UC-NRLF 
 
 $B 13 hEfl 
 
 o 
 
 o 
 
 o 
 >- 
 
FROM -THE- LIBRARY- OF- 
 A. W. Ryder 
 
p 
 
THE 
 
 CRITICAL, COMPARATIVE, AID HISTORICAL METHOD 
 
 OF 
 
 INQUIKY, 
 
 As applied to Sanskrit ScholarsMp and PMology and 
 Indian Archeology, 
 
 Being a lecture read at a Public Meeting held under the 
 
 auspices of the Free Church College Literary Society 
 
 of Bombay on the 31st of March ^ 1888. 
 
 RAMKRISHNA G, BHANDARKAR M. A., Ph. D. 
 
 &c., &c., &c., 
 
 Professor of Sanskrit, deccan college, poona. 
 
 BOMBAY. 
 
 Printed at the " Nirnaya-sagara " press. 
 1888. 
 

THE CRITICAL, COMPARATIVE, AND HISTORICAL METHOD OF INQUIRY. 
 
 A critical inquirer is one who does not accept an account of an 
 occurrence just as it is presented to him whether orally or in writing. 
 He subjects it to certain tests calculated to prove its truth or other- 
 wise. He takes care for instance to ascertain whether the person 
 giving the account was an eyewitness to the occurrence, and if so 
 whether he was an unprejudiced and at the same time an intelligent 
 observer. If his information is based on other sources, the critic endea- 
 vours to ascertain the credibility or otherwise of those sources. When 
 it is a thing or a verified occurrence that he has to deal with, he does 
 not satisfy himself with that view of its nature and relations that ap- 
 pears plausible at first sight. He seeks for extraneous assistance to en- 
 able himself to arrive at a correct view. One of the most efficacious 
 means employed by him is comparison of like things or occurrences. This 
 comparison enables him to separate the accidents of the thing or occur- 
 rence from its essential nature, and sometimes to arrive at a law which 
 includes the thing or occurrence as a particular case and explains it. 
 Though comparison may thus be considered one of the means of a criti- 
 cal examination, still its own proper results are so important that it 
 deserves to be considered an independent method of inquiry. The his- 
 tory of a thing i. e. a comparison of the various forms it presents at dif- 
 ferent well-ascertained periods, is also of the greatest use for the deter- 
 mination of its real nature. Often, when no written history is avail- 
 able, the inquirer determines the historical or chronological relations 
 between the several forms of a thing by referring them to an obvious 
 standard, and arrives at some important conclusions based on such re- 
 lations. Thus for instance, the Greek esmes' 'we are' is more ancient than 
 the Sanskrit smas, for it retains the original a of the root as in the form 
 of e which Sanskrit has lost ; and the Sanskrit santi ' they are, ' more 
 ancient than the Greek * entl ' which has lost the whole root as. From 
 an observation of many such forms the conclusion to be drawn is that 
 Greek is not derived from Sanskrit, nor Sanskrit from Greek, but both 
 from an older form of speech. The comparative and historical methods 
 correspond to the inductive method used in the physical and experimental 
 sciences. In those branches of knowledge in which you cannot from 
 the nature of the case make experiments, you have recourse to compari- 
 son and historical observation. The inductive method began to be used 
 in Europe about the end of the sixteenth century, and since that time 
 
 M43342 
 
[ 2 ] 
 
 very great progress has been made in the discovery of the laws of the 
 physical world. The critical, comparative, and historical methods began 
 to be well understood and employed about the end of the eighteenth 
 century, and within a hundred years since that time, an equally amaz- 
 ing progress has been made in other departments of knowledge ; and 
 geology, paleontology, comparative philology or the science of language, 
 comparative mythology, evolution and the origin of species, scientific 
 history, comparative jurisprudence, archeology, sound scholarship, and 
 even comparative religion are the grand results. Before the employ- 
 ment of the inductive and experimental method such theories as that 
 nature abhors a vacuum passed current ; and before the application 
 of the comparative and historical method the beliefs that the world 
 was created in six days and that the Hebrew was the primitive 
 language of which all the rest were ofishoots were equally prevalent. 
 But just as an experiment with a tube longer than thirty-three feet, in 
 the case of water, was enough to explode that theory, so did the observa- 
 tion and comparison of the different strata of which the earth is com- 
 posed and the discovery of Sanskrit and its comparison with Greek, 
 Latin, and other languages, dispel those beliefs. And this critical and 
 comparative method is necessary not only for increasing our knowledge 
 of the world and of historical man, but also for arriving at correct 
 views of things in ordinary practical life. I must use criticism and 
 comparison if I wish to have a true knowledge of the character of any 
 man public or private, or to understand any individual action of his 
 correctly. Criticism and comparison are necessary for the politician, the 
 legislator, the lawyer, the merchant, and, last but not least, the news- 
 paper writer if he is to rise above the level of scurrilous journalism. 
 
 Criticism and comparison are of use not only in enabling us to 
 arrive at a knowledge of what is true, but also of what is good and 
 rational. A man born in a certain country with certain social and 
 religious customs and institutions, and in a certain range of ideas, 
 thinks those customs, institutions, and that range of ideas to be perfect- 
 ly good and rational, and sees nothing objectionable in them, unless he 
 is a man of genius. When, however, he comes to know of other customs, 
 other institutions, and other ideas, and compares them with those to 
 which he has become accustomed, he is able to find out any evil that 
 there may be in the latter, and to see what is better and more rational. 
 The comparison of the jurisprudence of different countries is calculated 
 to afford valuable hints to the legislator for the improvement of the 
 laws of his own country. Similarly, the critical observation and com- 
 
[ 3 ] 
 
 parison of the social institutions of other countries and even of other 
 religions will afford guidance to the social and religious reformer. Critical 
 comparison is also of use in giving us juster notions of the beautiful. 
 These general observations, applicable as they are to all the branches 
 of knowledge I have indicated, I will expand and illustrate by taking 
 instances from those subjects to which I have devoted some attention, 
 viz. Sanskrit scholarship and philology, and Indian archeology. 
 
 Before admitting the narrative contained in an ancient work to be 
 historical, one ought to ask himself whether the object of the author 
 was to please and instruct the reader and excite the feeling of wonder, 
 or to record events as they occurred. If the former, the narrative can- 
 not be accepted as historical, but legendary. Our obvious and almost 
 axiomatic notions of ordinary probability should also be brought to 
 bear on the question. If a king, for instance, in such a narrative is 
 represented to have reigned a thousand, or even two or three hundred 
 years, one ought to understand that the author wants to excite the 
 feeling of wonder and admiration in his reader, and was in all like- 
 lihood under the influence of that feeling himself. If we apply these 
 tests to our existing Sanskrit literature, we must declare the Rama- 
 yana, the Mah^bharata, and the Pur^nas to be not historical works. 
 Of course, it is possible that they may have a historical basis, and 
 some of the persons mentioned in them may have really existed ; but 
 we cannot assert that they did exist, without corroborative evidence 
 such as is to be derived from contemporary inscriptions and the historical 
 writings of foreigners. Now, if the object of the author be the latter, 
 and the narrative answers to our tests of ordinary probability, the work 
 must be accepted as historical. But we have very few such works in 
 Sanskrit literature now extant. Probably, there were many more, but 
 they are lost to us. The Vikramankacharita, the Harshacharita, the 
 Gaudavadha, and the Rajatarangini are works of this nature. I will also 
 include deeds of grants inscribed on metallic plates, stone-inscriptions, 
 and coins among the historical documents now available to us. It ap- 
 pears to have been the custom in ancient times as it is even now to pre- 
 serve genealogies of royal families. We find some given in the Pur^nas. 
 These have a historical value as they are confirmed by inscriptions. 
 But as the readings have in the course of time become corrupt, and 
 the genealogies of different royal families seem to have been confused 
 together, they are not to be relied on implicitly, without check and 
 comparison. Now as to the contemporary charitras or the deeds of 
 grants spoken of above, it ought to be remembered that the writers being 
 
[ 4 ] 
 
 dependents or servants of the princes whose account is given therein, can- 
 not be expected to be impartial historians of their patrons and masters ; 
 and must be regarded as open to the temptation of bestowing extrava- 
 gant praise on them and their ancestors. Accordingly, the virtues that 
 they ascribe to the princes in the most general terms cannot safely 
 be accepted as historically true ; but the specific statements such as 
 those of their being at war with certain other princes or of their having 
 constructed certain public works must be ; and if corroborative evidence 
 becomes available, we find them confirmed. Thus, all the copperplate 
 grants of the successors of Pulakes'i II of the Chalukya dynasty of 
 Maharashtra, who ascended the throne in the year 610 a. c, speak of 
 his having defeated or remained unsubdued by Harshavardhana the 
 sovereign lord of the whole of Uttarapatha or Northern India, and 
 obtained in consequence the title of ParamesVara or sovereign lord. 
 This is confirmed by what the Chinese pilgrim Hwlian Thsang who 
 travelled in India from 629 a. c. to 645 a. c. tells us about the 
 prince, S'iladitya, as Harshavardhana is also named by him, invited 
 the ablest generals, and sending a large army under them, and on one 
 occasion taking the command himself, fought with the people of Maha- 
 rashtra who were at that time ruled over by Pulakes'i, but he was not 
 able to conquer them. Now, the point to be considered with reference to 
 such a book as the Rajatarangint is that though the author is to be con- 
 sidered a contemporary historian so far as the period in which he lived 
 is concerned, what were his authorities for the history of previous times 1 
 He does mention previous writers and speak of having consulted eight 
 historical works. But he begins his history with Gonarda I, who was 
 the contemporary of Yudhishthira, and gives three names after him. The 
 next 35 princes are, he says, unknown by name ; and then mentions 13 
 more. This is the period for which, he says, he did not find full autho- 
 rities, and mentions the books from which he got the 1 7 names given by 
 him. The next period begins in 1184 B. c. when a prince of the name 
 of Gonarda III, ascended the throne. The history is then carried on by 
 Kalhana without a break up to his own times. One of the princes, how- 
 ever, is represented by him to have reigned for 300 years ; and the aver- 
 age duration of the reigns of the princes in the different groups is some- 
 times 48 years, sometimes 38. When it is remembered that this varies 
 from 18 to 22 only, the chronology of Kalhana in the older portion of 
 his history must be considered not reliable. Though it appears very 
 probable that he himself did not put on paper anything for which he 
 found no authority, the works he consulted cannot be considered to be 
 
[ 5 ] 
 
 quite reliable themselves. And, looking generally to the manner in 
 which the text of old works gets corrupt in the course of time, this is 
 perfectly intelligible. Still, since Kalhana mentions his having used in- 
 scriptions, and edicts or proclamations of kings, and states with what 
 public works in Kas'mir the names of some of the princes are connected 
 and makes specific statements about them of another nature, the nar- 
 rative portion of his history should I think be considered generally 
 reliable, and also the chronology of the period nearer to his own time. 
 But the older chronology and even to a smaller extent the latter re- 
 quire rectification, as we have seen from the internal evidence ; and 
 there is ample external evidence also. For Kalhana mentions three 
 Turushka or Scythian kings who reigned before his historical period, i. e. 
 before 1184 b. c. and whose names he gives as Huslika, Jushka, and 
 Kanishka. There are coins of all these kings, and inscriptions also, from 
 which it appears that the last was the founder of the dynasty. Kalhana 
 mentions Buddhism as flourishing in Kas'mir during the reigns of these 
 monarchs, and represents them to have constructed monasteries in a 
 country in the vicinity which appears to me to be Afghanistan. Now 
 here. Buddhistic records and traditions, which represent Kanishka as a 
 great patron of the religion, confirm Kalhana's account. But his chro 
 nology is entirely wrong. For, from the evidence of the inscriptions, 
 coins, and Buddhistic traditions, Kanishka has been placed about the 
 end of the first century of the Christian era ; and I have found reason 
 to refer him to the middle of the second. Similarly, a comparison with 
 Chinese chronology which is believed to be very accurate has led to a 
 correction in the date of a later prince named LalitSditya who conquered 
 Yas'ovarman, the sovereign of Kanoj and the patron of Bhavabhuti 
 and Vakpatiraja the author of the Gaudavadha, and who, according to 
 Kalhana's chronology, reigned from 695 to 732 a. c. The Chinese ac- 
 count represents his brother who was king before him to have sent an 
 embassy to China in 713 a. c; while according to Kalhana that brother 
 died in 691 a. c. But even supposing that the Chinese account is not 
 correct. Prof. Jacobi has recently calculated the date of an annular 
 eclipse of the sun which is represented in the Gaudavadha to have 
 occurred about the time when Yas'ovarman's position in his kingdom 
 was shaken, apparently by a foreign invader whom that scholar 
 understands to be Lalitaditya. The eclipse occurred on the 14th of 
 August, 733 A. c. and reached its middle at Kanoj at 40 minutes after 
 4 o'clock in the afternoon. Now, if the danger to Yas'ovarman's posi- 
 tion that is spoken of, was really that caused by Lalitaditya, which ap- 
 
[ 6 ] 
 
 pears indeed very likely, Lalitaditya must have been living and in the 
 height of his glory in 733 a. c, while the Rajatarangini represents 
 him to have died in 732 a. c. 
 
 Here I have given you a specimen of the sort of criticism to which 
 books or documents of a professedly historical character ought to be sub- 
 jected. As to the other class of works — the Ramayana, the Mahabha- 
 rata, and all poems, plays, and religious books such as the Vedas,' — 
 though the narrative therein contained is not historical, still they are 
 of great historical value in so far as they place before us the thoughts 
 and feelings, the aims and aspirations, and the manners and customs 
 of the people at or, possibly in some cases, before the time when 
 they were written, and thus present to us a picture of the life and 
 civilization of the period. If, for instance, woven or sewn garments 
 ships, chariots, and weapons of war such as a sword and an axe, are 
 incidentally mentioned in the Eigveda, it must be inferred that these 
 were used at the time when the hymns in which they are mention- 
 ed were composed, and the arts of manufacturing them were known 
 and practised. If the Atharvaveda tells us that " when a woman has 
 had one husband before, and gets another, they shall not be separated 
 if they offer the ajapanchaudana offering," we may safely infer that 
 the practice of widow-marriages existed in those days. We may 
 arrive at the same conclusion, and also infer the existence of polygamy 
 in some cases at least and the absence of polyandry, from the words of the 
 Aitareya Br^hmana that " one [ man ] may have many wives ; but one 
 [ woman ] cannot have many husbands simultaneously. " When we 
 find that the Mantras or formulas contained in the Rigveda, which the 
 bridegroom addresses to the bride on the occasion of marriage, contain 
 allusions to the indissoluble character of the union into which they have 
 entered and to their bringing up a family of children, it must be sup- 
 posed that when these Mantras or formulas were composed, the two 
 parties to the marriage were not children but capable of understanding 
 and appreciating the sense -of what one says to the other. When in the 
 Ramayana and Mahabharata Kshatriya girls are represented as calling 
 a meeting of their possible lovers and choosing that one from among 
 them that they liked, the custom must be considered as existing at 
 or before the time when they were written. In Bana's Kadambari the 
 Queen Vilasavati is spoken of as going to the temple of Mahakala and 
 hearing the Mahabharata read. This shows that our present custom of 
 reading that work and others of the kind in temples for the edification 
 of visitors existed in the middle of the seventh century after Christ 
 
[ 7 ] 
 
 when Bana wrote ; and his mentioning the Bhagavadgita as a part of 
 the Mahabharata shows that the work is older than the seventh century 
 and formed an episode of the epic then as it does now. From the fact 
 that Rama, the hero of that very popular epic the Ramayana, is re- 
 presented as a rigid adherent of truth, regardless of the worldly sacri- 
 fices that such a course entails, as unswerving in his attachment and 
 fidelity to his one wife, and as possessed of such equanimity and 
 firmness of character that " when he was invited to his father's palace 
 to be installed as Crown-prince but was instead sent to live in a 
 forest, divested of all worldly greatness and enjoyments, there was 
 not the slightest change observable in his countenance," the conclusion 
 is legitimate that the higher Hindu mind loved and admired thes® 
 virtues. A wife's faithful devotion to her husband and her following 
 him cheerfully through dire vicissitudes of fortune, with her love and 
 respect for him unabated, were equally objects of admiration with 
 Hindus, as is shown by the characters of Sita, Damayanti and Tara. 
 
 Perhaps the distinction between the ways in which the two classes 
 of works are used for historical purposes will be best illustrated by a 
 specimen of the mistakes which writers in the vernacular papers make 
 on the point. One such writer waxing warm in the course of a disserta- 
 tion against the Poena High school for girls said, " What, had we not 
 learned women in ancient times such as Gargi, Sita, and Tara ? But did 
 they attend a High school ? What then is the necessity of a High school 
 in these days f And on another occasion in that same paper, another or 
 the same writer in giving advice to the women of these days said some- 
 thing to this effect, " You are ambitious of rivalling Gargi, Sita, and 
 Tara ; but you should attend to your domestic duties first." I do not 
 remember the exact words now, but tliis is certain that the writer 
 made no distinction between Gargt, and Sita or Tara as historical per- 
 sonages. Sita and Tara occur in poetical or legendary works ; and 
 though they may have existed, we have no grounds for believing that 
 they did. They are not represented as possessed of learning, though they 
 were educated women ; but supposing they were, all that it would prove 
 is that in those times it was possible for women to become learned, and 
 some did become so. But the existence of Gargi as a historical perso- 
 nage, as a woman of learning, and as a Vedic teacher, is not a matter of 
 reasonable doubt. For a Eigvedin Brahman has, as a portion of his 
 daily religious exercises, to recite portions of the Yedas and other sacred 
 treatises, and pour water in the name of and, technically, for the satisfac- 
 tion of three classes of beings viz : deities, Rishis, and Acharyas. This 
 
[ 8 ] 
 
 is prescribed in the Grihya Sutra of AsValayana and the names are 
 given there. The Rishis mentioned are the composers of the Rigveda 
 hymns, — Gritsamada, VisVamitra, Vasishtha &c. ; and since these 
 hymns exist, tlieir authors also must have existed. The Acharyas are 
 the Vedic teachers, or writers as we should call them in these days, 
 and in the list there are the names of authors of works called Brahma- 
 nas and Sutras some of which have come down to our times. Thiis, 
 S'akalya whose name occurs in the list was the author of the ex- 
 isting Pada text of the Rigveda ; Paingya wrote a treatise on the 
 sacrificial rites which is mentioned by the grammarians ; AsValayana 
 did the same and his work is now extant; Aitareya and Samkhyayana 
 were the authors of Brahmaiias which are extant ; S'aunaka wrote the 
 Rigveda Pratis'akhya; and Bashkala and S'akala were the names of 
 teachers of two different recensions of the Rigveda. All these being 
 historical personages, the three women, Gargi Yachaknavi, Sulabha 
 Maitreyi, and Vadava Pratitheyi, whose names occur in that same list 
 of Acharyas or teachers, must be so likewise though tlieir works have 
 not come down to us. 
 
 Since then it is from our ancient literature that we have to find 
 out the history of our people, it becomes a thing of the greatest import- 
 ance to determine the form in which the works we are concerned with 
 were written by their authors. The readings of passages are not the 
 same in different manuscripts of the same work. A variety of this na- 
 ture we find to a pretty large extent even in ordinary works such as 
 the poems and dramatic plays. Sometimes, numbers of whole stanzas 
 are to be found in certain manuscripts and not in others, as in the case 
 of Kalidasa's S akuntala. But in the Ramayana and the Mahabharata, 
 the two most important works of the non-Yedic period, the differ- 
 ences are bewildering. No two manuscripts from different parts of the 
 country agree. Often there are in some whole chapters which are 
 omitted in others. Sometimes new stories or legends are found. It is 
 therefore, the duty of the critical scholar to collect manuscripts from 
 different parts of the country and collate them, with a view to arrive 
 at a correct text. In the performance of this task which often is very 
 laborious, he ought to be guided by definite principles. He should, in 
 choosing or rejecting a certain reading or a certain passage, see whether 
 it gives good sense, whether it agrees with the context, whether it is in 
 keeping with the author's general way of thinking, whether it is found 
 in the oldest manuscripts, whether the idea or mode of expression was 
 current in the author's time, whether it involves redundance or tau- 
 
[ 9 ] 
 
 tology, and so on. In some parts of India such as Bengal, people have 
 been in the habit of taking very great liberties with the text of their 
 author ; and in Southern India they have been as a general rule very con- 
 servative. This fact I would take into account in determining the correct 
 text of a work. When, however, the differences are very great, the best 
 way is to treat the texts found in different provinces as independent 
 editions or rescensions. And this is what the old Indian Yedic scholars 
 did. In the Vedic texts also different readings sprang up in the course 
 of time ; and one active cause of this was the fact that they were handed 
 down orally. But when scholars like S akalya, S'aunaka, Katyayana, 
 and others arose and gave a definite form to the Vedic literature, 
 they took into account these differences and established separate re- 
 scensions or editions of the texts, known to us now by the name of 
 S'akhas which represent the forms of the text taught in different schools, 
 such as those of S'akala, Bashkala, the Taittiriyas, the Maitrayaniyas, 
 the Kanvas, the Madhyaradinas &c. It is the business of a scholar of 
 the present day to compare these several texts, as well as the several re- 
 scensions of later works, determine the circumstances under which they 
 arose and their mutual relations, literary as well as chronological, and 
 find out which of them have a greater claim to be considered genuine 
 and original. 
 
 Now as to the mode of interpretation of the texts so settled. The 
 first rule is that a word as occuring in a book must be interpreted in 
 the sense which usage has given to it. Etymology may serve as a 
 guide; but it ought never to be set above usage. Consequently, no word 
 should be understood in an etymological sense only. Oftentimes it is 
 difficult to find the correct etymology, and a man has recourse to one 
 that is fanciful. An interpretation of a book based on such fanciful 
 etymologies must be incorrect. Then again, the literature of a country is 
 divisible into periods, and the usage of one period differs from that of an- 
 other. A word therefore occurring in a certain book should be understood 
 in the sense which it has in the usage of the period in which the book 
 was written. A better way still is to interpret it in the sense in which the 
 author himself uses it in other parts of his work. Our oldest literary 
 period is that of the Vedas, and this again is clearly divisible into the 
 period of the hymns and the period of the Brahmanas. The language of 
 the hymns is archaic and very different from the later Sanskrit ; that of 
 the Brahmanas is much nearer to the classical Sanskrit. The hymns 
 contain a great many words which do not occur in later Sanskrit, and 
 there are also a good many which have a different sense there from that 
 2 
 
t 10 ] 
 
 which they have in the latter. Under these circumstances, the only 
 proper way to understand the hymns is to bring together and com- 
 pare the passages in which the same word occurs, taking etymology as 
 a guide only where necessary. In the same manner, the ideas and 
 modes of thinking, which from our acquaintance with the period we 
 have seen to be prevalent, should be referred to for help in interpreting 
 a passage. If instead of resorting to these methods we take an isolated 
 passage and interpret it according to modern usage, modern ideas, and 
 fanciful or even true etymology, we may make it mean anything ; 
 and we shall thus find in the Vedas not only pure theism, but even 
 railways and electric telegraphs. These observations are also applicable 
 to works belonging to other periods. 
 
 Having disposed of books and other written documents, I will now 
 endeavour to estimate the value of traditions. If we accept traditions 
 as we find them, we shall often be deceived. There is a tradition 
 among us, for instance, that Bhavabhuti and Kalidasa were contempo 
 raries ; but we have now found that Bhavabhuti lived in the first 
 quarter of the eighth century of the Christian Era and Kalidasa long 
 before 634 a. c, since he is mentioned as a famous poet in an inscription 
 bearing that date and also by Bana who lived in the middle of the seventh 
 century. Bhavabhuti, it has recently been discovered, was a pupil of 
 Kumarilabhatta; and Kumarilabhatta quotes from Kalidasa's S'akuntala. 
 Similarly, the authors, Dhanvantarin, Kshapanaka, Amarasimha, Kali- 
 dasa, Yarahamihira, Yararuehi, and others who formed the traditional 
 nine gems are said to have lived at the Court of Yikramaditya the found- 
 er of the Sam vat Era the initial date of which is 57 b. c. But Yara 
 hamihira, as stated by Amaraja in his commentary on Brahmagupta's 
 Khandakhadya, died in S'aka 509 or 587 A. c; and the epoch year of 
 his Panchasiddhantika is 427 S'aka corresponding to 505 a. c. And 
 the existence of Yikramaditya in 57 b. c. has not yet been traced. 
 Albiruni, who accompanied Mahmud of Ghizni in his invasion of Guja- 
 rat in the early part of the eleventh century, reports a tradition that 
 the Gupta era dates from the extermination of the Gupta dynasty. 
 But princes of the Gupta dynasty date their inscriptions in their era, 
 which therefore was in use before their extermination. Tradition 
 makes Panini a contemporary of Katyayana and represents him a» 
 having been a blockhead who by austerities pleased S'iva and obtained 
 a new grammar from him. Katyayana held a disputation with him for 
 eight days and, though Panini proved a powerful antagonist, he was 
 ultimately vanquished. The fact, however, as we know it, is that Kat 
 
[ 11 ] 
 
 yayana is the writer of Yartikas or comments on Panini's grammar in 
 which he explains, rectifies, and supplements Panini's rules. It was in 
 consequence of his having thus rectified and supplemented Panini that 
 the story of his having vanquished him must have arisen,-and traditions 
 often have a basis of this nature and no better-but it is impossible that 
 he sliould have written those comments a good many of which are ex- 
 planatory on the work of a contemporary. Besides, from a comparison 
 of the works of the two it appears that the Sanskrit language was in a 
 somewhat more archaic condition in the time of Panini. Hence, there 
 fore, the tradition cannot be true ; and there are also inherent impro- 
 babilities in it. Sometimes the tradition of one sect contradicts that of 
 another. The S'vetambara Jainas, for instance, claim to represent the 
 original Jainism and speak of the Digambara sect as having been found- 
 ed by one S'ivabhuti in 83 a. c; and give childish explanations of the 
 origin of the two peculiar doctrines of that sect viz: nudity and denial of 
 absolution to women. The Digambaras, on the other hand, represent 
 themselves to be the original Jainas ; and state that a sect of the name 
 of Ardhaphalaka separated from them about 272 b. c, and out of that 
 arose after a long time the S'vetambara sect. Still, however, traditions 
 are not to be entirely rejected. An endeavour should be made to as- 
 certain their antiquity as their credibility must be considered to be 
 proportionate to it; and if they are in themselves probable and stand all 
 critical tests, they may be provisionally accepted. Thus, the tradition 
 about the nine gems has been traced up to the year 1,005 Samvat or 948 
 A. D. In an inscription bearing that date found at Buddha Gaya it is 
 stated that Yikramaditya was a renowned monarch in whose court there 
 were nine learned men celebrated under the epithet of the JVava ratnani 
 or nine jewels of whom Amaradeva was one. That portion of the 
 tradition which refers to Vikramaditya's being a great patron of learn- 
 ing ascends still higher into antiquity. In the introduction to Yasava- 
 datta a work mentioned by Bana in his Harshacharita and consequently 
 written before the middle of the seventh century, Subandhu the 
 author, who also is mentioned by name as a previous poet in the 
 Gaudavadha by Yakpatiraja who lived in the middle of the eighth cen- 
 tury, speaks regretfully of the sweet or tasteful poesy of the time of 
 Yikramaditya having when he lived dried up like the waters of a lake 
 which once was full. And the manner of his regret looks like that of 
 one who lived near to the times of Yikramaditya. But nothing has yet 
 been found to confirm the date; and it occurs only in the Jyotirvida- 
 bharana the author of which pretends to be the great Kalidasa but which 
 
[ 12 1 
 
 has been satisfactorily proved to be but a recent forgery by all scho- 
 lars who have noticed it. Since the tradition is thus confirmed it may be 
 accepted as correct; but the date of the nine jewels and of their patron 
 Yikramaditya should be taken to be that which we have got for Yara- 
 hamihira on unimpeachable evidence. And a great king of the name of 
 Harsha Yikramaditya is mentioned by Kalhana in the Rajataranginl 
 as a paramount Sovereign ruling at Ujjayini. He drove away S'al^as 
 and all Mlechchhas from the country, and was a great patron of learn- 
 ing. According to Kalhana's chronology which in this place is, as I 
 have already observed, very faulty, this Yikramaditya lived in the 
 first half of the second century of the Christian era. But his description 
 of the king suits so well the patron of the nine gems, that we may 
 correct the date assigned by Kalhana in the light of that of Yaraha- 
 mihira, and suppose Harsha Yikramaditya to have reigned at Ujjayini 
 in the middle of the sixth century and patronized Kalidasa, Amara- 
 siiiiha, Yarahamihira, and the rest. And nothing that has yet been dis- 
 covered goes against the contemporaneity of those three ; so that the 
 tradition, when corrected as regards the date of Yikramaditya, agrees 
 with everything that is known and seems highly probable. It may in- 
 cidentally be observed here that Subandhu appears thus to have lived 
 after Yikramaditya and before Bana, that is between 550 and 650 A. c. 
 And this is in keeping with his regretful mode of expression noticed 
 above ; and both together lend strong support to another tradition 
 that he was the son of a sister of Yararuchi, one of the nine gema 
 of Yikramaditya's Court. But the chief use of a tradition is to confirm, 
 corroborate, and strengthen other evidence ; and it should not be put in 
 the place of such evidence. This will be illustrated further on. 
 
 I will now proceed to give instances of our method from philology 
 and point out the errors due to superficial analogies. People seem to 
 think that mere external similarity between a word in one language 
 and another in another language is enough to enable us to decide that 
 the two are one and the same word. Thus I have heard the word 
 Dutch, (Germ.) Deutsch, traced to the Sanskrit word Daily a which means 
 'a demon' or 'an enemy of the gods', Taittirtya identified with Tartar^ 
 Caspian with Kas'yapa, and Chitpavan or Chipte with Copt or Gypt. It 
 is melancholy to see such etymologies and a great many others of the 
 same nature put forth without the slightest evidence, and against all 
 the rules of philological comparison. Now there are languages which 
 bear a close affinity to each other, and there are others which do not. 
 Words in a language belonging to the former group observe certain 
 
[ 13 ] 
 
 laws in assuming the forms which they possess in another belonging to 
 the same group. Thus the High German in which we have the word 
 Deut, sch being an adjectival termination, is related to Sanskrit. But 
 Sanskrit consonantal sounds assume certain forms in German. And 
 there is a law with reference to that point known by the name of 
 Grimm's law. The German d corresponds to th in the Gothic, with 
 which also low German or Anglo-saxon, i. e., the modern English agrees, 
 and to t in Sanskrit. Thus German drei is English three, and Sanskrit 
 f^; Germ, dass, Eng. that, Skr. ^; Germ, du, Eng. thou, Skr. ^ 
 in ^jj;; Germ, durch, Eng. through, Skr. f^jX^^ Germ, durst, Eng. thirsty 
 Skr. ^"f, L e., -^^ &c. Similarly, Germ, t is Eng. d, and Skr. \jf ; as in 
 the instances, Germ, mit, old H. G. mitte, Eng. mid, A. S. midde, Skr. 
 Tp£,7f; Germ, tag, Eng. daij, Skr. some such word as \^. You will thus 
 see that there is a law which determines the forms that Sanskrit words 
 should assume in German, and that law ought to give to daitya some 
 such form as teudh ; so that we should not allow ourselves to be deceived 
 by the mere similarity of daitya and deut. Besides, from all the observa 
 tion that the great founders of comparative philology have made, they 
 have come to the conclusion that the affinity between the Sanskrit and 
 the European languages dates from prehistoric times, i. e., is due to the 
 fact that the ancestors of us all spoke one and the same language be- 
 fore they separated and formed distinct nationalities. This was long 
 before the time when the Vedas were composed. Now the word daitya 
 does not occur in the Vedas and was formed later. How could such a 
 word be found in the German, not having existed in the Sanskrit itself 
 a long time after the Aryan separation 1 There are indeed instances 
 in historical times of independent Aryan communities having again 
 been brought into connection with each other and exchanged a few 
 later words. Thus our astronomical works, and especially those of 
 Varahamihira, contain several Greek terms, as will be hereafter noticed. 
 If we suppose the word daitya to have been communicated to the 
 Germans by us in historic times, is there any record anywhere of our 
 having gone into the forests of Germany and established a kingdom 
 there, as we have of the Bactrian Greeks having established one in the 
 Panjab 1 And are there more such instances ? None has been pointed 
 out. The other etymologies must also be similarly condemned. Are 
 we to suppose that the students of the Taittiriya rescension of the 
 Yajurveda came from Tartary, and our Chitpavan Brahmans from Egypt, 
 and that these foreigners were admitted into our exclusive Hindu com- 
 munity and assigned the highest place ? Is there a tittle of evidence to 
 
[ u ] 
 
 show that 1 The Jews came to our country in very early times, and the 
 Parsis later; but have they become incorporated with our community, 
 taken our Gotras, and become students of our Yedas ? 
 
 There are many instances of false and unscientific Marathi etymo- 
 logies in our books. Unless you resort to comparison and historic 
 observation, and discover laws which explain the particular case before 
 you, your etymology must be empiric. Thus our ordinary root 5p^ 'to 
 sit' has been traced to Skr. ^^ 'to dwell'. Of course ^ and ^ constantly 
 interchange places in our languages and there is no difficulty on that 
 point. The sense, however, of sf^ is not appropriate ; but even that is 
 no great objection. Still, if we have recourse to observation and compa- 
 rison, we shall find that this etymology is wrong. ^^ has another form 
 %^, and since the tendency of languages is always to drop an element 
 and not add, ^^ is an older form of the root; and that it is a form of 
 that same root and not an independent root is shown by such contrac- 
 tions as T^^ 'a she-buifalo' for T^g-, sf^j 'as' for %^, ^p^ 'so' for 
 ^^l" &c. Again, the analogy of the cognate languages also shows that 
 %^ is the real form of the root; for the Gujarati has'^g-, and the Hindi 
 %;3-, though ^j- is more common. Now Marathi, Hindi, and Gujarati 
 have descended from the old Prakrits ; and "^ in the first two and t^ in 
 the last is a combination of the vowels ^ and j brought together by 
 the dropping away of uninitial consonants which forms a characteristic 
 of the Prakrits, as will be seen from the following instances : — 
 Skr. ^t^ certain tree', Pr. ?^^ * M. H. P. %^ G. ^^. 
 
 — ^^^1 'a bullock', • — ^f^ ^^ — %^. 
 
 — ^^^ 'like that', — rTlft^, Ap. ^^— • %^r — 
 
 — ?Tft^ 'a she-buff-alo,' — irft^t M- ^^^ET, H. P. ^^ — ^^. 
 
 — "^f^ 'sister,' — m^T, M. dial. ^n\, 11.%^ —^• 
 So that%g- or'^g- must be ^f^ in Prakrit or gf^^^, since sf and ^ 
 
 are always interchangeable. Now, we know that Skr. ^q-]^^ which 
 means 'to sit' appears in the Prakrit in the form of ^q^^, ^ being 
 changed to ^ as a general rule of which there are many instances, and ^ 
 being dropped as in ^f for ^, ^^ for ^ &c. Then the initial vowel 
 g- is dropped since it is unaccented, as it is in the following cases :— 
 Skr. ^*^?- 'inside', H. vftrTT. M. f^rT^ 
 
 — ;3TT^ 'a water-wheel', H. X'^z, M. x^jz 
 
 * Here and in the following pages M.=: Marathi, H.= Hindi, 
 Ap.=Apabhrams'a, G.=Gujarati, P.=Panjabi, S.=:Sindhi, B.=Ban- 
 gi^li, and 0. = Oriya. 
 
r 15 ] 
 
 Skr. ^t>^-jT;t 'to be soaked', H. \\\^r{{ , M. f^f^, <>• f^lTf 
 
 — 'q^^q' 'collected', H. ^^^ 
 
 — ^^tTini 'a teacher', 'a priest,' M. tfi-Tjjji-, P. x(^ 
 
 — ^3^ft 'above,' M. ^^. 
 
 Here then, there is an analogy at every step, and the M. ^f^ is 
 thus derived from the Skr. :jqrH^- 
 
 Another instance of the necessity of a close observation and com- 
 parison of facts is afforded by the derivation of masculine nouns end- 
 ing in 3TT ^^ Maratht and Hindi. It is supposed that the ^fj of 
 these is a remnant of the c^Jjr of masculine nouns in Prakrit, such as 
 fr^T' ?J'^ ^^- -^^t t^® question is, by what rules of transformation 
 does^,;?^ become ^, and why is it that a great many nouns such as 
 ^ITT 'hand,' q^q- 'foot,' qq^r 'ear', and others are without it 1 Besides, in 
 such a language as the Sindhi, those nouns which end in ^ in Hindi 
 and Marathi end in short ^, while those ending in ^ have ^ for 
 their final ; as ^^, 5|v5 for fT^j ^FT ^^- ^^^ %it '^ horse' for ^f§x, 
 5q^ 'a bead-stead' for T^'m, JT§T 'head', 'top' for j^m &c. The correct 
 way of finding the origin of these forms in ^ appears to me to be the 
 following: — 
 
 In modern vernacular pronunciation there is a law of accentu- 
 ation which has produced important results. The penultimate syllable 
 of a word is in all our dialects pronounced with a stress the tendency of 
 which is to lengthen that syllable and drop the final vowel, and in most 
 of them this tendency has worked itself out. The preceding vowel, 
 however, is not always written long, but still the long or at least the 
 emphasized pronunciation does exist. Thus nu^ ' virtue ' is pro- 
 nounced by us as tjo^ in M., 5j^ in H., ^^ 'a flower' as r^^ :^, 'strength* 
 as ^.^ ( baal ), q-^jf 'god of love' as ;q^ ( madaan ), &c. In these in- 
 stances, though we write the penultimate vowel short, it is really long. 
 The final ^ or ^ of Sanskrit words recently imported into the languages 
 have been dropped in virtue of this law of accentuation. Thus : — 
 
 M. H. G. tr^^ 'method,' 'mode', Skr. q^f^. 
 
 M. H. G. P. TTrT 'condition, Skr. jrf^. 
 
 M. G. ^w, H. P. ^^, 'misery,' Skr. ^%. 
 
 M. H. P. G. B. fpT 'manner,' Skr. frf^. 
 
 M. H. G. P. B. ^jfiTT 'species', 'caste,' Skr. ^\{^ , 
 
 M. H. ^^, H. G. P. -^TTrf, 'fame,' Skr. ^^. 
 
 M. H. P. G. xi^ 'a heap,' Skr. ^f%. 
 
 M. H. P. G. ^ 'morality,' Skr. ^%. 
 
 M. G. ^^, H. P. ^^, 'a thing,' Skr. ^^. 
 
[ 16 ] 
 
 H. P. ^^ 'a good man', Skr. ^^^. 
 
 M. G. H. P. jTci 'honey', Skr. ^^. 
 
 Not only does this law characterise the vernacular speech of the 
 day, but it must have been in operation for centuries, since the old 
 Prakrit words, which like the above have not recently been imported 
 but have descended to the modern languages from the spoken dialects of 
 ancient times, have also been similarly changed. Thus : — 
 
 M. B. v^, H. G. B. ^^, P. ^?3-, 0. ifj^, 'hunger,' Pr. ^p^^, Skr. 
 
 M. H. P. G. r^-q, B. 0. t%^, S. f^-q, ' tongue, ' Pr. fi^^, Skr. f^. 
 
 H. P. %;;t, M. G. %:if, S. ^TT^JTr^ ^ ' bed, ' Pr. ^:xrT» ^kr. ^[rj^. 
 
 M. ^, H. G. -^7^, P. f^j^ or :i^, B. O. f^, 'alms,' Pr. fvr^:^, 
 Skr. f^'^T. 
 
 M. fnr, H. p. ;ft^, S. f^, 'sleep,' Pr. f^, Skr. f^. 
 
 M. ^Tl"; H. ^, G. ^, P. ^, B. O. ^, T^ ' trunk of an elephant,' 
 Pr. ^pj^, Skr. jJilT- 
 
 M. G. ^, H. i-f^, 'a jaw,' 'a grinder,' Pr. ^f, Skr. ^^[. 
 
 H. S. ^rr^, B. ;^^ or ^^viT. O. ;^.:f , 'wish', 'longing,' Pr. ^^, Skr. «r?T. 
 
 M. G. ^55, H. ^^, S. ^ft-, B. O. ^^|-, P. ^x, 'dust, Pr. >grf^, Skr. ^,%. 
 
 M. ^^ or ^UT^ H. ^uT, ^^, or ^^r, Gl- %5T, B. fpr, 'sister', Pr. 
 K^, Skr. -iTfTr^- 
 
 M. Ih, H. ^, G. ^, P. ^ or §H, a 'she-buffalo,' Pr. jrf|^, Skr. 
 
 H. P. ^t^, G. gi^, M. ^;^, 'a side of the abdomen,' Pr. ^f^?3"> Skr. 
 
 M. ^[^^, H. i^, 'a fellow wife,' Pr. ^r^, Skr. j^(t^. 
 
 M. ^fUT, H. ^fq-, 'a mine,' Pr. ^^fur, Skr. ;^f^. 
 
 H. ^TRT, P. ^TE^, M. G. ^m^, ' mother-in-law,' Pr. ^j^^, Skr. ^jj^. 
 
 M. G. tt^, P. fir^lT, ' lightning,' Pr. f^^^J, Skr. ^?j^. 
 
 M. gjig-, H. ^^ or fj^, P. ^^, Pr. :j:5f , Skr. |;^. 
 
 H. G. ^m, ^- ^^Wm, B. 0. ^frm, ' the eye,' Pr. ^rf^- Skr. g^. 
 
 M. G. B. 0. fT^, H. ^T^, 1^, or i||^T: ?• ff or ^, Pr. ^y, Skr. 
 
 In this manner the final ^, |^, |^, ;? and :3R: of Sanskrit and Prakrit 
 nouns have, in almost all cases been dropped in the vernaculars or 
 changed to a silent ^. Final ^ is similarly treated. Even in the Apa- 
 bhrams'a period this rule of accentuation must have prevailed, since the 
 
[ 17 ] 
 
 ending vowels are similarly shortened in a good many cases. The Prakrit 
 ^ of the nominative singular of masculine nouns is mostly shortened to 
 g- in that dialect, and sometimes altogether dropped. The modern verna- 
 culars have thus got a great many masculine nouns ending in the silent 
 3^, such as ^m or ^^- ' hand,' ^^ ' ear,' ^f^ ' tooth,' '^ ' bold,' ^ 
 ' the Banyan tree.' When the final vowel is preceded by another, and 
 not by a consonant as in these words and the others given in the above 
 lists, that other vowel being accented by our rule shows a tendency to 
 become long; and the original final being dropped, the accented vowel 
 becomes final. Thus : — 
 
 Skr. frf^^H: 'a pearl,' Pr. ^^^, M. $r#, ^' ?• H. $t^. 
 
 Skr. iKH I ^H. ' water,' Pr. qrfjr^, M. G. j^, H. trpfV- 
 
 Skr. ^f«r^ ' a flowering bush,' Pr. ^f|^T, M. H. G. ^f or :^. 
 
 Skr. f ff^r ' a mare,' Pr. ^^^T, M. G. H. P. tr^T- 
 
 Skr. mf^^J ' a garment,' Pr. ^nfe^j ^- ^- H. ^f^. 
 
 Skr. ^jfWr 'earth,' Pr. j^ftr^, ^rffe^T or fqft^TT, M. qr^, H. ^ff &c. 
 
 Skr. %f^: ' a seller of oil,' Pr. and Ap.>%^f-;j, M. ^^. 
 
 Skr. ;nf^: ' a barber, ' Pr. ;nf^^ or ?f tR^j ^P- ^nf^, M. ?^, 
 H. ^. 
 
 Skr. ^f^sp: ' a scorpion, ' Pr. j^r^^, Ap. Qo.^y, M. ^^, H. P. 
 fir^j S. ^, B. O. f^. 
 
 Skr. ^^ ' a louse, ' Pr. ^^, H. P. G. 3^, M. gfj. 
 
 Skr. qrf^T * an enclosure, ' Pr. ^f^^ or ^T^^, M. G. H. ^T€t> 
 B. ^ft- 
 
 Skr. ^rf^T ' a roll of betel leaf, &c.,' Pr. ^ft^T; M- f^T; ^- r%€r, 
 
 H. ^r 
 
 Skr. vsfr^: ' life,' Pr. ^t^t, Ap. ^, H. ^. 
 Skr. ^ri^Hli: ' blood,' Pr. ^^, G. ^rfr, H. P. §tf . 
 Skr. ^^^ ' a leech,' Pr. ^jT§T^> M. :5t^, H. ^jfr^;, G. ^c^. 
 Skr. ^T^^T ' sand,' Pr. m^^, M. G. ^^, H. ^pj. 
 
 Thus then, the Sanskrit and Prakrit penultimate vowels become 
 final in the vernaculars and, being originally accented in consequence 
 of the law we have been considering, retain that accent in most cases, 
 and are thus lengthened. When the penultimate happens to be ^ it is 
 lengthened to ^f as in 
 
 Skr. -Jt^: ' a horse,' Pr. ^\^^, Ap. §7^, M. H. P. B. O. ^t^. 
 Skr. tHT^: 'mercury,' Pr. q^^, Ap. x^jj^, M. H. P. B. O. tnTT- 
 Skr. ^rm^^: 'a l^i"^<^^ o^ myrobalan,' Pr. 3TTH^^, ^P- 3TT#^^> ^• 
 9ri^55T» H. P. B. ^TT^T^ or an^r^- 
 
[ 18 ] 
 Skr. ^ranRT^: ' hog-plum myrobalan,' Pr. ^ITITT^^T % Ap. ^j:5lT§^, 
 
 Skr. f^^fm^: 'beleric myrobalan,' Pr. ^^, Ap. q|^, M.'^^f^ H. 
 P. 51%^. 
 
 Skr. ^^cT^fnj; ' a volume,' Pr. qtr«r^» Ap. q^^^, H. P. ^fx^x. 
 
 Skr. ^z^: ' a thorn,' Pr. ^^Q^^i, Ap. ^j^, M. H. B. ^f^. 
 
 Skr. %^^: ' a ball,' Pr. Tft^%, Ap. 37^^, M. P. 7\\oS\, H. B. jft^. 
 
 Skr. »^?n^^: ' brother-in-law,' Pr. gr^^T; Ap. ^EH^^; M. P. ^tToST, H. 
 
 Skr. frW: ' ^^ lamp,' Pr. ^^^, Ap. %^^, M. %r, ?• ft^TT, H. 
 
 B.f^qr- 
 
 Skr. JT^^: ' a bedstead,' Pr. jt^^Tj Ap. jj^^, M. ^rf^f 
 
 Skr. i j ^ H ^ ^ ' head, ' Pr. Jfr^^> ^- ( Groan. Mai. and Chit. ) m^j ; 
 by a change of gender, Pr. ^Tr^T^' Ap. j^^:^, M. H. B. jjj^, P. ^t^^xt- 
 
 Skr. ^^^f7i5; ' a wristlet,' Pr. ^f^^, M. ( Goan. Mai. and Chit.) ^^. 
 
 Skr. ^T^^: ' a worm,' Pr. ^% Ap. ^t^, M. ft^, H. P. ^^, 
 B. ^t^. 
 
 It may be urged that in modern pronunciation when the penulti- 
 mate 3T is accented, it does ndt become ^ even though pronounced 
 long, as observed before. How is it then that it becomes ^ here ? In 
 modern times several new modes of pronunciation have arisen, but as 
 regards the matter in hand, to lengthen g:f into ^if was the old process. 
 And often when the old processes have disappeared from what is con- 
 sidered the standard form of a language, they are found preserved in 
 some dialect of that language. Thus, while in the standard Martitlu 
 the penultimate ^ is simply pronounced long, it becomes ^tT "^ the 
 Goanese and Malvani dialects. 
 Thus:— 
 
 St. M. Mai. Goan. 
 
 mWrZ 
 
 ' a garment.' qTrTloS- 
 
 ^rr^ 
 
 * cloth.' qrTTR- 
 
 ^rT^ 
 
 ' hereditary property.' cfrTR- 
 
 ^cT^T 
 
 ' careful preservation.' ^rTF^- 
 
 '•JtrT^ 
 
 ' a garment.' ^ItrTT^- 
 
 ^^^T 
 
 ' a rock.' ?^^TT- 
 
 this manner 
 
 , then, the penultimate -3:r became ^\ in conse- 
 
 In 
 
 quence of the accent and, the final vowel being dropped, itself became the 
 final, and has preserved its accent. Thus the nouns ending in ^f in 
 Marathi and Hindi are derived from Sanskrit nouns with the penulti- 
 mate and final syllables ending in ^. The consonant of the final syl- 
 
[ 19 ] 
 
 lable is dropped in tlie Prakrits, and the vowel ^T if^ together with the 
 nominative termination changed to ^f Tliis ^j* being unaccented, is 
 first shortened to gf in the Apabhrams'a dialect, and afterwards entirely 
 dropped ; while the accented :s^ of the penultimate syllable is lengthen- 
 ed to ^i ond becomes the final. In most cases the final syllable is ^ 
 applied in Sanskrit to a great many nouns to modify the original sense 
 in some way. Instances of those nouns which are seen to have sp as 
 the ending syllable in Sanskrit have been given above, together with 
 their Marathi and Hindi forms in ^. And in those cases in which 
 the Sanskrit forms of other Marathi and Hindi words in ^ are not 
 seen with a final ^, it must have been applied to them in the spoken 
 language, since Panini gives a very general rule as to the addition 
 of ^ to all nouns, and we actually find its remnant ^ in a great many 
 words including past and present participles in the Prakrits. I have 
 spent so much time on this instance in order to give you an idea of 
 the extent to which it is sometimes necessary to make close observa- 
 tion, discover analogies, and trace the operation of laws, to enable us to 
 arrive at correct etymologies. A mere hap-hazard assertion without 
 any proof is unscientific and will not do in philology any more than in 
 other subjects. 
 
 In connection with this matter of insufficient or superficial analo- 
 gies, I may mention that those who are engaged in the studies I have 
 been speaking about are peculiarly open to their influence, especially 
 when they lead to or support a tlieory which is striking. Thus, the 
 Ramayana is supposed by some to represent the struggles between 
 tlxe Brahmanas of India and the Buddhists of Ceylon ; that the Raksha- 
 sas that disturb the rites of the Brahmanas in the Dandakaranya are 
 Buddhists; that the red clothes worn by the priests at the magic rites 
 of Indrajit are the brown garments of Buddhist mendicants &c. Again, 
 Sita's ravishmemt is the same incident as the ravishment of Helen, 
 and S'iva's bow which Rama bent is the bow of Ulysses. Therefore, 
 Valmiki must have been influenced by Homeric ideas, and the poem 
 written after the Hindus came in contact with the Greeks. I cannot 
 stop to give you a detailed account of this controversy ; but will 
 observe that if such analogies are to be used as arguments in favour 
 of a certain theory, an antiquarian in the remote future may declare 
 that the Maratha Hindus had no family names before the British 
 conquered India ; but that they learned to use them, having seen the 
 British doing so. Hence some of their family names are but mere 
 translations of Britisli namp« ; tlnis, Gore or DhavnJe is a translation 
 
[ 20 ] 
 
 of White; Kdle, of Black; Landge, of Wolf; Kolhe, of Fox; Par- 
 vafe or Bongre^ of Hill; Barve, of Wells; Gole, of Ball; Ghainte of 
 Bell, tfec. But as we now know that we had these names before we 
 heard of the British, the theory cannot be advocated at the present 
 day. And to my mind the analogies about the ravishment of the wife 
 of another and the bending of a heavy bow are more natural and less 
 due to intercommunication than this close correspondence of family 
 names. For, what incident is more common in an early condition 
 of society than for one powerful man to take away forcibly the 
 wife of another ? Even in a highly civilized condition of society the 
 incident is not uncommon, though the forces used are of a more deli- 
 cate nature. The same remark applies to a heavy bow. And the state 
 of society pictured in the Ramayana is very archaic ; the whole Maril- 
 tha country was a forest infested by Rakshasas or savages, the an- 
 cestors of our Bhlls and Gonds ; and the gentler races of the South were 
 so uncultivated that they were compared to monkeys. Valmiki does not 
 . mention a single Maratha town, while in the Mahabharata the names 
 of Karhad and Supard occur. He has heard a few names of places in 
 Southern India, but does not seem to be familiar with the geography of 
 the country; while, in the third and second centuries before the Christ- 
 ian era, the Aryas became more familiar with the South, as we see 
 from the inscriptions of As'oka, Pataiij all's Mahabhasliya, and Bud- 
 dhistic tradition. And to this must be added the weight of tlie 
 tradition which represents Valmiki as the first or earliest poet of the 
 non-Vedic Sanskrit. Similarly, resemblances have been traced between 
 the ideas expressed in the Bhagavadgita and those expressed in the 
 new Testament, and a Christian influence detected in that work. But 
 a good many of these resemblances are more apparent than real, the 
 whole tone and manner of the Gita are diflerent from those of the New 
 Testament, and most of the notions suspected to be borrowed from the 
 Bible are found expressed in the Upanishads and sucli older works, as 
 has been shown by the late Dr. Muir. Scholars seem sometimes wlien 
 they have to advocate a tlieory to forget our common humanity to which 
 a great deal that is common in our notions must be attributed. 
 
 On the other hand, when the evidence is irrefragable it is un- 
 scholarlike to deny foreign influence. For instance, the Indian astro- 
 nomical works written during the first five centuries contain several 
 Greek terms. The names of the twelve signs of the Zodiac are trans- 
 lations of the Greek names ; and the original Greek names even are 
 given by Varahamihira in the following Arya quoted by Bhau Daji: — 
 
[ 21 ] 
 
 Wliere \re have Kriya i. e. Krios, Tavuru i. e. Tauros, Jituma 
 i. e. Didymos, Leya i. e. Leon, Parthona L e. Parthenos, Juka i. e. Zyg6n, 
 Kaurpya i. e. Scorpios^ Taukshika i. e. ToxStes, Akokero i. e. Aigokerds^ 
 Hridroga i. e. Hydroclwos, Itthasi i. e. Ichthy's. Some of the other terms 
 are Heli ' the sun ' i. e. Helios, Kona i. e. Kronos, Kendra i, e. Kmtron, 
 Jumitra i. e. Didmetron &c. Altogether there are 36 such terms. These 
 are not Sanskrit words ; and to endeavour to give them an unnatural 
 Sanskrit etymology is vain and unscholarlike. We have the clearest evi- 
 dence of the close connection between Hindus and Greeks from about the 
 tliird century before Christ to the first after Christ, in the inscriptions 
 of As'oka and others, in coins which bear the names of Bactrian Greek 
 monarchs, in Greek as well as Indian characters, and in Buddhistic 
 literature. Garga as quoted by Varahamihira says: — "Tlie Yavanas are 
 Mlechchhas among whom this S'astra {i. e. astronomy and astrology) is 
 well known ; they even are worshipped like Eisliis." The Greeks were 
 at that time called Yavanas ; for in an inscription of As'oka, Antiochus 
 king of Syria is called a Tonarcija. So also Milinda who reigned at 
 S'akala in the Panjab and who has been identified with the Bactro-Greek 
 monarch Menandros is called a Yona king in the Pali books. The 
 Hindus had their own astronomy before they came in contact with the 
 Greeks ; but they borrowed from the latter what was necessary for their 
 own further progress. Prof. Weber thus expresses himself on the 
 point: — "Although most of these names denote astrological relations, 
 still, on the other hand, in the division of the heavens into Zodiacal 
 signs, decani, and degrees, they comprise all that the Hindus lacked, and 
 that was necessary to enable them to cultivate astronomy in a scientific 
 spirit. And accordingly we find that they turned these Greek aids to 
 good account ; rectifying, in the first place, the order of their lunar 
 asterisms whicli was no longer in accordance with reality, so that the 
 two which came last in the old order occupy the two first places in the 
 new ; and even, it would seem, in some points independently advanc- 
 ing astronomical science further than the Greeks themselves did." 
 
 I will now give a specific instance to show how previous history 
 enables us to understand the nature of a thing, though what I have 
 hitherto said in connection with other points contains a good deal 
 calculated to illustrate this also. At present we use an era which is 
 called S'alivahana S'aka ; and ordinarily we understand the word S'aka 
 
[ 22 ] 
 
 in the sense of ' an era,' and believe tliat tlie era was founded by a king 
 of the name of S'alivnhana. But the word S'aka has not the sense of 
 an era in the Sanskrit language; and the expression Salivahana S'aka 
 has been in use for about three or four hundred years only. Before 
 that, and even in some books of the subsequent period, the expression 
 used is S'akakala. Varahamihira calls it S'akendrakala or S'akabhtipa- 
 kala i. e. the era of the S'aka king. In an inscription dated 500 S'aka, 
 it is called the era of the coronation of the S'aka king ; and in another 
 dated 656 S'aka, the era of S'aka kings. In all old copperplate grants 
 the expression used is S'akanripakala or S'akakala i. e. the era of 
 the S'aka king or S'aka. Thus, then, this is an era founded by a 
 great king of the S'aka or Scythian race. And that India was exposed 
 to the inroads of the Scythians, and that they established a kingdom in 
 the country, is proved by many an inscription and coin. There was a 
 dynasty of kings who called themselves Satraps and ruled over Ujjayinl, 
 Kattiawar, and Gujarat. In the beginning at least they were in all 
 likelihood the viceroys of the S'aka kings. 
 
 S'alivahana or S'atavahana was the name of a royal family which ruled 
 over Dakshinapatha or Southern India. The principal branch reigned 
 at Dhanakataka in Telingana, and the younger princes of the family or 
 a subordinate branch ruled over Paithana. The S'atavahanas came in 
 contact with the conquering S'akas who established their power also 
 over a part of the country ruled over by the S'atavahanas. But after a 
 short interval of time one of the princes of this family succeeded in 
 driving away the foreigners and regaining the lost provinces. The S'akas 
 and the S'atavahanas were by these events associated together in the 
 popular memory ; and it must have been on this account that the 
 names of the two families came to be connected in after times with 
 the era which thereafter was called the S'alivahana S'aka. More in- 
 formation, however, is wanted to enable us to understand satisfactorily 
 how the name S'alivahana came to be connected with the era ; but at 
 present we do not possess it. 
 
 Thus, then, the great lesson we have to learn is that if we wish to 
 know and understand the truth about a point, whether in science or 
 practical life, we should seek analogies, find out if we can the history, 
 and criticise, not foolishly and ignorantly as we often do, but according 
 to well defined and rational principles. 
 
 And now gentlemen, and my Hindu friends in particular, a word 
 as to my object in taking up this subject for to-night's discourse. It is 
 
[ 23 ] 
 
 no use ignoring the fact that Europe is far ahead of us in all tliat consti- 
 tutes civilization. And knowledge is one of the elements of civilization. 
 Experimental sciences and the sciences that depend on the critical, com- 
 parative, and historical method have made very great progress in Europe, 
 and what deserves our earnest attention is that they are every day 
 making further and further progress. The Europeans have derived 
 much greater advantage from our connection with them than we have 
 from their connection with us. They have turned to account their 
 acquaintance with the Sacred language of our country, and have added 
 the sciences of comparative philology and comparative mythology to 
 their existing store of knowledge. The old principle of the classifica- 
 tion of races has been given up, and a new one based on the affinity 
 of languages adopted. Civilized mankind has in the first place been 
 divided into three races, — the Aryan, the Semitic, and the Turanian. 
 The first has again been subdivided into the Hindus, the Hellenes, 
 the Italians, the Slavs and Lithuanians, the Kelts, the Teutons includ- 
 ing the Scandinavians and the Germans, &c. So strongly and univers- 
 ally has the fact that affinity of language points to a community of 
 descent and consequently to a common nationality been grasped in 
 Europe that " to it," as Sir Sumner Maine remarks, " we owe, at all 
 events in part, the vast development of German nationality ; and we 
 certainly owe to it the pretensions of the Russian Empire to at least a 
 presidency over all the Slavonic communities." — So that, it may be re- 
 marked by the way, the many wars that have been fought in Europe 
 since 1855, and are likely to be fought during the next twenty years have 
 or will have for one of their causes the discovery of Sanskrit. Though 
 in itself this is by no means a very gratifying result, still I allude 
 to it simply to show how deeply the Europeans have been influenced 
 by the new ideas. Similarly, I think the liberality that is now obser- 
 vable in the religious thought of Europe is due to the study of Sanskrit 
 and Pali literature. All this will show the activity of the European 
 intellect, and convince us that the principle of progress is very strong in 
 their civilization. But what advantages have we derived from them ? 
 A great deal of what they have got from us has but very indirectly 
 been given to them by us ; while they have placed before us a whole 
 civilization which undoubtedly is far superior to ours in a great many 
 points. The impulse to be communicated to us by it ought to be a 
 hundred-fold stronger than that whicli we have communicated to them. 
 Just as they have used the critical and comparative faculty with energy 
 and produced the results I have just noted, we should use it and direct 
 
( 24 ) 
 
 it not only to find what is true in science, but what is good and ra- 
 tional in social and religious institutions. But have we received the 
 impulse, have we been using the faculty ? Who can say we have, while 
 our new literature is scanty and barren of any original idea 'and we are 
 still quarreling about female education, caste, and religion ? Why should 
 we not move on, side by side with Europeans, in the great fields of 
 thought ? Why should discoveries be made in France, Germany, and 
 England, and not in India ? If you say that in most of the branches 
 there are facilities in Europe for making fresh additions to the existing 
 stock of knowledge, while we have none in India, surely no costly labo- 
 ratories are required to enable us to study the ancient literature of our 
 country and its architectural remains and inscriptions, and to throw 
 light on its political and literary history and its philology. This is a 
 field in which we may successfully compete with Europeans, and in 
 which we enjoy certain peculiar advantages. But these advantages can 
 be turned to account only if we follow their critical, comparative, and 
 historical method. My object, therefore, has been to call your atten- 
 tion to the nature and requisites of this method, in order that by its 
 successful application to the branch of study I am speaking of, we may 
 take our legitimate place among the investigators of the political, liter- 
 ary, and religious history of our country, and not allow the Germans, 
 the French, and the English to monopolize the field. And here I feel 
 myself in duty bound, even at the risk of displeasing some of you, to 
 make a passing allusion to the most uncritical spirit that has come over 
 us of praising ourselves and our ancestors indiscriminately, seeing 
 nothing but good in our institutions and in our ancient literature, asserting 
 that the ancient Hindus had made very great progress in all the sciences, 
 physical, moral, and social, and the arts, — greater even by far than 
 Europe has made hitherto — and denying even the most obvious defi- 
 ciencies in our literature, such as the absence of satisfactory historical 
 records, and our most obvious defects. As long as this spirit exists in 
 us, we can never hope to be able to throw light on our ancient history, 
 and on the excellencies and defects of our race, and never hope to rise. 
 While, if we shake ourselves free of such a bias, and critically and im- 
 partially examine our old records and institutions, we shall do very 
 great service to our country; we shall be able to check the conclusions 
 of some European scholars who are swayed by an opposite bias; and 
 at the same time that by a clear perception of our great national 
 defects we prepare the ground for healthy progress in the future, we 
 shall, I promise you, find a great deal in the past of which we may 
 honestly be proud. 
 
( 25 ) 
 
 But an honest and discerning pride in the achievements of our 
 ancestors entails a heavy responsibility and duty. "We should render 
 ourselves the worthy sons of the fathers whom we respect. A son that 
 is no better than the father or is worse certainly dishonors him. Have 
 we then not been dishonoring our ancestors, of whom we profess to be 
 so proud, by going backwards arid thus becoming worse than they, or at 
 the best, standing where they left us 1 For, if you examine your history 
 you will find that your philology is where Panini and Katyayana left 
 it, and your philosophy and literature where Kapila, Kanada, Gotama, 
 Valmiki, Yyasa and others left them ; and your social institutions are 
 actually far more irrational than theirs were in the very olden age. Let 
 us therefore do honour to them by showing that we have their capacities 
 and can use our rare opportunities. We have just seen how fifteen 
 hundred years ago, the Hindus availed themselves of the astronomical 
 knowledge of the Greeks ; they 'worshipped' the Greek astronomers, 
 in the words of Garga, 'as Eishis'; and finally, according to Prof. 
 Weber who is by no means fond of praising us, advanced astrono- 
 mical science further than they did. Let us act likewise, and sitting 
 at the feet of the English, French, and German Rishis imbibe the know- 
 ledge that they have to give, and at least keep pace with them, if not go 
 beyond them. Let us learn, let us reform. If we do not do so, fifteen 
 centuries hence, the antiquarian of the period will, unlike Weber, say, 
 " the English placed before the Indian Aryas the highest civilization 
 which Europe had reached by the end of the nineteenth century ; but 
 in the hot plains of India the Indian Aryas had grown so degenerate, 
 that it produced no influence whatever on them, and their degeneracy 
 deepening, they eventually became hewers of wood and drawers of 
 water, or were swept ofi" the face of the earth by the inexorable law 
 of the survival of the fittest." 
 
RETURN TO: CIRCULATION DEPARTMENT 
 198 Main Stacks 
 
 LOAN PERIOD 1 
 Home Use 
 
 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS. 
 
 Renewals and Recharges may be made 4 days prior to the due date. 
 Books may be renewed by calling 642-3405. 
 
 DUE AS STAMPED BELOW. 
 
 sent ON ILL 
 
 SEP 1 6 200 4 
 
 U . CiiRKELE^ 
 
 FORM NO. DD6 
 50M 4-04 
 
 UNIVERSITY OF CALIFORNIA, BERKELEY 
 Berkeley, California 94720-6000 
 
Gaylamount 
 
 PampHlet 
 
 Binder 
 
 Gaylord Bros.. Inc. 
 
 Stockton, Calif. 
 T.M. Reg. U.S. pat. Off. 
 
 
 »«..?.f,[i'<f'-fY LIBRARIES 
 
 M43342 
 
 THE UNIVERSITY OF CALIFORNIA LIBRARY