notes by G. W. KITCHIN, and glossary by A. L. MAYHEW. 2s. 6d. each. Hakluyt's Principal Navigations : being narratives of the Voyages of the Elizabethan Seamen to America. Selection edited by E. J. PAYNE, containing the voyages of Gilbert, Hawkins, Drake, Frobisher, Raleigh and others. Crown 8vo, with portraits. First and second series. Second edition. 324 and 350 pp. 5s. each. Specimens from 1394 to 15r9 : see p. 3. Bacon's Advancement of Learning, edited by W. AI.DIS WRIGHT. Crown 8vo, with woodcuts. 424pp. 3s. (id. Bacon's Essays, by S. H. REYNOLDS. 8vo. 12s. 6d. Shakespeare as a Dramatic Artist. By R. G. Mon.TON. Third edition, enlarged. Crown 8vo. 7s. 6d. Seventeenth Century The Oxford Milton, edited by H. C. BERTHING. Demy 8vo, with facsimiles, 7s. 6d. ; crown 8vo, 3s. 6d. ; or, India paper, from 5s. ; miniature edition, on India paper, 3s. (id. Milton's Poems, edited by R. C. BBOWNE. 422 and 344 pp. Two volumes, 6s. 6d. ; or separately, vol. I, 4s., vol. II, 3s. Paradise Lost I Book I, edited by H. C. BEECHING. Is. 6d. Book II, edited by E. K. CHAMBERS. Is. 6d. Together, 2s. 6d. SaniSOn AgOnistCS, edited by J. CHURTON COLLINS. Stiff covers. Is. In paper covers Lycidas, 3d. ; Comus, 6d. : edited Lycidas, 6d. ; L' Allegro, 4d. ; II by R. C. BROWNE. Penseroso, 4d. ; Comus, Is. : edited by O. EI.TOX. ArCOpagitica, edited by J. W. HALES. 3s. 5 CLARENDON PRESS SCHOOL BOOKS Pilgrim's Progress, and Grace Abounding, edited, with biographical introduction and notes, by E. VENABLES. Second edit., revised by M. PEACOCK. Cr.Svo, with portrait. 3s. 6d. Holy War and the Heavenly Footman, by M. PEACOCK. 3s. 6d. Clarendon's History of the Rebellion, Book vi, edited by T. ARNOLD. Second edition. Crown 8vo. 5s. Selections from Dryden, including Astraea Redux, Annus Mirabilis, Absalom and Achitophel, Religio Laici, and The Hind and the Panther : edited by W. D. CHRISTIE. Fifth edition, revised by C. H. FIRTH. 372 pp. 3s. 6d. Dryden's Essays, selected and edited by W. P. KER (1900). Two volumes crown 8vo. 404 and 324 pp. 10s. 6d. Dramatic Poesy, edited by T. ARNOLD. Third edition (1904) revised by W. T. ARNOLD. 3s. 6d. Manchester Guardian: 'In its new form this book ought long to hold its place as the standard separate edition of one of the two or three finest achievements of English criticism.' Milton's PrOSOdy, by R. BRIDGES. Crown 8vo. 5s. net. Eighteenth Century Locke's Conduct of the Understanding, edited by T. FOWLER. Third edition. 2s. 6d. Selections from Addison's papers in the Spec- tator. By T. ARNOLD. 560 pp. 4s. 6d. Selections from Steele, being papers from the Tatler, Spectator, and Guardian, edited, with introduction, by AUSTIN DOHSON. Second ed. Cr. 8vo, with portrait. 556 pp. 7s. (id. Selections from Swift, edited, with biographical intro- duction and notes, by Sir HENRY CRAIK, containing the greater part of Tale of a Tub, Gulliver's Travels, Battle of the Books, etc. Two volumes crown 8vo, 484 and 488 pp. 7s. 6d. each. Selections from Pope, with introductions and notes by MARK PATTISON. (1) Essay on Man, sixth edition, Is. 6d. (2) Satires and Epistles, fourth edition, 2s. 6 ENGLISH 's Hermit. Paper covers. 2d. Thomson's Seasons and the Castle of Indolence, edited by J. LOGIE ROBERTSON. Extra fcap. 8vo. 4s. 6d. Also Castle of Indolence separately. Is. 6d. Selections from Gray, edited by EDMUND GOSSE. 3s. With additional notes for schools by F. WATSON. Is. 6d. Gray's Elegy and Ode on Eton College. 2d. Selections from Goldsmith, edited, with introduction and notes, by AUSTIN DOBSON. 3s. 6d. Goldsmith's Traveller, edited by G. BIRKBECK HILL. Stiff covers. Is. The Deserted Village. Paper covers. 2d. Johnson's RaSSelaS, edited, with introduction and notes, by G. BIRKBECK HILL. Cloth flush, 2s. ; also 4s. 6d. Rasselas, and Lives of Dryden and Pope, edited by A. MILNES. 4s. 6d. Lives separately. 2s. 6d. Life Of Milton, edited byC. H. FIRTH. Cloth, 2s. 6d. ; stiff covers, Is. 6d. Vanity of Human Wishes, ed. by E.J.PAYNE. 4d. Selections from Cowper, edited, with a life, introduction, and notes, by H. T. GRIFFITH. 314 and 332 pp. Vol. I : Didactic Poems of 1782, with some minor pieces 1779- 1783. 3s. Vol. II : The Task, with Tirocinium and some minor poems 1784-1799. Third edition. 3s. Selections from Burke, edited by E. J. PAYNE. I : Thoughts on the Present Discontents : the two Speeches on America. Second edition. 4s. 6d. II: Reflections on the French Revolution. Second edition. 5s. Ill : Letters on the proposed Regicide peace. Second ed. 5s. Selections from Burns, edited, with introduction, notes, and glossary, by J. LOGIE ROBERTSON. Crown 8vo. 3s. 6d. 7 CLARENDON PRESS SCHOOL BOOKS Nineteenth Century Byron's Childe Harold, edited by H. F. TOZER. Third edition. 3s. 6d. Keats' Odes, edited by A. C. DOWNER. With four illustra- tions. 3s. 6d. net. Hyperion, Book I, with notes by W.T. ARNOLD. Paper covers. 4d. Scott's Lady of the Lake, edited by W. MINTO. 3s. 6d. Lay Of the Last Minstrel, by the same editor. Second edition. Is. 6d. Separately, introduction and Canto I. 6d. Lord Of the Isles, edited by T. BAYXE. 2s. and 2s. 6d. Marmion, by the same editor. 3s. 6d. Ivanhoe, edited by C. E. THEODOSIUS. Crown 8vo. 2s. Talisman, edited by H. B. GEORGE. Crown 8vo. 2s. Shelley's Adonais, edited by W. M. ROSSEITI and A. O. PRICKARD. Second edition (1904). Crown 8vo. 3s. 6d. Campbell's Gertrude of Wyoming, edited by H. M. FrrzGiBBOx. Second edition. Is. A Vords worth's White Doe of Rylstone, etc., edited by WILLIAM KNIGHT. 2s. 6d. The Oxford Book of English Verse A .. 1250-1900. Chosen and edited by A. T. QUILLER-COUCH. 1096 pp. In two editions. Crown 8vo, gilt top. 7s. 6d. (fifth impression). Fcap 8vo on Oxford India paper, cloth extra, gilt top. 10s. 6d. (sixth impression). Also in leather bindings. Typical Selections from the best English writers with intro- ductory notices. Second edition. 3s. 6d. each. Vol. I : Latimer to Berkeley. Vol. II : Pope to Macaulay. The Treasury of Sacred Song. By F. T. PALGHAVE. 4s. 6d. On India paper, 7s. 6d. PoeniS Of English Country Life, selected and edited by H. B. GEORGE and W. H. HADOW. Crown 8vo. 2s. The Complete Poetical Works of Shelley, including materials never before printed in any edition of the Poems. Edited, with Textual Notes and Bibliographical List of Editions, by THOMAS HI-TCHINSOX. 8vo, with portrait of Shelley and two other collotype illustrations. 7s. t>d. net. 8 UCSOU' ,THERN REGIONAL UB^VFAC.UT| uiiiiH"'''' v-o/ o-ift