ON THE ORIGIN 
 
 OF THE 
 
 INDIAN BRAHMA ALPHABET 
 
 BY 
 
 GEORG BtiHLER. 
 
 SECOND REVISED EDITION OF INDIAN STUDIES, NO III. 
 
 TOGETHEE WITH TWO APPENDICES ON THE ORIGIN 
 
 OF THE KHAROSTHI ALPHABET AND OF THE 
 SO-CALLED LETTER-NUMERALS OF THE BRAHML 
 
 WITH THEEE PLATKS. 
 
 STBASSBURG. 
 
 KARL J. T R U B N E R 
 
 1898. 
 
 Ue Oriental Book-Supplying Agency, Poona.
 
 Printed by Adolf Hokhausen, Vienna.
 
 Preface to the Second Edition. 
 
 As the few separate copies of the Indian Studies No. Ill, 
 struck off in 1895, were sold very soon and rather numerous 
 requests for additional ones were addressed both to me and to 
 the bookseller of the Imperial Academy, Messrs. Carl Gerold's 
 Sohn, I asked the Academy for permission to issue a second 
 edition, which Mr. Karl J. Triibner had consented to publish. 
 My petition was readily granted. In addition Messrs, von Holder, 
 the publishers of the Wiener Zeitschrift fur die Kunde des 
 Morgenlandes, kindly allowed me to reprint my article on the 
 origin of the Kharosthi, which had appeared in vol. IX of that 
 Journal and is now given in Appendix I. To these two sections 
 I have added, in Appendix II, a brief review of the arguments 
 for Dr. Burnell's hypothesis, which derives the so-called letter- 
 numerals or numerical symbols of the Brahma alphabet from 
 the ancient Egyptian numeral signs, together with a third com- 
 parative table, in order to include in this volume all those 
 points, which require fuller discussion, and in order to make 
 it a serviceable companion to the palaeography of the Grund- 
 riss. The chapters on the Brahmi and the Kharosthl have been 
 throughout revised and the first has been changed most. A new 
 comparative table of the Semitic and Brahma signs, 1 the same 
 as has been used for the Grrundriss, has been given. The 
 Additional Note at the end has been omitted, as, since the 
 
 In using the plates, those of the Grundriss ought always to be com- 
 pared, as the signs given there are mechanical reproductions from im- 
 pressions and as such more reliable than any drawn by hand. 
 
 20048S9
 
 IV Preface. 
 
 appearance of M. Sylvain LeVi's article 1 on the Turkish kingdom 
 of Northwestern India, it is no longer required, and a number 
 of other alterations and additions has been made in accordance 
 with the results of further researches. 
 
 Thus the list of the passages from the Jatakas, which 
 mention writing and written documents, has been considerably 
 enlarged, the enlargement having become possible chiefly through 
 references, kindly communicated to me by Professors S. von 
 Oldenburg (p. 7ff.) and Rhys Davids (p. 120). 2 
 
 The extensive and intimate acquaintance of Lieut. Col. 
 R. C. Temple with the actualities of daily Indian life has enabled 
 me to adduce an interesting confirmation of my explanation of 
 the term rupa which occurs in the oldest known Indian trivium 
 (p. 14, note 3). 
 
 A valuable paper by Dr. von Rosthorn, based on Chinese 
 sources, has furnished a correction of the interpretation which 
 I formerly put on Hiuen Tsiang's statement that in the seventh 
 century A. D. the instruction of the young Hindus began with 
 the twelve chang (p. 30). It now appears that the twelve chang 
 were twelve tables of simple and compound letters, of which 
 the varnamdla or matrkaviveka of the period consisted. And a 
 
 1 See Journ. Asiatique, 1895, serie IX, t. VI, p. 380 f. I accept M. LeVi's 
 conclusion that Major Deane's undeciphered inscriptions from Swat are 
 Turkish as highly probable, and I may add that a key prepared from 
 the signs of Sanskrit and Prakrit inscriptions does not fit them. 
 
 2 I regret that vol. VI of Professor Fausbo'll's edition of the Jatakas 
 reached me too late for utilisation. It furnishes a number of additional 
 passages, proving the use of writing, among which that on p. 369 f. is 
 the most interesting. According to the story told there, Amaradevl, the 
 daughter of the Sheth of Yavamajjhaka, noted on a leaf (pany.a), when 
 and by whom king Vedeha's crest-jewel, golden garland, golden slippers 
 and precious rug were sent to her husband, Mahosadha. Afterwards she 
 produced the record before the king to the confusion of his four Pandits, 
 who had accused her husband of the theft of the articles, and whom she 
 had captured and confined in baskets when they came to seduce her. As 
 Prof. Minayeff has first seen, the scene in the king's court is illustrated 
 by the Bharahut relievo, inscribed yavamajhakiyam jatakam, Cunning- 
 ham, Bharhut Stupa, pi. XXV, No 3. The story teaches that already in 
 ancient times, just as in our days, the Vanias allowed their daughters to 
 learn to write, which accomplishment many Hindus still consider dan- 
 gerous for female virtue.
 
 Preface. V 
 
 communication from Dr. B. Liebich (p. 120 f.) has put me in 
 possession of the proof that the Bengal schoolmasters until a 
 very recent period used a set of twelve such tables, called phald 
 or in Sanskrit phalaka, which term the Chinese expressions 
 chang and fan probably are intended to render. 
 
 Dr. Grierson's important researches at Mahabodhi Gaya, 
 the results of which lately have been reprinted in the Proceed- 
 ings of the Asiatic Society of Bengal, 1896, p. 52 ff., have made 
 it necessary to rewrite the passage (p. 3 If.) on the remnants 
 of the masons' alphabet, found there by Sir A. Cunningham, 
 though the general conclusions to be drawn from them remain 
 the same. A communication, kindly placed at my disposal by 
 M. Sylvain Levi, has furnished from Chinese sources a distinct 
 tradition (p. 33), asserting that the signs for the liquid vowels 
 really are later additions to the Brahma alphabet, as the state- 
 ment of the Jaina scriptures regarding the original number of 
 its characters and the palaeographic evidence suggested. Mr. 
 Rapson's discovery of syllables, both in Brahml and Kharosthi, 
 on the Persian sigloi (pp. 51, 113) has further corroborated 
 the conclusions regarding the early prevalence of both alpha- 
 bets in Northwestern India and has raised a strong presumption 
 that both alphabets were used in the same districts already 
 in the fourth century B. C. during the Akhaemenian period. 
 Finally, Mr. Takukusu's article on Pali Elements in Chinese 
 has brought us the news that the tradition, asserting an early 
 preservation of the Buddhist scriptures in MSS., is more ancient 
 than the statement in the Life of Hiuen Tsiang could lead us to 
 suppose, and nearly, if not quite as old as the contradictory state- 
 ment of the Dlpavamsa and the Mahavamsa (p. 91, note). And the 
 most important discovery of a Kharosthi MS. at Khotan shows that 
 during the Kusana period Buddhist MSS. did exist in Northern 
 India and probably had been in use for some time (p. 122f.). 
 
 While the new facts, contained in these recent contribu- 
 tions, have been duly utilised, it has not been possible to pay 
 much attention to M. J. HaleVy's two controversial papers in the 
 Revue Se"mitique of 1895, pp. 223 ff., 372 ff., beyond omitting 
 in note 1 on p. 52 his derivation of the Brahma numeral symbol 
 for 8 (supposed to contain the first two letters of asta), from . 
 the Kharosthi Ma, which, as he now tells us, is a misprint
 
 VI Preface. 
 
 for as (a non-existing and impossible ligature 1 originally in- 
 vented by Dr. I. Taylor), and beyond adding a protest against 
 the derivation of the circular cipher, which belongs not to the 
 ancient Brahma system of numeral symbols but to the later 
 decimal notation, from the Kharosthi da? 
 
 I regret that I have not been able to find in the two, some- 
 what excitedly written, articles any facts or suggestions, likely 
 to benefit serious students of Indian antiquities. While repeating 
 the ingenious, but extravagant, theories of the earlier paper, they 
 teem in addition with statements which, though put forward with 
 the author's characteristic apodictiveness and self-confidence, 
 fill the Sanskritists rather with astonishment than with respect- 
 ful admiration, and which fully justify their continued unwilling- 
 ness to take his opinions on Indian matters into serious con- 
 sideration. There is hardly a single subject of Indian research, 
 on which they do not contain assertions, conflicting with per- 
 fectly well known and undisputed facts. 
 
 The quality of their scholarship and method is perhaps 
 best illustrated by the results of the 'examen serieux' of the age 
 of the Jatakas represented on the Stupas of Sanchi and Bharahut, 3 
 
 1 The supposition that such a ligature is possible, betrays a want of 
 acquaintance with the principles of Indian spelling. The Hindus divide 
 a-xta (not as-fa), i-sta (not is-t.a) u-pta (not up-ta) and so forth. Their 
 ligatures represent elements of one and the same syllable, and hence 
 they do not, and cannot, form ligatures like a? or 06, is, up etc. 
 
 2 It is a matter of course that M. Halevy sticks to the derivation of 
 the Brahma symbols for 4 9 from the Kharosthi initial letters of the 
 Sanskrit numerals. The untoward fact that, except in the case of 5, 
 either the supposed phonetical values or the forms of the signs do not 
 agree with the requirements of the theory, is easily got over. The 
 busy Gandharian inventors of the Brahma symbols, we are told, had 
 no time to look up Panini, and so they put cha for catuh and so forth. 
 Again if the contemporaneous forms of the symbols and the letters will 
 not agree, those of different periods are chosen for comparison, and 
 e. g. the Brahma 6 of the 3 rd cent. B. C. is declared to be a modifica- 
 tion of the "cursive" Kharosthi sa of the 1 st cent. B. C. The equally 
 untoward fact that the Brahma numeral signs include symbols for 20 
 100 and for 1000, which cannot be derived from Kharosthi letters, is 
 carefully kept out of sight. 
 
 8 Nouvelles Observations sur les Ventures Indiennes, p. 18f. (Revue Sm., 
 p. 241 f.).
 
 Preface. VII 
 
 which ; as it would lose in effect by curtailment, I translate in 
 full: "These monuments offer inscriptions in Asoka characters; 
 these insrs consequently cannot be earlier than B. C. 221, 
 because otherwise the writing would run from the right to 
 the left, as in the legend of the Eran coin. How far can the 
 lower limit of their date go? Let us not put it too close to the 
 relic caskets of Bhattiprolu, which are at least fifty years later, 
 and which show already some altered forms, and let us put 
 between the latter and the Stupas an interval of twenty years. 
 The inscriptions of Sanchi and Bharahut date therefore from 
 about 191, i. e, one hundred and thirty four years later than 
 the introduction of writing (quaere, of the Brahma letters'?) 
 into India. It is in this interval that the composition (sic). 
 must necessarily be placed, in the middle of the Greek epoch 
 and in no way in the Persian period. Very luckily for us, the 
 Katahaka-jataka (sic) mentions the writing- board, phalaka, and 
 the texts of the Stupas the canonical term pitaka "a box for 
 tablets of card-board or wood", in which the Jainas usually 
 keep their MSS., 1 with its derivative petaki "he who knows 
 the Pitaka or Pitakas", where naturally not the box, but the 
 tablets or leaves, contained in it, are meant. These are real 
 revelations; for the terms phalaka and pitaka, which have no 
 etymology in Sanskrit, 2 are nothing but the Greek words xXa 
 (-x6q) and Tucnavuov; 3 the identity of sound corresponds to the 
 identity of meaning. The Jatakas (sic) carry on their forehead 
 
 1 "Pitaka is only "a box" and corresponds to the modern dabado or card- 
 board of wood (sic) in which the Jainas usually keep the MSS. of their 
 parish libraries. (O. I. BA., p. 87.)" 
 
 2 "The root phal 'to profit' gives birth to phalaka "profit, gain" and (meta- 
 phorically) "the catamenia", but does not explain the homophonous pha- 
 laka "board, plaque and tablet". 
 
 3 "These two words have the identical meaning of "plaque" and "tablet". 
 In Syriac Dip^o is the same as m^> (Duval, B. B., p. 1575) and pnc (also 
 very common in the Talmud) means "leaflet, leaf". IIitTaxiov is a po- 
 pular word, which has come into literary use only rather late; but its 
 antiquity is guaranteed by the proper name IIiTTaxoj (sic) borne by one 
 of the Seven Sages of Greece, a contemporary of Solon. The word 
 auTtpov "un blanc" 1 presents an analogous case; though it has come late 
 into literary use, one finds it once in an ancient decision of the Misna 
 (Ma'aser seni II IDS >-ICDX) and in the Agpereno of the Avesta."
 
 VIII Preface. 
 
 the date of their birth, to wit, the century which follows the 
 conquest of Alexander." 
 
 What occurs to a Sanskritist with regard to this remark- 
 able statement of results, is as follows: 
 
 (1) The assumption that all the Asoka inscriptions date 
 from B. C. 221, which is the basis of M. Halevy's contention 
 that no inscription in Asoka characters can be older than B. C. 
 221, and which is another version of his statement on p. 12, 
 declaring the Asoka edicts to have been incised about (vers) 
 221, is a sad blunder which a writer on Indian palaeography 
 ought not to make. And it is the more unpardonable, as Pro- 
 fessor Max Mtiller, whose Hibbert Lectures, translated by M. J. 
 Darmesteter, M. Halevy quotes (p. 12, note 1) as his authority, 
 explicitly gives B. C. 221 only as the date of the so-called 
 three New Edicts of Sahasram, Rupnath and Bairat, following 
 my calculations, and does not touch the question of the dates 
 of the other numerous Asoka inscriptions. 
 
 The dates of the incision of the following Asoka inscrip- 
 tions are clearly ascertainable and undisputed, (1) of the Bara- 
 bar Hill Cave inscrs A and 5, dated in Asoka's 13 th year, 
 (2) of the Barabar H. C. inscr. C, dated in the 20 th year, (3) of 
 the Nigliva and Pacjeria pillar inscrs, dated in the 21 st year, 
 (4) of the first six so-called Pillar Edicts, dated in the 27 th 
 year, and (5) of the seventh edict on the Dehli-Sivalik pillar, 
 dated in the 28 th year. If Asoka's coronation is fixed with 
 Professor Max Miiller in B. C. 259, these inscriptions range 
 from B. C. 246 231; according to Professor Lassen's views 
 they would fall each four years earlier, and according to Pro- 
 fessor Kern, six years. As far as the evidence of the Asoka 
 edicts goes, the Sanchi and Bharahut inscriptions might, there- 
 fore, be placed at least a quarter of a century earlier than the 
 'examen se>ieux' assumes. 
 
 (2) An undated fragment of an A&oka edict, found on 
 one of the broken gate-pillars of the Sanchi Stupa, proves that 
 the structure certainly existed before the end of that king's 
 reign or, with Prof. Max Miiller's initial date, before B. C. 222/21, 
 and the presumption is that all the Sanchi inscriptions, show- 
 ing Asoka characters, were incised before that year. By the 
 attempt to fix the erection of the Sanchi Stupa between B. C.
 
 Preface. IX 
 
 221 191, M. Halevy collides with the clear epigraphic evidence, 
 and commits a mistake which he might easily have avoided, 
 if he had looked up the works on Sanchi. His attempt to drag 
 down the age of the Bharahut Stupa goes against the probability, 
 as the inscriptions on the representations of the Jatakas show 
 the same characters as the Asoka edicts. All these represen- 
 tations thus give the reign of Asoka as the lower limit for the 
 existence of the Birth-stories. They teach nothing definite re- 
 garding the time of the composition of these fables, except in 
 so far that they raise the presumption of their having been long 
 current and generally known. For the archaeology of all nations 
 proves that myths and scenes from religious legends are trans- 
 ferred to stone only when they have become thoroughly fami- 
 liar to the people. Consequently, the assertion that the com- 
 position of the Jatakas must "necessarily" be placed between 
 B. C. 221191, is not warranted. 
 
 (3) M. Halevy's attempt to bring the Jatakas into the 
 period after Alexander by deriving their term phalaka from 
 the Greek xXa and the word pitaka (which itself does not 
 occur "in the texts of the Stupas") from TUTWCVUOV. only shows 
 that his knowledge of Sanskrit is open to improvement and 
 that he does not even take the trouble to consult the standard 
 dictionaries. Both words are regular derivatives from Sanskrit 
 roots. Phalaka "a board", comes from the very common verb 
 phal, (phalati) "to split", mentioned in all Sanskrit dictio- 
 naries, and means etymologically "a piece (of wood) split off"; 
 compare the Latin scindula from scindo and the Greek cyJSa, 
 ox(8a!-, oyjSa from cr/'.o (<ryjt*). The etymology is duly given 
 in the larger Petersburg Dictionary under phalaka. Pitaka 
 means primarily and usually "a basket" (used for any pur- 
 pose, as for carrying earth, storing meat, measuring grain, etc.), 
 more rarely "a box", as I have translated it, in which latter 
 sense the cognate petaka is more common. Its employment to 
 denote a receptacle for MSS., or the contents of such a recep- 
 tacle, or a division of the sacred books, is mainly or entirely 
 confined to Buddhistic literature. Pitaka is derived from the 
 Sanskrit root pit (petati) 'to heap up, to store', as the larger 
 Petersburg Dictionary indicates under that verb, and it means 
 etymologically "an implement for heaping up or storing (ob-
 
 X Preface. 
 
 jects)". Whatever may be the truth about the age of the Ja- 
 takas, these two words are certainly useless for M. Halevy's 
 purpose. The triumphant tone, with which he brings forward 
 his supposed discoveries, and the display of misplaced learning 
 in the notes, especially the appeal to king Pittakos' name in 
 support of the antiquity of the !5iamx.u)tpov term Tunax-iov, make 
 the passage most amusing to every scholar, possessed of philo- 
 logical training. 
 
 To this specimen of M. Halevy's scholarship and method 
 may be added another, found in his remarks on Kebuka 
 (N. O. E. I, p. 56), where he lays down the law to a native 
 Pali commentator. "Kebuka, he informs us, is the name of 
 a place belonging to Seruma; the idea of the commentator 
 that it is the name of a river has not the slightest foundation." 
 And he then goes on to treat us to the ingenious conjecture 
 that Kebuka is the Caucasus which, as he alleges, in Persian 
 bears the name Kabak or Kabkh, essentially corresponding to 
 the Pali word. A Sanskritist who reads vs. 106 of the Jataka, 
 No. 327 (Fausboll, vol. Ill, p. 91): 
 
 Katham samuddam patari katham patari Kebukam | 
 
 Katham satta samudddni katham simbalim aruhi || 
 can only translate: 
 
 "How didst thou cross the ocean, how didst thou cross 
 Kebuka, how the seven oceans? How didst thou climb the 
 cotton-tree?" 
 
 and recognises at once from the use of the verb patari 
 "didst thou swim" or "cross in a boat", that Kebuka has 
 something to do with water and can only denote a river, a lake, 
 an ocean, or water in general. The Jataka No. 539 fully confirms 
 this inference, as it says explicitly, kebukam vuccati udakam, 1 
 "kebuka means water". 
 
 If it were worth the while, the list of positive mistakes 
 and more than hazardous speculations might be enlarged ad 
 infinitum by examples from passages on Indian phonetics, epi- 
 graphy, antiquities, Dharma and so forth. Similar instances 
 might also be adduced, where the papers show a weakness in 
 
 1 FausbSll, The Jatakas, vol. VI, p. 42, 1. 11; compare also vs. 133, p. 32, 
 where kebuka is used to denote a tank.
 
 Preface. XI 
 
 other subjects, e. g. in Greek, 1 or collide with the statements 
 of the Greek historians. 2 But what has been given, will be 
 amply sufficient to convince even others than Sanskritists that 
 a fuller discussion of their contents cannot serve any useful 
 purpose. 
 
 Recent utterances of Professors O. Donner, Konow, Lud- 
 wig, Macdonell, Oldenberg, Windisch and others on writing 
 in India and on the Indian alphabets make it apparent that 
 the object, for which the Indian Studies No. Ill and its supple- 
 ment were written, has been attained. Among the Sanskritists 
 it is now more generally recognised that there are good reasons 
 for believing that writing was practised in India long before Near- 
 chus saw the Hindus write on cloth, prepared from their native 
 cotton, and before another unnamed companion of Alexander, 
 whose statements Q. Curtius has preserved, saw them use for 
 the same purpose the tender inner bark of trees, i. e. the well 
 known bhurja from the Himalaya. Equally the conviction has 
 become stronger that the Brahmi is the most ancient Indian 
 alphabet and that the development of the Kharosthl is one of 
 the results of the Persian domination over Gandhara and Sindh. 
 The settlement of the general outlines of the enquiry, to 
 which belongs also the recognition of the fact that the Brahmi 
 is the alphabet of the Pandits and the Kharosthl a script of 
 the clerks and men of business, similar in character to the 
 simpler and defective Nagarl, Lande, Takkari, Kaethl, Mo<Ji 
 and Gujarat! which have been developed out of the more ela- 
 borate Brahmi, some in later and some in earlier times, is 
 all that can be carried through for the present. The details of 
 the derivation, which are particularly difficult in the case of the 
 
 1 See e. g. N. O. E., Note 1 on p. 15, where we find the instructive 
 sentence, "D'ailleurs le nom 'Icavfs (sic), et encore mains la forme non- 
 contractee 'la<ovt$ (sic), n'etait plus usite au temps d'Alexandre comme 
 designation collective des Gh-ecs (sic)". 
 
 2 See e. g. N. O. E. I, p. 60, where the establishment of an administration 
 of the Macedonian type is asserted to have taken place about B. C. 
 330 in the countries west of India, though the Greeks give us the names 
 of the Persians and other Orientals appointed by Alexander as his Satraps, 
 who, as far as is known at present, were not forced to pass previously 
 a Greek Civil Service examination.
 
 XII Preface. 
 
 Brahmi with its forms, pointing partly to very archaic and 
 partly to transitional North -Semitic characters, may at any 
 moment require modifications on account of new epigraphic dis- 
 coveries in India or in Western Asia. And important epigra- 
 phic discoveries will, no doubt, soon be made in India, since 
 of late ancient sites have been identified, which were famous 
 places of pilgrimage before the times of the Mauryas.
 
 Table of Contents. 
 
 Preface, pp. Ill XII. 
 
 I. Former derivations of the Brahml lipi, pp. 15. 
 II. Literary evidence for the antiquity of writing, 
 from the Vasistha Dharmasutra, pp. 5 7, 
 from the Jatakas and the Vinayapitaka, pp. 7 22, 
 from the Lalitavistara and the Jaina Agamas, pp. 2235. 
 
 III. Palaeographic evidence for the antiquity of the Brahml lipi, 
 
 from the Asoka Edicts, pp. 3543, 
 
 from the legend of the Eran coin, running from the right to the left, 
 
 pp. 4445, 
 
 from the Bhattiprolu alphabet, pp. 45 47, 
 from the legends of the Taxila coins, pp. 4750. 
 from the Persian Sigloi, pp. 51 53. 
 
 IV. The derivation of the Brahma letters from the most ancient North- 
 
 Semitic signs. 
 
 General principles, pp. 5356. 
 The borrowed signs, pp. 5672. 
 
 The derivative consonants and initial vowels, pp. 7277. 
 The medial vowels, pp. 7782. 
 The approximate date of the introduction of the Semitic letters into 
 
 India, pp. 8391. 
 
 Appendix I. The Origin of the Kharosthl Alphabet, pp. 92114. 
 Appendix II. The Origin of the ancient Brahma Numerals, pp. 115119. 
 Addenda and Corrections: 
 
 Further passages from the Jatakas, pp. 120124. 
 
 The twelve Phala in the Bengal schools, pp. 120124. 
 
 The Petroffski MS. from Khotaii, pp. 120124.
 
 Ijver since Mr. J. Prinsep succeeded in deciphering the 
 Edicts incised by order of the god-beloved king Piyadasi of 
 Pataliputra on the pillars and rocks of Eastern, Central and 
 Western India, the attention of the European Orientalists has 
 been directed to the question of the origin of their curious 
 alphabet, which is the parent of all those now used from Cape 
 Comorin to the Himalayas and of many others occurring be- 
 yond the confines of India proper. And, while there has been 
 not much difference of opinion regarding the derivation of the 
 second alphabet, in which Piyadasi's servants placarded their 
 master's sermons in the Northwestern corner of his dominions, 
 the views regarding the source of the more common characters 
 have diverged very widely. There has been almost from the 
 beginning a pretty general consensus that the alphabet of the 
 Shabazgarhl and Mansehra Edicts, called by the Europeans the 
 Arian, Ariano-Pali, Bactro-Pali, Gandharian, Northwestern or 
 Northern and by the Hindus Kharosthl lipi, 1 is, as the direction 
 of its letters from the right to the left at once suggests, of Se- 
 mitic origin, and that it has been derived from one of the later 
 types of the Northeastern Semitic alphabet. But for the charac- 
 ters running from the left to the right, called by the Europeans 
 Lath, Southern, Indian Pali, Indian or Maurya and by the 
 [2] Hindus Brdhmi lipi, not less than five different derivations 
 have been proposed, of which a detailed demonstration has 
 been attempted. The number of general, more or less vague, 
 
 1 Regarding the Hindu names of the two alphabets see below p. 23. 
 Bukler, Indian Studies. III. 1
 
 2 Indian Studies. No HI. 
 
 suggestions is even greater. 1 Leaving the latter aside, the five 
 theories may be briefly stated as follows: 
 
 (1) According to Sir A. Cunningham the Indian Pali or 
 Brahma alphabet, is an Indian invention and is based on a 
 system of indigenous hieroglyphics; 2 
 
 (2) According to Professor A. Weber it is derived directly 
 from the oldest Phoenician alphabet; 3 
 
 (3) According to Dr. Deecke it is descended from the 
 Assyrian cuneiform characters through an ancient Southern Se- 
 mitic alphabet, which was also the parent of the Himyaritic; 4 
 
 (4) According to Dr. Isaac Taylor it comes from an alpha- 
 bet of Southern Arabia, the parent of the Himyaritic; 5 
 
 (5) According to M. J. Halevy it is of a composite cha- 
 racter eight consonants having been taken directly from the 
 Aramaic alphabet of the fourth century B. C., six consonants 
 and two initial vowels as well as the medial vowels together 
 with the Anusvara from the Ariano-Pali or Kharosthi, and five 
 from the Greek; and this hotch-potch is alleged to have been 
 concocted about 325 B. C. 6 
 
 It must, however, be noted that the first among these 
 theories seems to have been almost given up by its chief 
 advocate some time before his death. For in his last dis- 
 cussion of the Indian alphabet 7 Sir A. Cunningham says, "The 
 origin of the Indian alphabet is still unsettled. According to 
 Lassen, Dowson, Thomas and myself, its origin was indigen- 
 ous, [3] that is, it was invented by the people of India. But con- 
 tinental scholars are generally in favour of its derivation from 
 some unknown Western source", and in the sequel, after cri- 
 ticising the latest theory, which he takes to be that of Dr. 
 Taylor, he adds, 8 "It seems not improbable that this old Indian 
 
 1 See the exhaustive review of earlier opinions in Dr. R. N. Gust's Ling- 
 uistic and Oriental Essays. Second series, pp. 2752. 
 Corpus Inscr. Indicarum, vol. I, p. 52 S. 
 
 Zeitschrift der Deutschen Morgenlandischen Ges., Bd. X, 389 ff.; Indische 
 Skizzen, p. 225 -250. 
 
 Zeitschrift der Deutschen Morg. Ges., Bd. XXXI, 598 S. 
 The Alphabet, vol. H, p. 314 ff. 
 
 Journ. Asiatique, serie VIII, tome VI (1885), p. 268 ff. 
 Coins of Ancient India, p. 38 f. 
 Op. cit., pag. 41. The italics are mine.
 
 I Former Derivations of the Brahmi. 3 
 
 alphabet, when it was first framed or adopted, did not possess 
 any cerebral letters." These utterances indicate that in 1891 
 Sir A. Cunningham himself no longer felt as certain of the sound- 
 ness of his views as in 1876, when he wrote the introduction 
 to the Corpus Inscriptionum Indicarum. As far as has become 
 known, they have not gained of late any new adherents, and 
 with the death of the illustrious archaeologist they have pro- 
 bably become entirely a matter of the past. Sir A. Cunning- 
 ham himself has furnished a very strong argument in favour 
 of the opposite theory by publishing, op. cit., Plate XI. 18, a 
 coin from Eran, which shows an inscription in Brahma cha- 
 racters running from the right to the left. This is really the 
 link, which was wanted in order to complete the chain of argu- 
 ments, proving the Semitic origin of the Brahma alphabet. 
 
 The remaining theories coincide in the main point that the 
 ancient Indian characters are derived from a Semitic source, 
 and Sir A. Cunningham is no doubt right, when he says that 
 this is the prevailing belief among Sanskritists, not only 
 however among those of the European continent, but pretty 
 nearly all over the Western world. This belief is probably 
 founded not so much on special studies in Indian palaeography, 
 which, as well as epigraphy, are mostly neglected owing to 
 the force of unfavourable circumstances, as on the general im- 
 pression that certain Indian characters strongly resemble Semitic 
 letters and on the conviction, supported in several cases by the 
 clearest evidence, that the Indo- Aryan civilisation includes many 
 and various elements, borrowed from Western nations, Semites, 
 Persians and Greeks. But I doubt that even half a dozen 
 Sanskritists could be found, who would care to make a definite 
 choice between the rival theories, except in so far that 
 they might be inclined to reject M. Halevy's ingenious, but 
 untenable combinations, which rest on a priori improbable as- 
 sumptions [4] and partly on errors regarding facts, and which in 
 their final results, e. g. the conclusion that the Vedas were com- 
 posed in the time of the Mauryas, disagree with all the lessons 
 taught by Indo-Aryan research. 1 
 
 1 With respect to the statement, that M. Halevy's theory has not found 
 much favour with Sanskritists, I would point to Professor A. Weber's 
 remarks in his paper "Die Griechen in Indien", p. 17 (Sitz. Ber. der 
 
 1*
 
 4 Indian Studies. No III. 
 
 My own attitude with respect to this problem has been 
 for a long time exactly the same. During the last fifteen years, 
 whilst I have devoted a portion of my time to early Indian 
 epigraphy, it has been with me an open question whether the 
 Brahma characters came from western or from southern Asia. 
 I have always believed in their Semitic origin. But I have 
 vacillated more than once between their derivation from a 
 pre-Himyaritic alphabet of Arabia and that from the ancient 
 northern Semitic characters, which show almost identical forms 
 in Palestine, Phoenicia, Cyprus and Assyria. And I have he- 
 sitated to take up the enquiry in real earnest, because it seemed 
 to me that one preliminary condition to a new attempt on the 
 problem was the preparation of perfectly trustworthy impres- 
 sions and facsimiles of the oldest Indian inscriptions, and a 
 second, the careful study of all these documents from a pa- 
 laeographic and from a philological point of view. The first 
 condition has now been fulfilled thanks to the unwearied la- 
 bours of Messrs Burgess, Fleet, Fiihrer, Hultzsch, Rice and 
 Senart. Really good facsimiles of all the versions of the Edicts 
 of Asoka have been prepared and mostly published, as well as 
 faithful reproductions of the closely allied, quite or nearly con- 
 temporaneous inscriptions in Dasaratha's caves, on the Bharhut 
 or Bharaut and Sanchi Stupas, on the Ghasundi slab, on the 
 Sohgaurfi Copper plate, in the Hathigumpha, Nanaghat and Pabhosa 
 caves. Moreover, the palaeographic store has been unexpectedly 
 enriched by Mr. Rea's discoveries in the Bhatfiprolu Stupa, 
 which have brought to light a new type of Brahma characters, 
 showing a certain independence, and, as it would seem to me 
 on further consideration, at least some very archaic forms. 
 The explanation of these [5] ancient documents, too, has so far 
 advanced as is requisite for the palaeographic enquiry. 
 
 Under these circumstances I believe it possible to resume 
 the discussion regarding the origin of the Brahma alphabet with 
 some hope of success, and I may state at once that the re- 
 
 Berliner Akademie, 1890, XXXVII), to Professor A. Ludwig's interesting 
 paper on "Yavananl", Sitz. Ber. der k. Bohm. Ges. der Wiss., 1893, No. IX, 
 to Mr. S. Soerensen's Om Sanskrits Stilling i den almindelige Sprogud- 
 vikliug i Indien, Copenhagen 1894, p. 288, note 1, and to Professor Kern's 
 remarks in Dr. Gust's Essay's, p. 39.
 
 II. Literary Evidence for the Use of Writing. 5 
 
 suits, at which I have arrived in general confirm the views of 
 Professor Weber, who has already given the correct identifi- 
 cations for the majority of the signs. In the case of most of 
 the letters it is, of course, now possible to adduce forms which 
 come closer to each other than those which his table contains. 
 This is chiefly due to the numerous discoveries in Semitic epi- 
 graphy, which have been made during the last thirty nine years. 
 Mesa's stone, the oldest Sinjirli inscription and the inscriptions 
 on the Assyrian weights, which are datable more or less ac- 
 curately, each furnish something valuable. And these discoveries 
 make it also possible to adhere strictly to the general rule, to 
 be observed in such inquiries, that only the signs of one period 
 should be chosen for comparison. 
 
 Before I proceed to this comparison, it will be desirable 
 to call attention to some passages in Indian literature, recently 
 made accessible, and to some peculiarities in the oldest forms 
 of the Brahma alphabet, revealed by the new facsimiles and 
 by a tabular arrangement of the signs, which I have lately 
 undertaken for my ,,Grundriss der indischen Palaeographie". 1 
 Both the passages in the literary works and the characteristics 
 of the oldest alphabet point to the conclusion that the Hin- 
 dus extensively used the art of writing at least about three 
 centuries before the time of Asoka-Piyadasi. 
 
 II. 
 
 When thirty five years ago Professor Max Miiller wrote 
 his excursus on the Introduction of Writing in India, 2 the 
 oldest Brahminical works which he could quote as witnesses 
 for the use of letters were Panini's Grammar, Mann's and Ya- 
 jnavalkya's Institutes of the Sacred Law, the Mahabharata and 
 Kalidasa's Dramas, and he had to declare that in the Vedas [6] 
 and in the later literature of the Vedic schools no certain trace 
 of the use of writing could he found, while they contained very 
 strong evidence for the prevalence of oral teaching and for 
 
 Grundriss der Indo-Arischen Philologie und Alterthumskunde , Bd. I, 
 
 Heft 11, Strassburg i. E. 1896. 
 
 History of Ancient Sanskrit Literature, p. 497 ff.
 
 6 Indian Studies. No IH. 
 
 their having been preserved by a purely oral tradition. From 
 the heterodox literature he was only able to adduce a passage 
 of the Lalitavistara which describes the first visit of prince 
 Siddhartha, the future Buddha, to the writing school, and from non- 
 Indian sources the seemingly conflicting utterances of Nearchos 
 and of Megasthenes, one of the most careless reporters on Indian 
 subjects. Professor Max Miiller's final conclusion was, there- 
 fore, that the art of writing became known in India about 400 
 B. C. in the middle of his Sutra-period and that, then and 
 even later, it was not applied to literary purposes. 
 
 With the further exploration of Indian literature various 
 additional pieces of evidence have come to light, which some- 
 what modify the above inferences and tend to show that writ- 
 ing was extensively used for the most various purposes at an 
 earlier period. A closer scrutiny of the ancient Dharmasutras 
 has proved that there is at least one among them, the so-called 
 Vasistha Dharmasastra, which in general mentions written do- 
 cuments (lekhya) as a proof of ownership (XVI, 10) and en- 
 joins in particular (XVI, 14 15) that in disputes about houses 
 and fields the judicial decisions shall be given in accordance 
 with the documents if the evidence of the neighbours disagrees, 
 and that they shall be based on the statements of the old men 
 and of the guilds, in case conflicting documents are produced. 1 
 Regarding the age of Vasistha' s Institutes of the Sacred Law 
 nothing definite is known. It is only possible to say that this 
 work is a real Dharmasutra, that it was originally composed 
 for the use of the students of a northern school connected with 
 the Rgveda and was considered to be of general authority be- 
 fore the eighth century of our era. 2 With respect to its rela- 
 tive position among the works on the sacred law, it is possible 
 to assert that it is older than the famous Manusamhita, where 
 one of its rules is quoted and the name of its supposed author 
 [7] is mentioned, while in its turn it quotes the ancient Dharma- 
 sutra of the Manavas, on which the homonymous metrical law- 
 
 Sacred Books of the East, vol. XIV, p. XXVI and p. 80. The first-men- 
 tioned passage is a verse, quoted by the author either from the tradi- 
 tion of the learned or from an older work. 
 Rumania, Tantravarttika, p. 179, Benares edition.
 
 II. Literary Evidence for the Use of Writing. 7 
 
 book is founded. 1 It is also later than Gautama's Dharmasutra 
 and probably belongs to the period, when special law-schools 
 had come into existence and taught the sacred and civil law 
 iu rivalry with the teachers of the Vedic schools. These cir- 
 cumstances make it difficult to deduce from its mention of 
 written documents more than the obvious general conclusion 
 that the art of writting was commonly used in daily life and 
 its importance for legal purposes was recognised during the 
 period, when new Vedic schools were still founded, and that 
 it is erroneous to consider the admission of written documents 
 as legal evidence to be a distinctive mark of the metrical 
 manuals of the special law-schools. On general grounds it is 
 probable that the composition of the Vasistha Dharmasastra 
 falls some centuries before the beginning of our era. To such 
 a conclusion points inter alia the fact that it is older than our 
 Manusamhita. But for the present it would be hazardous to 
 say anything more definite regarding its age. 
 
 More instructive are the numerous passages in the canon- 
 ical works of the Southern Buddhists which testify to an ex- 
 tensive use of writing in very early times. All those sections 
 of the Tripitaka, which contain descriptions of, or allusions to, 
 the national life of ancient India, furnish some contribution to 
 the subject. In the Jatakas, where, of course, most may be 
 expected, most is also found. Private and official correspon- 
 dence by means of letters is referred to again and again as 
 something quite common. In the Katahaka Jataka we are told 
 how Katahaka, a slave of the Sheth, or great banker and mer- 
 chant of Benares, by means of a forged letter passed himself 
 off as the son of his master and obtained the daughter of the 
 Sheth of another town: 
 
 "He (Katahaka) who performed in the Sheth's house the 
 work of a store-keeper thought, 'These people sha'nt make rue 
 always do the work of a store-keeper and treat me as a slave, 
 striking, imprisoning and branding me, if they find fault with 
 me. In a neighbouring kingdom there is a Sheth, a friend of [8] 
 our Sheth. If I take to him a letter (lekha) written in our Sheth's 
 name, and if I go to him and say that I am the Sheth's son, I 
 
 1 Sacred Books of the East, vol. XXV, p. XXIX f.
 
 8 Indian Studies. No III. 
 
 may deceive him, obtain his daughter and live comfortably'. 
 He himself took a leaf (panna), wrote as follows, 1 have sent 
 my son N. N. to thee; mutual connexion by marriage is suit- 
 able for us; give, therefore, thy daughter to this boy and let 
 him live there; when I have time, I will also come', and he 
 sealed the letter with the Sheth's seal. Then he took money 
 for the journey, perfumes, clothes and so forth according to 
 his pleasure, travelled to the neighbouring kingdom and stood 
 before the Sheth there, respectfully saluting him. Then the 
 Sheth asked him, 'Friend, whence hast thou come?' 'From Be- 
 nares. 5 'Whose son art thou?' 'The Benares Sheth's.' 'For 
 what purpose hast thou come?' Thereupon Katahaka handed 
 over the letter, saying, 'You will know it, when you have read 
 this.' The Sheth read the letter, and exclaiming, 'Now I live 
 indeed!', he gave him joyfully his daughter and established 
 him there." 1 
 
 Again the Mahasutasoma Jfitaka mentions a correspond- 
 ence by means of letters (panna) between a teacher of Tak- 
 kasila and his former pupils, 2 and the Kama Jataka 3 narrates, 
 how a prince, who had renounced the throne and lived in a 
 village, was asked to write and actually wrote a letter (panna) 
 to his brother, the reigning king, requesting a remission of the 
 royal taxes for the people who had hospitably received him. 4 
 
 An official letter is mentioned and its preparation is de- 
 scribed in the Punnanadi Jataka, 5 which gives an account of 
 the manner, in which the future Buddha was re-installed in 
 his position as Purohita of the king of Benares, after having 
 been banished in consequence of the intrigues of his enemies. 
 "Afterwards the king remembered his (the Bodhisattva's) vir- 
 tues, and reflected thus, 'It is not proper to send somebody 
 in order to call my teacher; but I will compose a verse, write 
 a [9] letter (panna), order crow's flesh to be cooked, tie up the 
 
 1 FausbOlI, Jatakas, vol. I, p. 451, 1. 22 S. 
 * Op. cit., vol. V, p. 458. 
 8 Op. cit., vol. IV, p. 169. 
 
 4 Compare also the Harita Jataka, No. 331, op. cit., vol. Ill, p. 498, 1. 18, 
 where the ministers send a panria to the king regarding Harita's mis- 
 conduct (reference kindly supplied by Prof. S. v. Oldenburg). 
 
 5 Op. cit., vol. II, p. 173f.
 
 II. Literary Evidence for the Use of Writing. 9 
 
 letter (panna) and the flesh in white cloth, seal it with the 
 royal seal (rajamuddika) and send it to him. If he is clever, 
 he will come after reading the letter and recognising the crow's 
 flesh; if he is not clever, he wo'nt come 5 . He then wrote the 
 verse, which begins with the words punnam nadim, on a 
 sheet." Of course the future Buddha was clever, and came 
 back to Benares. 
 
 Further, some other passages mention official correspond- 
 ence between kings. In the Cullakalinga Jataka 1 we are told 
 how powerful Kalinga, the king of Dantapura, eager for war- 
 fare, tried to pick a quarrel with the princes of India, whom 
 he found disinclined to gratify his bellicose inclinations. In 
 order to effect his purpose he sent his four beautiful daughters 
 in a covered cart through the territories of his neighbours and 
 ordered their guards to proclaim that any prince, who might 
 take them into his harem, would have to fight their father. 
 Assaka (Asmaka), the king of Potali, dared to ai'rest their pro- 
 gress, and made all the four maidens his queens. Thereupon 
 the Kalinga marched out with his army. But Nandisena, the 
 minister of the Assaka king, the Jataka continues, "hearing of 
 his approach sent a royal decree (sasana) to the following ef- 
 fect, 'Let him stop within the boundaries of his territory, let 
 him not cross our frontier, (else) a fight will take place between 
 the two kings 5 . When he (the Kalinga) had heard this letter 
 read (lekham sutva), he stopped within his own territory." Here 
 the term sasana, literally 'an order", is of considerable interest, 
 as it is the representative of the Sanskrit sasana, which occurs 
 so frequently as the technical term for landgrants. 2 
 
 Another case occurs in the Asadisa Jataka, 3 according to 
 which seven kings besieged the town of Benares and sent to 
 Brahmadatta, its ruler, a letter (panna), asking him either to 
 give up his kingdom or to fight. The future Buddha, who 
 was then Prince Asadisa, king Brahmadatta's elder brother, 
 came to the asistance of the latter. He cut (achindi) on an 
 
 1 No. 301, op. cit., vol. UI, p. 4 ff. 
 
 2 Two other cases, in which letters, called panna and sasana, are men- 
 tioned occur in No. 186, op. cit., vol. II, p. 104, 1. 1. 8 and 9, and No. 462, 
 op. cit., vol. IV, p. 133, 1. 1. 4 and 9 (Oldenburg). 
 
 3 No. 181, op. cit., vol. II, p. 89 ff.
 
 10 Indian Studies. No III. 
 
 arrow the following letters (akkhardni): "I, Prince Asadisa, 
 have come, and shall destroy with one arrow the lives of all 
 [10] of you; let those fly who wish to live," and, being unrivalled 
 in the archer's craft, he shot his arrow on the knob of the 
 golden dinner-vessel of the besiegers. The latter, who where 
 just sitting at dinner, read the letters and, of course, speedily 
 raised the siege. 
 
 The Jatakas contain also a passage, mentioning the use 
 of writing for legal purposes. In the Ruru Jataka 1 a debtor 
 invites his creditors to come with the bonds (inapannani), 
 which he had given to them, to the banks of the Ganges in 
 order to receive payment. The same Jataka, (p. 257) mentions 
 further the custom of inscribing particular important records 
 or compositions on gold-plates. Khema, the queen of Benares, 
 had dreamt of a gold-coloured deer and had notified to her 
 husband that she must die, if the deer was not found. The 
 uxorious king composed this verse: 
 
 'To whom shall I give a rich village and women decked with ornaments? 
 Who will tell me of that deer, the best deer among deer?" 
 
 which he caused to be engraved on a gold-plate. The plate 
 he made over to his prime minister and caused the inscription 
 to be read to the townspeople. 2 While in this case the text en- 
 graved is a kind of proclamation, we learn from the Kanha Ja- 
 taka 3 that in rich families statements regarding the acquisition of 
 property were preserved in this peculiar manner. The future 
 Buddha, the story says, who had been born as the son af a Brah- 
 man possessing eighty millions, after the death of his parents 
 "one day examined his treasury. Seated on a splendid couch, 
 he caused a gold-plate to be brought and looked at the letters 
 (akkharani), incised on it by his ancestors, which stated, 'So 
 much wealth has been gained by such a one and so much by 
 such another one'." 
 
 1 No. 482, op. cit., vol. IV, p. 256. 
 
 2 According to No. 159, op. cit., vol. II, p. 36, the king of Benares caused 
 the information about a gold-coloured peacock to be engraved on a 
 gold-plate, which one of his successors caused to be read out (Olden- 
 burg). 
 
 3 No. 440, op. cit., vol. IV, p. 7.
 
 II. Literary Evidence for the Use of Writing. 11 
 
 Three other cases, mentioned respectively in the Kurudhamma 
 Jataka, 1 the Tesakuna Jataka and the Sambhava Jataka are 
 again different. 2 All three stories narrate, how particularly 
 valued moral maxims were engraved on gold-plates apparently 
 in order that they might [11] not be forgotten. In the first the 
 inscription records at the king's command the Kurudhamma, 
 the law of the Kurus, which is identical with the five great 
 precepts, imposed by all Indian religions on laymen, f Not to 
 slay, not to steal, not to commit adultery, not to lie, not to 
 drink intoxicating drinks. 3 According to the second story the 
 future Buddha caused the vinicchayadhamma "the maxims con- 
 cerning righteous judgment and the behaviour of kings", which 
 he had preached, to be perpetuated in the same way. 3 In 
 addition to these testimonies for the use of writing the Kafahaka 
 Jataka, already quoted, gives a hint regarding the existence 
 of writing-schools and the manner in which writing was taught 
 in ancient India. But, this had be better reserved for a fuller 
 discussion of that subject. 
 
 Equally valuable are a number of passages of the Vinaya- 
 pitaka, which Professor Oldenberg's Index makes easily acces- 
 sible. Writing (lekhd) and writers (lekhaka) are mentioned 
 in the Bhikkhu Pacittiya II. 2 and in the Bhikhun! Pacittiya 
 49. 2. In the former passage writing is enumerated among 
 "the excellent branches of learning, which are not blamed, nor 
 despised, nor contemned, nor disregarded, (but) esteemed in the 
 various countries." In the Parajika section a curious practice 
 is forbidden to the Buddhist monks, in which writing plays 
 an important part. "(If one) cuts (chindati), the text says, an 
 inscription (lekham) to this effect, c He who dies in this manner, 
 will obtain wealth, or will obtain fame, or will go to heaven/ 
 (the cutter) is guilty of a Dukkata (dushkrta) sin for each 
 single letter (akkharakkharaya). (If anybody] sees the inscrip- 
 
 1 No. 276, op. cit., vol. II, pp. 371, 381. 
 
 2 No. 522, op. cit., vol. V, p. 125, compare No. 515, op. cit., vol. V, p. 59 
 (Grierson). 
 
 8 According to Jataka No. 509, op. cit., vol. IV, p. 488 f., information about 
 hidden treasure was written on gold-plates. The same story narrates 
 that Visvakarma wrote with jatihiiigula on the walls of the pa^asalas 
 which he built by order of Indra (Oldenburg).
 
 12 Indian Studies. No III. 
 
 tion, and forms the painful resolution to die, (the engraver will 
 be guilty) of a Thullaccaya (sthulatyaya) sin; (in case the 
 reader actually) dies, (the engraver will be guilty) of a Para- 
 jika offence." l 
 
 The passage indicates that it was the practice of religious 
 teachers to incite their lay-hearers by the promise of rewards 
 in the next birth to commit suicide, and that they distri- 
 buted tablets of wood or bamboo with inscriptions specifying 
 the manner of the death and the rewards to be gained. The 
 statement is perfectly credible, as religious suicide, or suicide 
 with the hope of rewards in the next birth, was very com- 
 mon in ancient India and even occurred not rarely within 
 [12] the last thirty or forty years. 2 The Dharmasutras and the 
 metrical Siurtis mention the voluntary death by starvation or by 
 other more violent means and even recommend it to the her- 
 mits and to the other ascetics, 3 and there are passages in the 
 Mahabhfirata, where ascetics, kings and others are recommended 
 to put an end to their existence by starvation, jumping down 
 from precipices, voluntary cremation and so forth. The Jainas 
 were and are universally in favour of the c death of the sages', 
 though it has gone out of fashion in our days. And Hiwen 
 Tsiang, Siyuki I, p. 232 (Beal) testifies to the prevalence of 
 the belief that a jump from the Aksaya Vata, the sacred fig- 
 tree at Allahabad, secured re-birth among the gods, as well as 
 to the fact that it was acted on in the seventh century. Like 
 the majority of the Brahnainical teachers who, though giving the 
 old rules, strongly disapprove of suicide, the Buddhists naturally 
 opposed such practices. Their statement that written exhortations 
 to suicide used to be given, furnishes another valuable piece of 
 evidence for the very general use of writing in ancient India. 
 
 Finally there are still two remarkable passages in the 
 Mahavagga I, 43 and 49, which are also of considerable im- 
 
 1 See also Prof. Oldenberg's remarks in Sacred Books of the East, vol. XIII, 
 p. XXXII f. 
 
 2 In January 1869 there was still a guard on the Girnar, in order to pre- 
 vent pilgrims from jumping from the rock, called the Bhairav jhdmp 
 'Bhairava's leap'. 
 
 8 See Manu VI, 31 and the passages quoted in the Synopsis to my Trans- 
 lation, Sacred Books of the East, vol. XXV, pp. 204, 567.
 
 II. Literary Evidence for the Use of Writing. 13 
 
 portance for our question. 1 The first tells us that a likhitako 
 coro, a thief whose name had been placarded or proclaimed 
 in writing in the king's palace had been received into the 
 order of the Buddhist monks. The people murmured against 
 this and Buddha, of course, forbade for the future the ad- 
 mission of proclaimed thieves. The story confirms the hints, 
 to be gathered from the Jatakas, regarding written royal procla- 
 mations. The second passage, which is literally reproduced in 
 Bhikkhu Pacittiya 65. 1, and hence must be very ancient, de- 
 scribes in detail the deliberations of the parents of a boy of 
 Rajagrha, called Upali, about their son's education. They agree 
 that it would be beneficial for his future, if he learnt lekha 
 'writing', ganana 'arithmetic 5 and rupa, literally 'forms'. But 
 [13] they find that the first art might injure his hands, the second 
 his chest and the third his eyes. Hence they finally resolve to 
 permit him to enter the order of the Buddhist monks, who are of 
 good moral conduct, dine well and sleep in well sheltered beds. 
 Even at first sight it seems probable that rupa, ga- 
 nana and lekha were the three "Rs", or subjects taught in 
 the elementary schools of ancient India, when the Mahavagga 
 was composed. This conjecture is confirmed by a remark, 
 which king Kharavela makes about his own education in the Ha- 
 thigumpha inscription, dated in the year 165 of the Maurya era. 
 He says concerning himself: 
 
 [i] 
 
 wrftra [11] 
 
 "Endowed with the body of a glorious prince, he played 
 during fifteen years children's games. Then, being expert in 
 writing, rupa, arithmetic and legal rules and excelling in all 
 sciences, he ruled during nine years as Caesar." 
 
 Dr. Bhagvanlal renders the untranslated word rupa by 
 "painting", while Professors Oldenberg and Rhys Davids, Sacred 
 Books of the East, vol. XIII, p. 201, take it to mean in the Maha- 
 vagga "money- changing", because Buddhaghosa says in the com- 
 
 -1 See also the remarks on these two stories by Prof. Oldenberg in Sacred 
 Books of the East, vol. XIII, p. XXXII f. and by D'Alwis, Introduction 
 to Kachchhayana's Grammar, pp. XXVII, CXVf., 73101. 
 
 2 Actes du VI. Congres Int. Or. Ill, 2, p. 154.
 
 14 Indian Studies. No III. 
 
 mentary that "he who learns the rupa-sutta must turn over and 
 over many kar sap anas and look at them". The rendering 
 "money-changing", though not far from the truth, is a little 
 too specific. For it is not probable that a royal prince would 
 qualify himself to become a banker. But, the curriculum of 
 the so-called indigenous schools of the present day includes 
 a branch of elementary learning, which may be called rupa 
 "forms" and to which Buddhaghosa's explanation may also 
 refer. 1 After the children have mastered the art of writing 
 and the most elementary arithmetical operations, addition, sub- 
 traction and particularly the amk or complicated multiplication 
 tables, they are instructed in the practical application of arithmetic 
 [14] to simple commercial and agricultural affairs. They learn, 
 how many Dams, Korls, Paisas, Paulas and so forth go to the 
 Rupee, the rules for calculating interest and wages as well as 
 the simplest rules of mensuration. 2 This commercial and agri- 
 cultural arithmetic is no doubt what is meant by rupa and it 
 may be that in ancient times, when coins were rare, specimens 
 were placed before the pupils, which they had to handle and 
 look at, in order to learn their form, weight and marks. As 
 far as I am aware, this is not done in our days. 3 
 
 With respect to the instruction in writing, there is, as 
 stated above, something more in the beginning of the Katahaka 
 Jataka. "When the son of the Sheth, the story says, learnt 
 writing, the slave (Katahaka) too went with him carrying his 
 
 1 My statements regarding the indigenous schools are based on what I 
 have seen in Western, Northern and Central India. The Rev. J. Long's 
 edition of Adam's Reports on Vernacular Education, pp. 19ff. and 98 ff. 
 furnishes confirmatory statements. See also Captain Harkness' article 
 on the schools in Southern India, Jour. Roy. As. Soc., vol. I, p. 15 f., 
 Mr. Gover's on the Dyal schools of Madras, Indian Antiquary, vol. II, 
 p. 52 ff., and Mr. Raghunathjl's on the Marathl schools, Indian Antiquary, 
 vol. VIII, p. 246 ff. 
 
 2 What is taught in this way in the indigenous schools of Gujarat has 
 been collected by Rao Saheb Bhogilal Pranvallabh and published by 
 the Bombay Educational Department under the title Detl Hisab, "Native 
 Arithmetic", Pts. I and II. 
 
 3 According to Major Temple the Burmese market -girls were actually 
 taught in this way within the last thirty years, Indian Antiquary, 
 vol. XXIV, p. 247, Notes.
 
 II. Literary Evidence for the Use of Writing. 15 
 
 board and (thus) learnt writing." The sentence indicates, that 
 the Sheth's son did not receive instruction at home, but went 
 to a master, who presumably kept a school. The mention of 
 the "board" (phalaka) is very interesting. It agrees with the 
 narrative of the Northern Buddhist Lalitavistara, according to 
 which young Siddhartha, the future Buddha, on -going to the 
 school of the Brahman Visvamitra, brought with him "a golden 
 pencil and a tablet of red sandal wood". 1 And the actualities 
 in the indigenous schools of Western, Northern, Central and 
 Eastern India 2 furnish the necessary commentary on the two 
 passages. Nearly everywhere the board or wooden tablet is 
 still in use for the first instruction in writing, and it is either 
 covered with sand in which the letters are drawn with a small 
 stick, or it is varnished and the letters are drawn with a stick, 
 smeared with a solution of white chalk instead of ink. 3 It is 
 [15] evident that the Lalitavistara and the Jataka refer to the me- 
 thod of instruction, which is still followed, and the information, 
 conveyed by the latter and Mahavagga I. 49, makes it pro- 
 bable that elementary schools existed at the period, when the 
 Buddhist canon was composed, and that their curriculum was 
 about the same as that of the indigenous Pathsalas, Lehsa^s, 
 Nisals (i. e. *Lihsalas) and Tolls of modern India. 
 
 In the portions of the Nik.ayas, which I have read, I have 
 met with fewer references to writing. But they are not entirely 
 
 Professor Terrien de la Couperie, Babylonian and Or. Record, vol. I, 
 p. 59, states that these words are found in the older Chinese translation, 
 Pu yao king, dated 308 A. D. The legend possibly goes back to the 
 beginning of our era. 
 
 Regarding Bengal and Behar, see the passages from Adam's Report quoted 
 above. 
 
 In Gujarat the latter method is the more common one, and a vessel, 
 filled with a solution of chalk usually belongs to the paraphernalia of 
 the schoolboys. But I have likewise seen the sanded board, on which 
 the grains are made to adhere slightly by gum arabic. Very poor boys 
 simply scattered dust on their boards and wrote in that, or if they had 
 no boards, they brought brass kettles or pans and wrote on these with 
 chalk. This happened even in the Government schools, where more 
 usually slates and chalk-solution were used by the beginners or Amk- 
 valas, as the school-phrase is. Beruni, India, vol. I, p. 182 (Sachau), 
 writing iu A. D. 1030, mentions the use of black tablets in the schools, 
 on which the children wrote with a white material.
 
 16 Indian Studies. No III. 
 
 wanting. Thus the Brahmajala Sutta 14 and the Samanna- 
 phala 49, mention a game, called akkharika, 1 which according 
 to Buddhaghosa means "reading letters in the air or in the 
 vault of the sky", see also Childers, Pali Diet. s. v. pitthl. 
 
 As regards the question to which exact period the testi- 
 mony of the "Pali Canon may refer, the answer, I think, must 
 be, "to the fifth and possibly to the sixth century B. C." In 
 the introduction to the Vinayapitaka Professor Oldenberg has 
 shown that there are good reasons for assuming the composi- 
 tion of the Mahavagga, Pacittiya and Parajika sections to pre- 
 cede the Council of Vesall (ca. 380 B. C.) and to fall even 
 somewhat earlier than the year 400 B. C. 2 While it will be 
 sufficient to refer to his discussion on the Vinaya and to Pro- 
 fessor M. Miiller's review of the question in the introduction to 
 the Dhammapada, Sacred Books of the East, vol. X, p. XXIX if., 
 the case of the Jatakas requires a fuller consideration. 
 
 It is a well known fact that the sculptures on the Sanchi 
 and Bharahut or Bharaut Stupas contain representations of va- 
 rious Jatakas. On the Sanchi Stupa the Sanaa Jataka has been 
 identified 3 and others, like the Mahakapi Jataka, may be re- 
 cognised even on the Plates in Dr. Fergusson's Tree and Ser- 
 pent [16] Worship. On the Bharahut Stupa twenty-one sculptured 
 scenes, to which the titles are mostly added, have been found 
 to correspond with Birth Stories in Professor Fausboll's printed 
 edition, and 'Dr. E. Hultzsch has shown that even a Pada of 
 a verse, used as a title, agrees with the printed text. 4 Among 
 
 1 Mentioned also in the Cullavagga I, 13. 2 and elsewhere. 
 
 2 Vinayapitaka, vol. I, p. XXXIV XXXVIII. 
 
 3 Jour. Roy. As. Soc. 1894, p. 21 Iff. 
 
 4 Indian Antiquary, vol. XXI, p. 225 f., where Dr. Hultzsch has reprinted 
 his excellent edition of the inscriptions together with a synopsis of the 
 stories hitherto identified and other very valuable remarks. The fact 
 that the titles of the stories frequently differ, the Pali text naming the 
 story after one chief actor or incident and the inscription after another, 
 has been discussed by Professor Rhys Davids in the Introduction to his 
 Buddhist Birthstories p. LX ff. And he has shown that it in no way 
 goes against the assumption that the canonical collection existed at the 
 time when the sculptures were made, because vacillations with respect 
 to the titles occur likewise in the Pali collection. I would add that 
 variations in titles are also found in Brahminical literature. Thus Bana
 
 II. Literary Evidence for the Use of Writing. 17 
 
 them there are also two, the Rurumiga and the Asadisa, which 
 have been quoted above as witnesses for the use of writing. Both 
 the Stupas date from the third century B. C., some additions 
 only belonging to the second century. The age of the Sanchi 
 Stupa is proved by a fragment of an Asoka Edict, which agrees 
 in part with the so-called Kosambl Edict on the Allahabad Pil- 
 lar, and by the fact that the characters of more than nine tenth 
 of its nearly four hundred inscriptions fully agree with those 
 of the Asoka Edicts, while about two score show slightly more 
 advanced forms. 1 Similarly, nearly all the hundred and fifty 
 three published inscriptions of the Bharahut Stupa are written 
 in the alphabet of the Edicts, and it is chiefly Dhanabhuti's 
 inscription on the gateway-pillar dated "in the reign of the 
 Sungas", which is incised in more modern letters. The pieces 
 with the later characters are, of course, additions or repairs, made 
 [17] after the completion of the original structure. 2 Under these 
 circumstances it is very probable that in the third century 
 B. C. our collection of the Jatakas formed part of the Buddhist 
 Canon, which, as the Bairat Edict, addressed to the Magadha 
 Saiigha, and various inscriptions on the Stupas indicate, was 
 then fully settled. Both on the Sanchi and Bharahut Stupas 
 we read of monks who had the title pacanekdyika (pdncanai- 
 
 calls the Bhagavadglta, Anantaglta. Kumarila uses the title Ascarya Par- 
 van for the Putradarsana Parvan of the MSS. of the MahSbharata, and 
 there are Vedic hymns with two or three names. Such a vacillation 
 comes quite natural to a Hindit, who is accustomed to substitute endless 
 synonyms for technical terms and names of plants animals etc., and 
 even changes portions of personal names, saying Vikramaditya, Vikra- 
 marka or Vikramanka, though he means the same individual. The 
 discrepancies between the Pali stories and the representations in Bharahut, 
 which Prof. Minayeff believed to have discovered (Recherches I, 140ff.), 
 are, I think, merely such as may be expected to be found between all 
 narratives and their illustrations by sculptures. 
 
 1 See my articles on the Sanchi Stupa inscriptions in the Epigraphia In- 
 dica, vol. II, p. 87 ff. and p. 366 ff., which latter is accompanied by two 
 plates, giving specimens of inscriptions in the old and the more ad- 
 vanced characters. 
 
 2 See the Plate in vol. XL of the Zeitschrift der Deutschen Morgenlan- 
 dischen Gesellschaft, and Indian Antiquary, vol. XIV, p. 139. The in- 
 scription of Dhanabhuti on the gateway-pillar belongs in my opinion to 
 the middle of the second century B. C. 
 
 Buhler, Indian Studies. III. 2
 
 18 Indian Studies. No III. 
 
 kayika) 1 i. e. 'teacher of the five Nikayas 5 , and in Bharahut 
 appears also a petaki i. e. c a person who knew or taught the 
 Pi^aka or Pitakas', about whose designation more will be said 
 below. 
 
 The evidence of the relievos on the Stupas furnishes a 
 lower date for the Jatakas, later than which their use for 
 the edification of the Buddhist laymen cannot be put. But 
 there are other circumstances connected with them, which 
 make it very probable, that the picture of the national life of 
 [18] India, furnished by them, refers to a much earlier period than 
 the third century B. C. Though, as Professor Rhys Davids 
 has shown op. cit., p. LV. ff., nothing definite is known re- 
 garding the date when the present collection was formed and 
 incorporated in the Buddhist canon, it is already now evident 
 that the stories which they contain are not, as was believed 
 formerly, inventions of the Buddhist monks, but almost through- 
 out loans from the ancient Brahminical literature or the old pre- 
 Buddhistic national tradition of India. A very good statement 
 of this view by Dr. S. von Oldenburg has been translated in 
 
 By an oversight I have given in the Epigraphia Indica, vol. II, p. 93, 
 pancanaikayika as the Sanskrit equivalent of pacanekayika. Though such 
 a form might be defended by analogous compounds like tridraunika, it 
 is equally possible and no doubt better to assume that the prototype of 
 the Prakrit word was formed in accordance with the rule, exemplified 
 by cMurvaidyalca 'one who knows or teaches the four Vedas'. For, in 
 the ancient epigraphic Prakrits panca and panca would both be repre- 
 sented by pamca or paca i. e. pacca. These remarks may possibly meet 
 the doubts, which M. A. Barth has expressed regarding the word in the 
 Bulletin des Religions de Tlnde, 1894, Bouddhisme, p. 1, note 1. M. A. 
 Barth's further doubts, whether the five Nikayas, known in the third 
 century B. C., may be identified with those of the Pali Canon, do not 
 appear justified to me. Assuming the correctness of his statement that the 
 'five Nikayas' are unknown to the Northern Buddhists, this fact would, 
 in my opinion, not prove anything against the antiquity of these collec- 
 tions, because the Northern Buddhists have preserved only the disjecta 
 membra of an old tradition. No doubt, when the Northerners and the 
 Southerners agree with respect to a particular text or collection, its great 
 age is clearly established. But it seems to me dangerous to invert the 
 proposition and to allege that books or collections of the Southern Ca- 
 non, not known to the Northerners, must be considered later additions 
 for this reason alone.
 
 II. Literary Evidence for the Use of Writing. 19 
 
 the Jour. Roy. As. Soc. of 1893, p. 301 ff., and I can only 
 say that a closer study of the Jatakas had led me to the same 
 conclusion before the publication of Dr. von Oldenburg's pa- 
 per. Moreover, the detailed researches of Professor Leumann 
 and of Dr. R. G. Bhandarkar have yielded exactly the same 
 results, see the Wiener Zeitschrift fur d. Kunde d. Morg. Vol. 
 V, p. 11 Iff., the Zeitschrift der Deutschen Morg. Gesellschaft 
 Vol. XLVIII, p. 65 ff. and the Transactions of the IX th Int. 
 Or. Congress, 1892, Vol. I, p. 422 ff. 
 
 This discovery makes the question regarding the exact 
 date, at which the loan may have been effected, a matter of 
 minor importance. The chief point for consideration is, if in 
 effecting the loan the Buddhist monks altered much and espe- 
 cially, if the description of Indian life which the Jatakas con- 
 tain, has been made to agree with that of the times when 
 Buddhism had become a power in India. The answer can only 
 be, that there are remarkably few traces of Buddhism in these 
 stories and that they do not describe the condition of India in 
 the third or fourth century B. C., but an older one. Peculiarly 
 Buddhistic are only the introduction of the future Buddha into 
 most tales, who invariably is identified with the wisest and best 
 of the actors, occasional spiteful remarks against the Vedic 
 animal sacrifices and against the deified national heroes, whom 
 the orthodox sects worshipped, and intentional perversions of 
 the legends told of them. The doctrine of the power of the 
 Karrnan, the moral maxims and the few religious observances 
 such as the sanctification of the Uposatha or Parvan days, 
 which the stories inculcate, are common to the Buddhists and 
 to all other Indian religions, whether orthodox or heterodox. 
 [19] With respect to these there was not much to change, except 
 perhaps some of the technical expressions. 
 
 On the other hand the descriptions of the political, reli- 
 gious and social condition of the people clearly refer to the 
 ancient time before the rise of the great Eastern dynasties of 
 the Nandas and the Mauryas, when Pataliputra had become 
 the capital of India. The Jatakas mention neither the one nor 
 the other, and they know nothing of great empires which com- 
 prised the whole or large parts of India. The number of the 
 kingdoms, whose rulers play a part in the stories, is very con- 
 
 2*
 
 20 Indian Studies. No III. 
 
 siderable. The majority of the names, such as Madra, the two Paii- 
 calas, Kosala, Videha, Kasi, and Vidarbha agrees with those 
 mentioned in the Vedic literature, while a few others, like Ka- 
 liftga and Assaka i. e. Asmaka or Asvaka, occur in Brahmi- 
 nical literature first in the Epics and in Panini's Sutras. The 
 characteristic names of the Andhras, the Panqlyas and Keralas 
 are not mentioned. 
 
 Though a political centre was wanting, frequent state- 
 ments regarding the instruction of the young Brahmans and 
 nobles show that there was an intellectual centre and that it lay 
 in Takkasila, the capital of distant Gandhara. Takkasila is 
 according to the Jatakas the town, where the youth of the 
 tAvo highest classes received instruction in the three Vedas and 
 the twelve or eighteen branches of learning (sippa or vijjathana). 
 This agrees with some statements in other parts of the Canon, 
 where, as in the Vinayapitaka, the famous physician Jlvaka Koma- 
 rabhacca is represented as having studied medicine in Takka- 
 sila. And it is very credible that Gandhara, the native country 
 of Panini, was a stronghold of Brahminical learning certainly 
 in the fourth and fifth centuries B. C., and perhaps even earlier. 
 The statements regarding the religious condition of India 
 point to an equally early period. Just as the three Vedas are the 
 basis of the higher instruction, so the prevalent religion is that 
 of the Path of Works with its ceremonies and sacrifices, among 
 which several like the Vacapeyya (vajapeya) and the Rajasuya 
 are specially and repeatedly mentioned. Side by side with 
 these appear popular festivals, celebrated, when the Naksatra 
 had been proclaimed, with general merry-makings and copious 
 libations of Sura, as well as the worship of demons and trees, 
 [20] all of which go back to the earliest times. Nor are the her- 
 mits in the woods and the wandering ascetics unknown. Most 
 of the heroes take the isipabbajja, i. e. } renounce the world 
 according to the rule of the Rsis, and live with their sacred 
 fires in the hill-forests, whence they descend occasionally in 
 order to procure salt and pungent condiments (lonambilaseva- 
 nattha). The wandering ascetics (tapasa) appear to belong to 
 different orders, as various distinctive marks are mentioned. 
 But only those of one division, the Ajivikas, are actually named, 
 among whom the future Buddha himself was once born in a for-
 
 II. Literary Evidence for the Use of Writing. 21 
 
 raer Kalpa. These are, as Professor Kern has first pointed out, 
 the ancient Vaisnava devotees of Narayana and particularly the 
 Paramahamsas, who according to the lately recovered Vaikha- 
 nasa Dharmasutra actually went naked and swallowed cowdung, 
 as the Buddhists allege of these dangerous rivals. 
 
 The state of civilisation, described in the Jatakas. is in 
 various respects primitive, and particularly noteworthy is the 
 prevalence of wood architecture, which on the evidence of the 
 oldest sculptures had almost disappeared in the third century 
 B. C. The Jatakas even describe the palaces of kings as us- 
 ually constructed of wood. Many other details might be added. 
 But the facts given are sufficient for our purpose. They make 
 it at least probable that the mention of writing as common in 
 daily life is not an addition, made by the Buddhists in later times, 
 but occurred in the old stories which they appropriated. And it 
 ought to be remembered, that in the Punnanadi and Asadisa 
 Jatakas writing is not merely an ornamental accessory, but a most 
 essential point, without which the stories would have no meaning, 
 and that the Asadisa Jataka is found on the Stupa of Bharahut. 
 
 An additional argument for the antiquity of the writing, 
 mentioned in the books of the Pali Canon, is furnished by the 
 technical terms which they employ. They exclusively use for writ- 
 ing, writers and letters words which mean "to cut", chind, "to 
 scratch" likh, "the scratcher" lekhaka, "scratching" or "scratches", 
 lekha, and "the indelible" akkhara. On the other hand the word 
 lipi, which Panini, circiter 350 B. C., 1 has in the compounds 
 [21] Upikara and libikara, "writer", is not found in the Buddhist ca- 
 nonical works known to me, nor is there in Childers' Dictionary 
 any quotation for it from the Canon. For lipi two explanations 
 have been proposed. According to the older opinion it is de- 
 
 1 I can only adhere to the traditional date of the great grammarian, which, 
 as we know now, was contained in the ancient Brhatkatha, since both Kse- 
 mendra's and Somadeva's Sanskrit translations contain the story of Pa- 
 nini, the pupil of Upavarsa, who flourished during the reign of Yoga- 
 nanda, the predecessor of Candragupta, Indian Antiquary, vol. I, p. 305. 
 Panini' s Dhatupatha gives for the verb likh the meaning aksaravimjdsa 
 'arranging letters' or 'syllables' and thus furnishes, as Geheimrath von 
 Bfihtlingk points out to me, another proof for the prevalence of writing 
 in the fourth cent. B. C.
 
 22 Indian Studies. No III. 
 
 rived from lip 'to smear', like krsi 'agriculture' from kj's to 
 plough', and like a number of other substantives formed by the 
 feminine affix i. According to others, 1 it is a corruption of the 
 Old Persian dipi 'writing', which corruption was favoured by 
 a fancied connexion with the verb lip. Though lipi might be 
 a perfectly regular derivative from lip, and might appropriately 
 have been used to denote 'letters' and 'the alphabet' on the 
 introduction of writing with ink, the derivation becomes doubt- 
 ful through the fact that the verb limpati is not used in the 
 sense of 'he writes'. And the impressions of the Shahbaz- 
 garhi version of Asoka's Edicts have furnished a strong argu- 
 ment for the adherents of the second view. For they show 
 that in the Gandhara dialect lipi is represented by dipi and 
 that the verbs dipati 'he writes' and dipapati 'he causes to 
 write' did also exist. 2 On linguistic grounds it is not probable 
 that lipi and limpati should have been turned into dipi and 
 dipati } the latter also changing its meaning. On the other 
 hand, dipi corresponds exactly with the Persian word, and its 
 introduction into India is easily explained by the Persian oc- 
 cupation of Northwestern India during the Akhaemenian period 
 from about 500 B. C. probably until the fall of the Persian 
 empire. 3 These points appear to me so strong that I too must 
 declare myself in favour of the loan theory, and assume that in 
 Sanskrit lipi was substituted for dipi at a period, when writing 
 with ink had come into use, in order to connect the term with the 
 [22] root lip. The statement of Nearchos, according to which the 
 Hindus wrote letters ev c'.vBis'. X(av x/.po-r / ;j.svatc, as well as that 
 of Q. Curtius VIII, 9, according to which they used libri ar- 
 borum teneri i. e. birch-bark, for the same purpose, makes the 
 use of ink certain for the latter half of the fourth century. 4 
 
 1 See, Burnell, Elements of South Ind. Palaeography, p. 5f., Note 2. 
 
 2 The aorist dipista is found in ShahbSzgarhi Ed. IV, 1. 10; VI, 1. 16 and 
 the participle dipapita ibidem, Ed. XIV, 1. 13. 
 
 3 See also App. I, p. 95. It may be noted, that even in later times the 
 Hindus have borrowed a Persian word, connected with writing. This is 
 the term divlra divira, "a writer, clerk", found in the Valabhl land grants 
 of the seventh century A. D. and in later Kasmirian works. It represents the 
 Persian deblr, see the smaller Petersburg Dictionary sub voce divira. 
 
 4 The oldest preserved specimen of writing in ink occurs on the relic 
 vase from Andher, Cunningham, Bhilsa Topes, PI. 30, No. 6.
 
 II Literary Evidence for the Use of Writing. 23 
 
 The loan theory agrees also well with the occurrence of the 
 word in the Sutras of Panini, who was a native of Salatura, the 
 modern village of Lahul, close to Ohind and west of the Indus, and 
 with its absence from the ancient Buddhist canonical works, 
 which were composed in Eastern India and before 400 B. C. 
 Some further valuable information regarding writing and 
 especially regarding the ancient alphabets, is furnished by the 
 works of the Northern Buddhists, by the Jaina scriptures and 
 by some metrical Smrtis. In the account of prince Siddhartha's 
 first visit to the writing-school, extracted by Professor Terrien 
 de la Couperie from the Chinese translation of the Lalitavistara 
 of 308 A. D. (see above p. 15, note 2), there occurs besides 
 the mention of the sixty four alphabets, known also from the 
 printed Sanskrit text, the utterance of the master Visvamitra, 
 "that there are two kinds of writing, that of Fan or Brahman 
 and that of Kuliu, both equally good and not differing". With 
 the help of a Chinese Buddhist Encyclopedia, dated 668 A. D., 
 Professor Terrien de la Couperie has shown that the alphabet 
 of Brahman was written from the left to the right and that, 
 invented by Kuliu, Kiuliii, Kialu or Kialusheta (all of which 
 names are explained by "ass' lips", in Sanskrit kharostha) from 
 the right to the left. He thereby has rendered us the great 
 service of showing what the terms Brahm! and Kharosthi really 
 mean. The former is evidently the alphabet, which used to 
 be called Pali, Lath, Southern Indian, Asoka or Maurya, and 
 the latter the so-called Northwestern, Ariano-Pali, Bactro-Pali, 
 or Gandharian. With this explanation, the remark of the future 
 Buddha's writing-master indicates, that at the time, when the 
 Lalitavistara, translated into Chinese in 308 A. D., was com- 
 posed, i. <?., at the latest in the third century A. D., both the 
 ordinary Indian and the Gandhara alphabets were equally com- 
 mon in the author's native country. If, as is not improbable 
 on account of the position of the Lalitavistara in the Canon 
 of the Northern Buddhists, this was the Panjab or one of the 
 adjoining districts of Northwestern India, the statement agrees 
 with the facts known through the inscriptions and coins, which 
 [23] from the earliest times of the historical period of India until 
 the end of the second century A. D., show both the Brahma 
 and Kharosthi characters.
 
 24 Indian Studies. No III. 
 
 The enumeration of the sixty-four alphabets 1 has gained 
 a greater interest by the discovery of a similar, apparently in- 
 dependent list in the Jaina Agamas. Professor Weber's ana- 
 lysis of the latter shows, 2 that they too allege the early exist- 
 ence of a larger number of scripts and that some of the names 
 agree literally, or at least in meaning, with those given by 
 the Buddhists. Both the Jainas and the Buddhists agree in 
 allotting the first place, to "Brahman's writing", the Brahml 
 llpi or Bambhl livl, indicating thereby the pre-eminence of the 
 characters, running from the left to the right, which are used 
 in the majority of Asoka's Edicts. Both name also the alpha- 
 bet written from the right to the left, which in the Jaina Pra- 
 krit is called Kharotthi instead of Kharosthl. Its position in 
 the two lists somewhat differs, as the Buddhists give it the 
 second place and the Jainas only the fourth. Further, both 
 lists include the Puskarasarl (No. 3) or Pukkharasariya (No. 5) 
 and the Dravida (No. 12) or Damila (No. 17) characters. The 
 mention of the last alphabet has become important since the 
 discovery of an ancient partly independent variant of the Brah- 
 ml lipi at Bhattiprolu in the Kistna districts of the Madras 
 Presidency. And it is not improbable that this alphabet is 
 meant by "the writing of the Dravidian country 3 ". The men- 
 tion of the Puskarasarl or Pukkharasariya lipi, too, possesses 
 some interest, as its name is evidently derived from the patro- 
 nymic Pauskarasddi or Puskarasadi, which appears in Brah- 
 minical works on grammar and law as the designation of one, 
 or perhaps of several famous teachers. It is certainly a Brah- 
 minical name and indicates that a scion of the race of Puskara- 
 sad invented some particular alphabet or introduced modifica- 
 tions [24] in an existing one. The name, therefore, furnishes the 
 
 1 See Lalitavistara, p. 143 f. Calc. ed. 
 
 2 Indische Studien, vol. XVI, pp. 280, 399 ff. The list occurs twice, in the 
 Samavayanga and the PannavanS Sutras. There is a third list of alpha- 
 bets with about 30 mostly very corrupt names in the Mabavastu, I, 
 p. 135 (Senart), which has been omitted intentionally, as its exact contents 
 cannot be made out with certainty. 
 
 8 I may state here that in my opinion Dr. Burnell's contention for the 
 antiquity ot the Vatteluttu or Pandya-Cera alphabet is erroneous; see 
 the Grundriss der Indo-Ar. Phil, und Alt., I, 11, 32.
 
 II. Literary Evidence for the Use of Writing. 25 
 
 proof that in early times Brahmans directed their attention to 
 the art of writing, for which view other arguments will be ad- 
 duced further on. The other names, which like the Gandharva 
 or Gandhavva agree fully, or like the nagalipi "the writing of 
 the snake-deities 33 and the Bhogavalya, "the writing of Bhoga- 
 vatl, the residence of the Nagas", in their general meaning, 
 are perhaps in part fanciful and, at least for the present, not 
 particularly instructive. 
 
 The independence of the Jaina list from that of the La- 
 litavistara is proved by various discrepancies. First, it is much 
 shorter and includes only eighteen varieties, and among them 
 occurs the Javanaliyd or according to the printed edition of 
 the Pannavana Sutra, the Javananiya, which corresponds to 
 Panini's term Yavananl, 1 or, as Katyayana's Varttika states, 
 "the writing of the Yavanas 33 . Secondly, the Jainas use the 
 term Bambhi livl in a double sense, not only for a particular 
 variety of writing, but also for every kind of writing. The 
 texts are explicit on this point and say, 2 Bambhle nam lime 
 attharasavihalikkhavihane pannatte \ tarn bambhi etc. "An eigh- 
 teenfold order of scripts is taught for the Brahma writing, viz. 
 the Brahma" etc. This apparently senseless assertion finds its 
 explanation through passages of the metrical Smrtis of Narada 
 and Brhaspati, as well as through Brahminical sculptures and 
 pictures of Brahman. The two lawbooks state that "the Cre- 
 ator (Brahman) created writing in order to keep the affairs of 
 the world in their proper course" or "in order to remove doubts 
 regarding legal transactions 33 . 3 On the sculptures in the Ba- 
 
 The identification, which is given by Malayagiri, is unobjectionable, 
 even if Javanaliyd is the correct form. For instances of the substitution 
 of la for Sanskrit na occur e. g. in Pali anela for an&nas, mulala for mr- 
 ndla, vef.u for venu and in Maharastrl velu for venu and limba for nimba. 
 It may be noted that the later Tibetan version of the Lalitavistara in- 
 cludes the Yavananl in its list of seventy alphabets, see Foucaux, Rgya 
 Cher Rol Pa, p. 112 f., and that the Yondrl of the Mahavastu, loc. cit., 
 is probably a representative of Yavananl or YavandU. 
 See A. Weber, op. cit, p. 399. 
 
 Sacred Books of the East, vol. XXXIII, pp. 58 f., p. 304. NSrada's law- 
 book is later than the Manusamhita and older than Bana (circiter 620 
 A. D.), who alludes to it in the KadambarT, p. 91 (Peterson). The Brhas- 
 pati Smrti is again later than Nfirada's.
 
 26 Indian Studies. No HI. 
 
 [25] dami caves, 1 which date from the end of the sixth century 
 A. D., the same idea is expressed by representing Brahman 
 with strips of palmleaves in his right hand, for which modern 
 pictures of the deity substitute an inscribed piece of paper. 2 It 
 thus appears that there were two legends which the Jainas have 
 combined, one which ascribed to Brahman the invention of all 
 writing and another which restricted his activity to one parti- 
 cular alphabet, that which was the commonest and most gen- 
 erally used. Both myths no doubt were current side by side, 
 and express with a slight variation the belief- that writing is a 
 national Indian invention. 
 
 These points, to which the differences in half a dozen 
 names of alphabets must be added, show very clearly that the 
 Jaina list has not been copied from that of the Buddhists, but 
 gives expression to an independent tradition, which in all pro- 
 bability is considerably older than that of the Buddhists. The 
 greater antiquity of the Jaina list is apparent from its more 
 reasonable number of alphabets, which, however, is also a purely 
 conventional one 3 and need not be taken to mean more than 
 "a large number". It is also probable, because the Anga and 
 the Upanga, in which it occurs, certainly are much older than 
 the third century A. D., the time for which the existence of 
 the Buddhist list is absolutely certain. In my opinion we have, 
 since the discoveries in the Kankall Tlla at Mathura, very good 
 reasons for believing the 6vetambara tradition which places the 
 first collection of the Angas in the reign of the Maurya Can- 
 dragupta or about 300 B. C. 4 And, though the Angas evidently 
 have undergone changes between that time and their final re- 
 daction by Devarddhi in the fifth century A. D., it seems to 
 me probable that the list of the alphabets belongs to the orig- 
 inal contents of the Samavayanga, because it has been embod- 
 ied also in the Pannavana Sutra, the traditional date of which 
 is 358 after Vlra or 168 B. C. Nor is the existence of such 
 
 1 Indian Antiquary, vol. VI, plate facing p. 361. 
 
 2 Moore, Hindu Pantheon, plate I, and the representation of Brahman in 
 Sir W. Jones' article, Asiatic Researches, I, p. 222 S. 
 
 3 Compare the eighteen Puranas, and Upapuranas, the eighteen Smrtis and 
 Upasmrtis. 
 
 4 See Sacred Books of the East, vol. XXII, p. XL ff.
 
 II. Literary Evidence for the Use of Writing. 27 
 
 [26] a traditional list at the beginning of the Maurya period a priori 
 incredible. Panini's rule regarding the formation of Yavananl 
 "the writing of the Yavanas" very probably indicates, as has 
 already been remarked by others, that in his time, i. e., about 
 350 B. C., more alphabets than one were known, 1 and for the 
 third century B. C. the contemporaneous use of three alphabets, 
 the ordinary Brahml lipi, the Kharogthl and the Bhattiprolu 
 variety of the Brahma alphabet is certain. The bearing of the 
 early existence of such a list of alphabets and of the myth, 
 ascribing their invention to Brahman, on the question of the 
 antiquity of writing in India, is obvious. The introduction of 
 writing cannot have taken place about 400 B. C., but must be 
 earlier at least by some centuries. 
 
 Another passage of the Jaina Samavayanga Sutra makes 
 it possible to show how the popular Brahma alphabet looked 
 about 300 B. C. The Samavayanga includes a detailed ab- 
 stract of the lost Drstivada, the twelfth of the Jaina Angas, and 
 asserts that according to this work, the Bambhl livl or Brahma 
 
 1 As stated above, I adhere to the tradition, which asserts that Panini was 
 the pupil of Upavarsa and lived during the reign of the last Nanda, the 
 predecessor of the Maurya Candragupta. This tradition gives a reason- 
 able date and probably goes back itself to the beginning of our era, since, 
 according to the concurrent testimony of Ksemendra and Somadeva, it oc- 
 curred in Guriadhya's Brhatkatha. I agree with Dr. Burnell (El. S. Ind. 
 Pal. 2 , p. 6) and Professor A. Ludwig, (see his paper on "Yavananl" quoted 
 above) that yavanani means "the writing of the Greeks". But I believe 
 with Prof. Ludwig that Pnnini's acquaintance with the existence of the 
 Greek alphabet is by no means irreconcilable with his traditional date. 
 Irrespective of the general reasons, adduced by Professor Ludwig, it seems 
 to me not wonderful that an author, whose native country had been ex- 
 plored in 509 B. C. by Skylax and whose countrymen, the Gandharas, 
 had furnished a contingent for Xerxes' invasion of Greece should mention 
 the old Oriental name of the Greeks and should be acquainted with their 
 writing (see also Professor Weber, Monatsberichte Berl. Akad. 1871, p. 616). 
 And there is, as Mr. Rapson points out to me, even positive proof for 
 such an acquaintance, as Athenian drakhmes with the inscription AOE 
 struck before the end of the reign of Alexander, possibly even before 
 350 B. C., have been found in India as well as Indian imitations of such 
 coins, see B. V. Head, Cat. Greek. Coins: Attica, p. XXXIf., pp. 2527. To 
 me it seems absolutely impossible to make the mere occurrence of the 
 word Yavana in Sanskrit works a mark, which proves that they must 
 have been written after the invasion of Alexander.
 
 28 Indian Studies. No III. 
 
 [27] alphabet consisted of 46 mauyakkhara (matrkaksara) or ra 
 clical signs. 1 The commentator Abhayadeva says that this number 
 comes out by deducting from the (in his time, saec. XI) ordi- 
 nary alphabet the vowels r, f, I, | and the lingual la, but 
 including km. The reckoning is correct, as may be seen from 
 a comparison of the oldest written alphabet, that on the Ho- 
 riuzi palmleaf, which gives 51 signs, viz; : 
 
 a, a, i, l, u, u, r, f, I, I (10), e, ai, o, au, am, ah, ka, kha, 
 ga, gha (20), ha, ca, cha, ja, jha, na, ta, tha, da, dha (30), na, 
 ta, tha, da, dha, na, pa, pha, la, bha (40), ma, ya, ra, la, va, 
 a, sa, sa, ha, lam (50), km. 2 
 
 If the four vowels and la are deducted, only 46 radical 
 signs remain. Nevertheless Abhayadeva's explanation undoubt- 
 edly contains a mistake. It is not the consonant la but the 
 group ksa, which ought to be deducted. For la is one of the 
 ancient radical signs and occurs on the Sanchi Stupa as well 
 as in the Bhattiprolu alphabet. Ksa, on the other hand, can 
 have been reckoned as a radical sign only from the time, when 
 ka was written with a loop or triangle on the left <Ji. It was 
 only then that the origin of ksa ^, in wich ka retained its 
 ancient dagger-shape, was obscured and that the still prevalent 
 erroneous conception. of the indigenous schoolmasters could arise, 
 who persistently declare ksa to be a mdtrka. The period, when 
 the ka with a loop came into general use probably falls not 
 earlier than between 400 and 500 A. D. In the inscriptions 
 of the nail-headed and flat-topped (Nagari) types, it is only 
 traceable since the second half of the sixth century. But in 
 the ordinary literary characters it appeared earlier, as the Ho- 
 riuzi palrnleaf shows. 
 
 With respect to the omission of the vowels r, r, I, I, 
 Abhayadeva is undoubtedly right, as they are missing in all 
 the ancient and modern alphabets, used in the elementary Ind- 
 
 1 See Weber, Indische Studien, vol. XVI, p. 281 f. and Verzeichniss der 
 Sanskrit und Prakrit Handschriften, vol. II, pt. II, p. 408, where better 
 readings of the text are given. 
 
 2 See Anecdota Oxoniensia, Aryan Series, vol. I, pt. 3, plates 1 and 2. The 
 alphabet of the Cambridge MS. Add. No. 1049 has one letter less, as it 
 omits la which is unknown in Nepal and other parts of Northern India 
 as in Kashmir.
 
 II. Literary Evidence for the Use of Writing. 
 
 29 
 
 [28] ian schools and by the classes without a scientific Brahminical 
 education. According to the printed Lalitavistara, p. 145 f., 
 prince Siddhartha explained on entering the writing-school to 
 his master the hidden meaning of the radical signs of the al- 
 phabet which he was to learn. 1 The vowels, which he is said 
 to have enumerated, are only twelve, viz: a, a, i, l, u : u, e, 
 ai, o, an, am, ah. These twelve vowels alone occur in the 
 alphabet, taught in the indigenous Indian schools, where they 
 are combined with the consonants and form the so-called Ba- 
 rakhadl or Barasakhadi, 2 which the children on beginning their 
 school-course are made to copy incessantly, until by its means 
 they have learned both elementary reading and writing. The 
 Barakhadl, in Sanskrit dvddasaksarl, "a collection or aggreg- 
 ate of twelve syllables (for each consonant)", is arranged as 
 follows 3 :- 
 
 A A II U U E AI AU Am Ah 
 
 1 
 
 ka 
 
 kha ' 
 
 ga 
 
 gha 
 
 ha 
 
 
 2 
 
 ka 
 
 kha 
 
 ga 
 
 gha 
 
 ha 
 
 
 3 
 
 ki 
 
 khi 
 
 0* 
 
 ghi 
 
 hi 
 
 
 4 
 
 kl 
 
 khi 
 
 9~ l 
 
 ghl 
 
 m 
 
 
 5 
 6 
 
 7 
 8 
 
 ku 
 ku 
 ke 
 kai 
 
 khu 
 khu 
 khe 
 khai 
 
 gu 
 ga 
 
 9 e 
 gai 
 
 ghu 
 ghu 
 ghe 
 ghai 
 
 hu 
 nu 
 he 
 hai 
 
 and so on 
 through all the 
 consonants. 
 
 9 
 
 ko 
 
 kho 
 
 9 
 
 gho 
 
 ho 
 
 
 10 
 
 kau 
 
 khau 
 
 gau 
 
 ghau 
 
 hau 
 
 
 11 
 
 kam 
 
 kham 
 
 gam, 
 
 gham 
 
 ham 
 
 
 12 
 
 kah 
 
 khah 
 
 gah 
 
 ghah 
 
 hah 
 
 
 1 This alphabet, too, included 46 mStrkas. But the Calcutta edition omits, 
 no doubt erroneously, among the consonants the dental la. The last letter 
 is k?a. 
 
 2 The word is usually pronounced B5ra-khadl instead of Barakhadl, be- 
 cause its etymology is no longer remembered. 
 
 3 The Barakhadl has been printed in Bombay and can be obtained through 
 Pandit Jeshtaram Mukundji. It is described by Molesworth, MarathT 
 Dictionary sub voce ^"fjr^T^i'^', and by Narmadashankar in his GujaratI 
 Dictionary sub voce ^TTT^^^"- This latter work states that the syllables 
 ka, ka and so forth down to kah are the Barakhadl of ka.
 
 30 Indian Studies. No III. 
 
 [29] This Barakhadi, which is used in all the parts of India 
 known to me, 1 has of course always the same number of vow- 
 els, but the number of the consonants varies. In the Maratha 
 country and in Gujarat, there are 36 instead of 34, the groups 
 ksa and jna being reckoned as simple radical signs and placed 
 after la. As usually a Mangala or invocation om namah siddham 
 is prefixed to it, it is sometimes called by the Pandits the Sid- 
 dhamatrka or Siddhaksarasamamnayali or "the alphabet, pre- 
 ceded by the word Siddha (success)", or jocularly Matrkapu- 
 rana the Purana of the Mothers (radical signs)". 
 
 Its great antiquity is attested by the Chinese writers. 
 Hiwen Tsiang states in the Siyuki 2 that the instruction of the 
 young Hindus began with the practice in the twelve chang or 
 tables. What these twelve tables were, appears from an account 
 of the Indian alphabet and elementary instruction, given by 
 Hui-lin 3 who wrote between A. D. 788 810. According to 
 him the thirty four consonants, the last of which was ksa, 
 were each first combined with the twelve vowels a ah, and 
 thus a table or syllabary was formed identical with the mo- 
 dern Barakha<Jl, which bore (according to another authority) the 
 name Siddhavastu "the matter (preceded by the word) siddha". 
 Next came ligatures of each consonant with ya, ra, la, va, sa, 
 ha and the five nasals, and each of these ligatures was again 
 combined with the twelve vowels. The result was a set of 
 eleven further tables, which were, as the author says, in com- 
 mon use. Though Hui-lin calls these tables fan or "cycles", 
 
 1 It is also used for the pagination of MSS. in Burma, Siam and Ceylon. 
 In the former two countries the vowels are 12, in Ceylon 16, see the 
 Grundriss, I, 11, p. 82 f. 
 
 a See, St. Julien, Memoires etc., I, 72 and note, where, however, the trans- 
 lation has "un livre en douze sections" and the note alleges that the book 
 was a syllabary, composed by Brahman and called Siddhavastu. In a 
 letter to me, quoted in the first edition of this essay, Prof. J. Legge has 
 pointed out that the Chinese text has nothing about "a book", but 
 merely "12 chang" or tables. 
 
 3 See Dr. A. von Rosthorn's letter to me in the WZKM (Vienna Oriental 
 Journal) X, 280, the contents of which are summarised in the above ac- 
 count. My former conjecture that the expression "12 chang" is a trans- 
 lation of dvadataktari (Barakhadi) and stands for "12 (syllable) table" 
 is of course not tenable.
 
 II. Literary Evidence for the Use of Writting. 31 
 
 it is not doubtful, as Dr. A. von Rosthorn remarks, that they are 
 identical with Hiwen Tsiang's twelve chang. It thus appears that the 
 system of instruction in writing, described by Hui-lin and Hiwen 
 Tsiang, is based on the Barakhaoll still used, but is further devel- 
 oped in accordance with the_requirements of Sanskrit teaching. But 
 the four vowels, r, r, I, I are not included, and Hui-lin, who 
 mentions them expressly, states that ordinary people and beginners 
 did not know them, as well as that they were employed as 
 Expletives' by grammarians. Hiwen Tsiang, too, excludes 
 them from the ordinary alphabet which, as he says in the 
 Siyuki, I, 77 (Beal), contained 47 letters i. e. the 12 vowels 
 and 35 consonants, ka ha, la and ksa. 
 
 [30] The omission of the four liquid vowels may also be in- 
 ferred from the fragments of an alphabet of the Maurya period, 
 originally discovered by Sir A. Cunningham at Mahabodhi 
 Gaya 1 and lately re-examined by Dr. G. A. Grierson. 2 In the 
 cloistered walk, which according to the tradition Asoka erected 
 over the supposed Cahkama of Buddha, there is a double row 
 of pillar bases, 3 eleven on each side, on which according to 
 Dr. Grierson the following letters are still visible: 
 
 South 
 
 11 10 987654321 
 
 cha na ka 
 
 22 21 20 19 18 17 16 15 14 13 12 
 North 
 
 The A on No. 1 is according to Dr. Grierson laid on its side, 
 and the fia on No. 16 is in the same position. If the sign |Hl, which 
 is here given according to a photographic quarter size repro- 
 duction of Dr. Grierson's photo-etching (Rep., Figure 3), is set 
 up straight , we obtain a form closely agreeing with those 
 
 1 Mahabodhi Gaya, p. 8f. Plates III and X, 1. 
 
 2 Report to the Seer. Gov* of Bengal, General Dept, Nov. 25, 1895. 
 
 3 The pillar base No. 1 with the letter A has been transferred to the 
 Asoka Gallery of the Indian Museum at Calcutta, together with two 
 others which at present bear no letters, Grierson, Report, p. 2.
 
 32 Indian Studies. No III. 
 
 in the Mathura inscriptions of the Kushana period. 1 There is 
 besides one pillar shaft, at present erroneously fixed in the 
 base with ka, which, as Dr. Grierson's facsimile (loc. cit.) 
 shows, bears a mutilated A, laid on its side. Sir A. Cun- 
 ningham does not mention the ha on No. 16, but instead gives 
 a round-topped ga on No. 13, as well as ja on No. 19 and ta 
 on No. 22 as visible at the time of his visit (1881). Dr. Grier- 
 son however states that these signs are not to be traced at 
 present. He also declares it to be improbable that the ja was 
 actually seen by Sir A. Cunningham "as the fractures [on the 
 spot where the letter ought to stand] are evidently very old", 
 and he conjectures (Rep., p. 5) that Sir A. Cunningham's round- 
 topped ga is "merely a faulty impression" of the sign on 
 No. 16. 2 ' 
 
 However that may be, and even if the additional ja and 
 ta mentioned by Sir A. Cunningham are left out of count, it 
 is evident that the masons had numbered the pillar bases, 
 and probably also the corresponding shafts, with the letters of 
 the Brahma alphabet arranged according to the phonetic prin- 
 ciples of the grammarians and of the Barakhatfi. As the first 
 base of the southern row is marked with the letter A and 
 No. 12, the first of the northern row, with ka, it is further 
 evident that the masons' alphabet included eleven vowel-signs, 
 one less than the Barakhadl. And it becomes also probable that 
 among them were, as Sir A. Cunningham conjectured, the signs 
 for A, I, I, U, U, E, AI, 0, AU and that the last was either 
 Am or Ah, 3 one of them being inadvertently omitted. The li- 
 quid vowels cannot have had a place here, since they would 
 make the total of the vowel signs sixteen and it would be ne- 
 cessary to assume that the masons skipped five of them. 
 
 1 See the Grundriss, I, 11, Tafel III, 10, IV and especially 40, III. In the 
 second Mathura letter we have likewise the little knob on the left, cau- 
 sed by the scribe's making the bottom-stroke separately. 
 
 2 On comparing Sir A. Cunningham's facsimiles, op. cit., Plate X, 1 with Dr. 
 Grierson's, it seems to me evident that the former reproduce drawings 
 or at least corrected impressions of rubbings. 
 
 3 As Dr. Grierson has pointed "out, there is no sufficient reason for 
 declaring with Sir A. Cunningham that the last letter must have 
 been Ah.
 
 II. Literary Evidence for the Use of Writing. 33 
 
 As the Indian masons are neither great scholars nor 
 quite illiterate, 1 it may be assumed that the alphabet, which 
 they used, is the popular one of the elementary schools of 
 the day. 
 
 [31] The discovery of the Gaya alphabet possesses, as Sir 
 
 A. Cunningham has pointed out, considerable importance for 
 the history of Indian writing. As its fragments are arranged 
 on phonetic principles and as they include the guttural ha, 
 which never occurs in the old Prakrit inscriptions, it must have 
 been framed for the requirements of Sanskrit. And it is at 
 least highly probable that its formation is due to the Brahmans, 
 whose influence and peculiar theories are also recognisable, as 
 will appear further on, in the manner of the derivation of the 
 secondary signs from the original ones. In the third century 
 
 B. C., it appears, the state of things in the elementary schools 
 was the same as in our days. The children learnt an alphabet 
 which was not intended for their vernacular dialects, and this 
 was no doubt due to the circumstance that already then (as 
 later when the legend regarding Buddha's Brahminical writing- 
 master Visvamitra arose, and also in modern times) the ele- 
 mentary instruction was chiefly in the hands of the Brahmans, 
 who did not think it worth the while to alter for the sake of 
 their Prakrit speaking pupils the alphabet, invented and suited 
 for the peculiar bhdsa of their schools and class. 
 
 Finally there are explicit statements in Chinese Buddhist 
 works, discovered and kindly communicated to me by M. Sylvain 
 Levi, according to which the signs for the four liquid vowels 
 were invented either by Nagarjuna or by Sarvavarman, the 
 minister of Satavahana. This tradition points at all events to 
 the conclusion that even in later times 1 the Hindus were aware 
 of the fact that the ancient Brahma alphabet contained only 
 twelve vowel signs. 
 
 1 The Indian masons, the sutradkaras or vardhakins of the older literature 
 and the siletas of our days, occupy an intermediate position between the 
 Aryan and the Sudra classes. They wore and still wear the sacred thread, 
 and they possessed and still possess a small amount of Sanskrit learning. 
 The rules of their craft, which they commit to memory, are written in 
 Sanskrit, and are at present largely mixed with Prakrit and barbarous 
 bastard forms. 
 
 Buhler, Indian Studies. III. 3
 
 34 Indian Studies. No III. 
 
 If we return to the passage of the Samavayanga Sutra, 
 the various facts adduced make it plain that Abhayadeva's ex- 
 planation of the extract from the Drstivada is substantially 
 correct. The forty-six radical signs of the ancient Brahmi lipi 
 included twelve vowels a, a, i } I, u, u, e, ai, o, au } am, ah and 
 thirty four consonants viz, the twenty five of the five Vargas, 
 the four liquids, the three sibilants, the spirant ha and in all 
 probability not ksa, but la. It appears further, that there is 
 no reason to distrust the Jaina tradition, according to which 
 the statement of the Drstivada goes back at least to the reign 
 [32] of the Maurya Candragupta, as his grandson's masons had 
 learnt an alphabet, apparently agreeing with that described in 
 the Jaina Anga in the most important particular. 
 
 The result of this enquiry, which shows that the popular 
 Indian alphabet of the third century B. C. had no signs for the 
 vowels r, r, I and 7, will not surprise those who have paid atten- 
 tion to Indian palaeography. The long l } the existence of which 
 (as a sound) is denied by the grammarians of Panini's school, is 
 an invention probably due to the Brahminical Kabbala, the so- 
 called Mantrasastra, which seems to have been studied and used 
 for charms by Brahmans, Buddhists and Jainas at least since 
 the beginning of our era. The sign for the intial L occurs first 
 on the Horiuzi palmleaf and in the ancient Cambridge MS. 
 from Nepal, Add. No. 1049, where it consists of two cursive 
 la interlaced (see Tafel VI of the Grundriss). Medial I is, as 
 far as I know, not traceable in any old document. Among the 
 remaining three sounds, only the short r occurs as an initial 
 in words of the ordinary language, while initial R and L are 
 used only in the technical terms of the Vyakarana and other 
 Sastras. The ancient signs for the initial Z-, which again are 
 found on the Horiuzi palmleaf and in the Cambridge MS. Add. 
 No. 1049, are cursive forms of la. Medial I is expressed in 
 the inscriptions mostly by li. But in the Haidarabad grant of 
 Pulikesi II (Ind. Ant. vol. VI, 72, pi. II a , 1. 7), we find a cursive 
 la in klpta, see also the Grundriss, I, 11, Tafel VII, 42. XIV and 
 p. 65. A similar sign occurs also in the same word in Rudra- 
 deva's Annamkonda inscription, 1. 118, see Ind. Ant. vol. XI, 
 facs. at p. 14 15. The principles, on which the more common 
 signs for initial and medial r, r have been framed, are (1)
 
 III. The Antiquity of the Brahmi. 35 
 
 initial r and r are expressed by a ra with the signs for medial 
 r and r, (2) medial r is a modification of ra, produced either 
 by a twist of the ra-stroke to the left with or without a curve 
 to the right at the end, or by the addition of a curl at the end 
 of the ra-stroke, (3) medial f is invariably expressed by the 
 double medial r of the period. Thus we find for vr in the 
 Northern inscriptions of the first and second centuries A. D. 
 ^ or ^ and in the fourth century and later J, while the 
 inscriptions and even the modern alphabets of the Southern 
 type offer ^. The northern initial R, which appears first in 
 the Bower MSS., 1 is ^ i. e. ra with the curve of the medial 
 r attached, and [33] the southern forms of the letter, the oldest 
 example of which occurs in a Kadamba grant of the fourth or 
 fifth century, 2 appear to be modifications of the northern sign. 
 These facts indicate that the signs for medial r and f were 
 developed first and that those for the initials came into use 
 somewhat later. As will be shown below, the process is exactly 
 the reverse of that followed in the cases of the other vowels, 
 where the medial signs are identical with, or modifications of, 
 the initial ones. It is evident that the formation first of initial 
 and next of medial vowels is the natural method, when an 
 alphabet without vowel-signs is turned into one with vowels. 
 Hence the palaeographic facts, too, show that the signs for r 
 and f were not framed at the same time with those for i, I, 
 u, u, e, ai, o, au, and that in all probability they are later 
 inventions. 
 
 HI. 
 
 If we now turn to the consideration of the oldest Indian 
 inscriptions, it is not difficult to show that the palaeographic 
 facts fully confirm the results, which the preceding examination 
 of the ancient literature has yielded. They likewise show that 
 writing, and especially the Brahmi lipi, had had a long history 
 in India, before king Piyadasi-Asoka caused his Edicts to be 
 incised in the various provinces of his large empire. In ad- 
 
 1 Compare also for the several forms of r and R the Grundriss, I, 11, Tafel III, 
 7, IX; 34, III; 37, VI; Tafel IV, 7, III; 39, III; Tafel VI, 18, IV; 43, III and 
 Tafel VI, 7, I-II. 
 
 2 Ind. Ant., vol. VI, p. 23, pi. Ill, 1. 12. 
 
 3*
 
 36 Indian Studies. No III. 
 
 dition they permit us to recognise that the Brahmi lipi is the 
 real old Indian alphabet, which was popularly used in the third 
 century B. C. all over India, and that it was fully developed 
 [34] before the introduction of the Kharosthi, whence we may 
 infer that it is the writing which the Buddhist canonical works 
 mention so frequently. 
 
 That the Brahmi lipi was in the third century an ancient 
 alphabet with a long history is proved by the very consider- 
 able variations in the forms of its signs found in the several 
 versions of the Edicts, as well as by the recent discovery, on 
 Sir A. Cunningham's Eran coin, of a variety which runs from 
 the right to the left, and of another one, on the Bhattiprolu 
 relic caskets, which includes a number of more archaic forms 
 and independent developments. 
 
 As regards Asoka's Edicts, even a cursory inspection of 
 good impressions or of trustworthy facsimiles, such as those 
 of Drs. Burgess, Fleet and Hultzsch, reveals the existence of 
 numerous differences in the formation of the several characters. 
 A more careful study shows not only that almost every letter 1 
 has two, three, half a dozen or more sometimes widely diver- 
 gent forms, but also that certain peculiarities are confined to 
 particular districts, as well as that the great majority of the 
 apparently or really more advanced forms, which appear more 
 or less constantly in the inscriptions of the next three or four 
 centuries, are found already in the Edicts. And it is significant 
 that the local differences observable permit us to speak of a 
 Northern and a Southern variety of the ordinary Brahma al- 
 phabet, between which, as in later times, roughly reckoning, 
 the river Narmada marks, the boundary line. 
 
 In order to show, how considerable the divergences are, 
 I give here the eight chief varieties of the initial A? among which 
 the first and the last show hardly any resemblance, though 
 if all eight are placed side by side their connexion is easily 
 recognised. 
 
 X X X a H * H 
 
 1 According to Tafel II of my Grundriss der indischen Palaeographie the 
 only exceptions are the letters U, jha, na, ta, }ha, na, tha and na. 
 
 2 The same varieties are of course found also in the initial a.
 
 III. The Antiquity of the Brahml. 37 
 
 It is not difficult to see, that the first seven varieties 
 have been caused by two conflicting tendencies, a liking for 
 [35] angles and a liking for curves, which are observable also in 
 the divergent forms of other Brahma letters. Nos 1 3 are 
 purely angular. The first and the second differ only thereby, 
 that in the one the obtuse angle touches the vertical line, and 
 that in the other it has been first made separate and then con- 
 nected by a short crossbar. 1 In the third form the right hand 
 stroke is slightly bent in the middle and the whole letter re- 
 sembles a Greek XL Nos 6 and 7 show on the left, instead 
 of the straight sides of the angle, two well developed curves, 
 and in No 6 they are united in the middle, while in No 7 
 they do not touch. Nos 4 5 appear to be mixed forms, as 
 the former has on the left a curve below with a straight stroke 
 slanting towards the left, while in the latter the curve stands 
 above and the straight stroke below. This apparent mixture 
 may be due to an incomplete change of the older, angular 
 form. But the two varieties may .also be explained as cursive 
 developments from No 6, the writer not caring to make the 
 more elaborate curve twice. No 8, finally, is a purely cursive 
 development from No 6, a straight stroke being substituted for 
 the notched line on the left. 
 
 Now the local distribution of these forms, which with the 
 exception of Nos 2 and 3 are of very frequent occurrence, is 
 as follows. The angular forms Nos 13, as well as the var- 
 iety mentioned in note 1, are confined to the Southern ver- 
 sions of the Edicts. They appear only in Girnar and Sidda- 
 pur, Dhauli and Jaugada, and it may be noted that in Girnar 
 and Siddapur they are in the majority, Nos 4 7 appearing 
 only occasionally, while in the two Southeastern versions (with 
 the exception of the Jaugada Separate Edicts) the contrary is 
 the case. On the other hand I know of no case where a purely 
 angular form is found in the KalsT, Delhi, Bairat, Sanchi, Al- 
 lahabad, Mathia, Radhia, Rampurva Barabar, Sahasram and Ru- 
 pnath Edicts. In the majority of these documents Nos 4 7 
 alone are used. No 8 occurs very frequently in Kalsi (some- 
 
 1 There are also instances in which this crossbar is omitted and the angle 
 stands by itself, see e. g. amina, Siddapur I, 1. 3.
 
 38 Indian Studies. No III. 
 
 times with small variations) and once or twice in Rampurva. 
 Hence it would appear that the angular forms are southern 
 [36] peculiarities and that, as they are undoubtedly the more an- 
 cient ones, they furnish an instance of the conservatism, fre- 
 quently observable in the southern alphabets of later times. 
 This inference is confirmed by the fact that other ancient 
 inscriptions of the same or nearly the same period, like those 
 on the Kolhapur and Bhatfiprolu relic caskets and from the 
 Nanaghat cave, likewise show the angular forms, (mostly side 
 by side with the curved ones), while the documents, found 
 north of the Narmada, such as those on the Bharahut and 
 Sanchi Stupas and in the NagarjunI caves, as well as the coins 
 of Agathocles offer almost exclusively the A with two curves, 
 rarely the mixed form No 4. A solitary exception in Maha- 
 bodhi Gaya (Cunningham, op. cit., plate V, No 2) may be ex- 
 plained by the consideration that, as the dialect of several 
 Gaya inscriptions shows, Southerners came to visit that famous 
 place of pilgrimage, and that the person who wrote the copy 
 may have been a Southron. 
 
 In addition to the local differences in the form of the 
 initial A and A, there are also others observable in kha, ja, 
 ma, ra and sa, which may be briefly noticed here, though the 
 full details must be reserved for the discussion of the palaeo- 
 graphy of the Edicts in my Grundriss. 1 The kha with a circle 2 
 or loop at the foot, a very ancient form, alone is used in 
 Kalsi and besides occurs only in Jaugada together with the 
 simplified forms showing a dot instead of the circle or no ap- 
 pendage at all. The ja with a loop in the middle 3 is used 
 only in Kalsl. It seems to be a form peculiar to the extreme 
 North and the Northwest, as it is found also on the coins of 
 Agathocles and of the Taxila merchants (see below p. 48 f.). The 
 other northern versions have mostly a secondary development 
 from it, the ja with a dot in the middle (plate I, No 7, Col. V, c) 
 more rarely the independent form (given under No 7, Col. V, d), 
 which is used exclusively in Girnar together with its derivative, 
 
 1 See now the Grundriss der Indo-Ar. Phil, und Alt. I, 11, p. 34 ff. 
 
 2 See plate I of this paper, No. 19, Col. V, a. 
 
 3 See plate I, No. 7, Col. V, b.
 
 III. The Antiquity of the Brahmi. 39 
 
 the later angular ja with three horizontal bars (see below p. 41). 
 The ma with the angle at the top, an ancient form, (plate I, No 13, 
 Col. V) occurs in the southern versions (with the exception of 
 Siddapur), where an open square appears instead of the angle) 
 [37] and is used exclusively in Girnar. All the northern versions 
 of the Edicts have a semicircle instead of the angle. The ang- 
 ular and wavy forms of ra and their insertion in the verticals 
 of consonants are also peculiar to the southern versions (Gir- 
 nar and Siddapur); the solitary ra in the Rupnath Edict 
 consists of an almost straight stroke. Finally, the ancient sa 
 with a straight limb on the left (plate I, No 15, Col. VI, b) is 
 likewise confined to the South (Girnar and Siddapur). 
 
 These facts, to which some more might be added, are suf- 
 ficient to show, that the very common idea 1 of the homogene- 
 ousness of the characters of the Edicts and of the absence of 
 local varieties, is erroneous. The differences between the writing 
 of the northern and the southern versions are quite as consider- 
 able as those, found four hundred forty years later, between the 
 letters of the northern and southern inscriptions of the first and 
 second centuries A. D. 2 And it must be kept in mind that the 
 circumstances, under which the Edicts were engraved, were not 
 favourable to a full expression of the local varieties of the letters. 
 Copies were sent out from Pataliputra into the provinces, which 
 were recopied and, as the dialectic differences in the language 
 and occasional peculiarities in the wording prove, also recast 
 by the clerks of the district-governors, before they were made 
 over for engraving to the masons. It seems only natural to 
 
 See e. g. Burnell, El. South Indian Palaeography, p. 7, note 4, with whose 
 remarks Dr. Taylor, M. Halevy and others agree. 
 
 I state this in accordance with the facts, shown by Tafel III of my Grund- 
 riss, which includes inter alia the signs from the inscriptions of the Sakas 
 and Kusanas of Mathura, Kaman and Sanchi, the Western Ksatrapas, 
 the Andhras and Abhlras. Dr. BurneU's statement (loc. cit.) that "in 
 the course of a few hundred years [after Asoka] the alphabets used in 
 Gujarat and Bengal had already become so different as to be very little 
 alike in appearance", I fear, cannot be substantiated. Between 200 B. C. 
 and 200 A. D. there are no inscriptions from Bengal proper. The in- 
 scriptions from Mahabodhi Gaya in Behar (given by Sir A. Cunningham) 
 which may be assigned to this period, look very much like those of the 
 Sakas and the Western Ksatrapas.
 
 40 Indian Studies. No III. 
 
 assume that the characters of the copies prepared at Patali- 
 putra influenced the writing of the provincial clerks, and caused 
 the introduction of forms, otherwise not usual in the several 
 [38] provinces. 1 Such a suspicion is the more natural, as the pro- 
 vincial clerks have in no case completely changed the lang- 
 uage, but have always allowed some Magadhisms to stand. 
 But, however that may be, local differences are traceable in 
 the writing of the several versions and they prove that the 
 Brahma alphabet had had a long history before the third cent- 
 ury B. C. If the slowness of the change of the forms, and 
 the comparative insignificance of the local variations which the 
 inscriptions of the next four or five centuries show, are taken 
 as the standard, it will be necessary to assume that the letters 
 of the Edicts had been used at least during four or five hund- 
 red years. 
 
 As regards the second important point, to which attention 
 has been called above, viz., the occurrence of numerous ap- 
 parently or really advanced forms, identical or closely agreeing 
 
 As I have stated already in Dr. Burgess 1 Arch. Surv. Rep. W. I., Vol. IV, p. 79 f. 
 and in the Ind. Ant., Vol. XII, 190, the clerks who prepared the fair copies 
 of the inscriptions, not the masons or coppersmiths who engraved them, are 
 the men who influenced the formation of the letters. How the masons worked, 
 may be seen from two passages of the Kalsi version. In the twelfth Edict, 
 1. 31 six letters have been scored out and the corrections have been written 
 above the line. The letters in the upper row are as large as those in the 
 lower and the distance between lines 30 and 31 becomes from the beginning 
 of the corrected passage twice as great as it was before. It is evident 
 that the mistake and its correction occurred iu the MS. given to the 
 mason. If they were due to the latter, the line would run on straighter 
 and the letters of the correction would be smaller. Again in the four- 
 teenth Edict 1. 20 the syllables tile of the word ghatile are corrections, 
 one standing above the line and the other below it. But the distance 
 between rjha and the first letter of the following word, ma, is twice as 
 great as those intervening between the other letters of the line. Here 
 it is again certain that the MS. had the mistake and the correction. 
 If the mason had skipped the two signs and added them afterwards, the 
 gha would not stand further off from ma thait from its predecessor. I 
 think that these two instances are sufficient to prove that Asoka's ma- 
 sons copied quite mechanically. It seems, therefore, impossible to attri- 
 bute to them any other influence on the shape of the letters than such 
 as may be caused by a slip of the chisel or by their accidentally over- 
 looking a stroke in the MS. before them.
 
 III. The Antiquity of the Brahml. 41 
 
 with those of the later times, its significance will become best 
 apparent, if all such signs, found in the Edicts, are placed to- 
 gether. The subjoined little table, for the preparation of which 
 [39] I have to acknowledge the help of Dr. W. Cartellieri, gives in 
 the lines, marked A, the advanced Asoka letters photolitho- 
 graphed according to cuttings from the facsimiles, and in those, 
 marked B, the corresponding characters from the later in- 
 scriptions of Hathigumpha, the Nanaghat, Mathura and the 
 Western caves. The dates of the latter vary between the 
 middle of the second century B. C. and of the second century 
 A. D., and in every case the oldest available counterpart has 
 been chosen. 
 
 n ka kha ga gha cha ja $a ti 
 
 H+3 ? ^ ^ L dt < X 
 
 12345678 9 10 11 
 
 va vi sa ha 
 
 \> d * lr^ 
 ^ > L 13 * J ,, Lr < 
 
 12 13 14 15 16 17 18 19 20 21 22 
 
 The table shows that the later signs for sixteen letters 
 occur already in the third century B. C. Four of these forms, 
 the dagger- shaped ka (common in Kalsi and occurring in most 
 other versions except in Girnar), the angular gha (Kalsi) and 
 ha (No 21, in various versions) as well as the curved da 
 (Kalsi) are in reality archaic, 1 while the corresponding common 
 signs of the Edicts are advanced developments, which like the 
 peculiar va (No 18, Girnar, Sohgaura), and sa (No 20, Kalsi, Jau- 
 gacla), have left no trace in the later writing. The remaining ones 
 are really cursive or derived from cursive forms. No 1, the 
 initial A, has been discussed above, its counterpart in line B 
 
 1 See below pp. 58 f., 61, 73. The counterparts in line B are from the 
 Hathigumpha inscription.
 
 42 Indian Studies. No III. 
 
 occurs in the Kusana inscriptions from Mathura, No 3, the 
 kha with a loop to the right (Kalsl) is a cursive form for the 
 kha with the circle at the foot, 1 and itself the parent of No 4 
 [40] (Delhi-Sivalik) which preserves the twist in the down stroke 
 caused by the loop, but substitutes a dot for the latter. 2 No 5, 
 the kha with the triangle (once in Mathia) is a fanciful va- 
 riant for the oldest form, its counterpart has been taken from 
 the archaic Mathura inscriptions. No 6, the ga with the round 
 top 3 is a cursive form of the pointed letter, the corresponding 
 form is from Hathigumpha, but found in all inscriptions of the 
 second century B. C. No 8, the cha with two loops 4 is a 
 tertiary development, immediately derived from the form with 
 the bisected circle, which again is merely cursive. 5 The same 
 remark applies to No 9, the ja with three bars. It is derived 
 from the notched Girnar form, which itself is a cursive deve- 
 lopment from the Blmttiprolu form." Nos 11 and 19, the 
 stunted ti and vi 1 , give examples of the triangulation of the 
 lower portion of va and of the reduction of the vertical 
 strokes so characteristic of the alphabets of the next cent- 
 uries, which appears already with great regularity in the 
 Nagarjunl cave inscriptions of Asoka's grandson. No 12, the 
 
 1 See below plate I, No 19, Col. V, I. 
 
 2 The corresponding forms in line B have been taken from Hathigumpha. 
 Better ones for No 3 are found in the inscriptions from the Western 
 caves, see Tafel III, 8, VIII of the Grundriss. 
 
 3 Once in Delhi Sivalik Ed. VII and in the mason's alphabet at Mahabo- 
 dhi Gaya. 
 
 4 Kills! once and mason's alphabet from Mahabodhi Gaya. The form in 
 line B is from Hathigumpha. 
 
 5 See below p. 68. 
 
 6 See below plate I, No 7, Col. V, a and the discussion on No 7, p. 60 f., 
 The sign, given above, occurs in Girnar, Ed. IX, 1. 1. The great length of 
 the central bar is caused by the a-stroke which it includes. Similar 
 forms occur in Kalsi, where they represent the looped ja. The corres- 
 ponding form in line B occurs already in Dasaratha's Nagarjunl inscrip- 
 tion. 
 
 7 The stunted ta is very frequent in Kalsl and occurs occasionally also in 
 other versions. The triangular va with the very short vertical is found 
 r>nce in a correction in Kalsl, Edict XIII, 2, 1. 13. The forms in line B 
 have been taken from the Kusana inscriptions. Some instances of pa, 
 pha and *a with very short verticals occur likewise already in the Edicts.
 
 III. The Antiquity of the Brahmi. 43 
 
 da with a shallow curve and the tail twisted towards the right 
 (Jaugada, Girnar, etc.) is a transitional form, corresponding to 
 those in the NagarjunI cave, the archaic Mathura and the Pa- 
 bhosa inscriptions (B), and leading up to the da of the first 
 [41] and second centuries A. D. Nos 13, 14 and 16 give examples 
 of the introduction of angles in the lower limb of pa, pha and 
 la* which originally consisted of curves, but are changed with 
 perfect regularity already in the Hathigumpha, the archaic 
 Mathura and the Pabhosa inscriptions. 2 No 15, the l>ha with 
 the round side-limb (Jaugada, Kalsl, etc.) is of course cursive 
 and found in all the later inscriptions except in Hathigumpha. 
 Nos 17 and 22. the la and ha with the side-limbs turned down- 
 wards (Jaugada Separate Edicts) are again highly cursive. Coun- 
 terparts of No 17 are found in the Nanaghat inscription, those of 
 No 22 only in the Abhira inscription from the Nasik caves. These 
 facts, to which others, such as later forms of the medial vowels, 
 the position of the Anusvara at the top of the consonants and 
 the occasional use of serifs or short bars at the top of verti- 
 cals, might be added, do not agree with the assumption that 
 writing was a recent practice in Asoka's times. To me it seems 
 that they are most easily explained, on the supposition that 
 several, both archaic and more advanced, alphabets existed in 
 the third century B. C., that an archaic alphabet was chosen 
 for the perpetuation of the Edicts, but that the clerks mixed 
 the forms. And in support of this view I would adduce the 
 Jaina tradition, discussed above p. 23 ff., according to which 
 many alphabets were used about 300 B. C. But, even if we 
 leave aside all conjectural explanations of the facts, it remains 
 undeniable that the writing of the Edicts is in a state of 
 transition, and this alone is sufficient to warrant the assertion, 
 that their alphabet certainly had had a long history. 3 
 
 1 Nos 13 and 14 occur a few times in Kalsi and other versions, No 16 is 
 from Delhi-Sivalik and occurs also in Kalsl, etc. 
 
 2 Among the inscriptions of the second century B. C., that from the Nana- 
 ghat cave preserves the round forms. 
 
 8 As so distinguished an epigraphist as Dr. Burnell has come to exactly 
 the contrary conclusion and as his view is still quoted by other writers 
 on the subject, it will not be superfluous, if I briefly review his argu- 
 ments. His chief argument for the late introduction of writing is that
 
 44 Indian Studies. No III. 
 
 [42] To an earlier stage than the writing of the Edicts belongs 
 the inscription on Sir A. Cunningham's Eran coin, which runs 
 
 very few allusions to the use of letters are contained in the literary 
 works which date from the fourth century B. C. The answer to this 
 statement is contained in the second part of this paper, and it need 
 only be pointed out that Dr. Burnell could not know of the passages 
 mentioning private and official documents, as the works, in which they 
 occur, had not been published in 1878. His second and accessory ar- 
 gument is drawn from internal evidence furnished by the Edicts. He 
 says, El. South Indian Palaeography, p. 2, "The inscriptions of Asoka are 
 also in themselves proofs that writing was about 250 B. C. a recent 
 practice; for they present irregularities of every kind," and in the note 
 to the passage, he quotes as instances, conflicting spellings like anapitam 
 (correctly anapitam) and anapitam, dasana and dasana (from the Girnar 
 version), the irregular insertion of nasals before consonants (which, as 
 he admits, may be due to the negligence of the masons) and the cons- 
 tant neglect of the reduplication of consonants e. g. in piyasa for piyassa, 
 arabhisante (read arabhisare) for drabbhissare and so forth. These facts 
 are indisputable, and other similar ones like the irregular employment 
 of the signs for sibilants in Kalsl, Bairfit II and Siddapur and the constant or 
 nearly constant use of short i and u for long z and u in Kalsl, Bairat 
 and Kfipnfith may be added. But they do not prove the proposition, in 
 support of which Dr. Burnell adduces them. The numerous double and 
 even treble forms of the same words, which occur in one and the same 
 version are not graphic. Some are, as the analogies in the ancient lite- 
 rary Prakrits show, real variants which occurred in one and the same 
 dialect, and some are due to slips in the translation of the Magadhi 
 originals of the Edicts into the Western and Northwestern dialects. The 
 irregular use of the Anusvara before consonants may also be ascribed 
 partly to the same cause, since the Pali too occasionally omits a nasal 
 and then doubles the following consonant. In other cases it may be due 
 to the carelessness of Asoka's clerks, who treated their Vernaculars ex- 
 actly as a modern Karkun treats his. Everybody who has had any ex- 
 perience of Indian office work, must know that the Sheras or official 
 papers, prepared by the clerks of the older generation, who had received 
 their elementary instruction in the indigenous schools and had after- 
 wards been trained in the offices, show an extreme irregularity in the 
 use of the Anusvaras, of the short and long i and w, of the three signs 
 for sibilants and of the reduplication of consonants. During my service 
 as Educational Inspector I have seen a great many, sometimes impor- 
 tant, documents from British Government offices and from those of native 
 princes, which in their spelling were quite as bad as, and even worse 
 than, Asoka's Edicts, and I do not recollect that any papers, except those 
 sent by carefully trained schoolmasters, were quite exact. One cause of 
 this state of things was the wretched instruction in the indigenous schools,
 
 III. The Antiquity of the Brahmi. 45 
 
 [43] from the right to the left. 1 The letters agree exactly with Aso- 
 ka's and the dha has the position required for the writing from 
 the left to the right. It dates, therefore, from a period during 
 which the Brahma characters were written in both directions. 
 This period is probably not very far distant from the middle 
 of the third century B. C., as the Edicts still show single let- 
 ters, which belong to the writing from the right to the left, 
 viz., the dha, given in the first Comparative Table No 4, Col. V, a, 
 the 0, ibidem, No 6, Col. VI, f, and the ta, ibidem, No 22, 
 Col. V, 1. The coin may, therefore, be somewhat younger than 
 Sir A. Cunningham thinks, who assigns it to 400 B. C. The 
 great antiquity of its find-spot, the town of Eran, is attested 
 by an inscription on the Sanchi Stupa, where it occurs as 
 Erakina. 
 
 A still more important palaeographic witness for the an- 
 tiquity of the Brahma alphabet is the variety, found in the 
 inscriptions on the Bhattiprolu relic caskets, the value of which, 
 I am sorry to say, I have somewhat underestimated in the in- 
 
 where writing is taught according to the Barakhadi, described above, 
 which contains no ligatures and more sibilants than the Vernaculars 
 possess, and where composition received little or no attention, being at 
 the best confined to the copying of a few forms for letters. Another 
 cause was the want of a settled system for the minutiae of vernacular 
 grammar, both in writing and in speech (compare my remarks, Ep. Ind., 
 Vol. Ill, p. 136) all efforts in this direction being of quite recent date. 
 A third cause is, I fear, the deeply rooted tendency of all Hindus to in- 
 accuracy in small matters. Now the Asoka Edicts are official papers, 
 written by his lipikaras or clerks. As there is no reason for assuming 
 that they had received a better education than the men in the modern 
 offices of British and native India, and that the Vernaculars of the third 
 century B. C. had been polished and perfected by grammarians, it seems 
 difficult to ascribe the defects in their spelling to other causes than 
 those which produce the same imperfections in the office work of the 
 modern Karkuns. The probability that the causes are the same in- 
 creases, if it is remembered, that even the Sanskrit landgrants, issued 
 from the secretariats of later kings, show the same mistakes, sometimes 
 in a very high degree. Thus the Valabhl grant, published in the Ind- 
 ian Antiquary, Vol. VII, p. 68 f., has no long , very few long u, invar- 
 iably dha for ddha and almost regularly Sa for $a. 
 
 Coins of Ancient India, p. 101 and plate XI, 18. According to a plaster 
 cast, which I owe to the kindness of Mr. Rapson, the inscription is Dhn- 
 mapalasa.o, the last consonant being illegible.
 
 46 Indian Studies. No III. 
 
 troductory remarks to my edition in the Epigraphia Indica, 
 Vol. II, p. 323 ff. Though the article has been published only 
 in 1894, it was written more than two years earlier, before I had 
 begun to study the question of the derivation of the Brahma 
 [44] alphabet. Then, I saw only that the inscriptions probably be- 
 long to nearly the same period as the Edicts and that their 
 alphabet, which offers the six independent Matrkas gh, j, 
 m, I, s and I and the curious notation of a and a, as well 
 as some minor differences in the radical signs for c, d and 
 bh, must be considered as coeval with Asoka's Brahma let- 
 ters. Hence I drew the inference that in the third century 
 the Brahma characters showed in certain districts even greater 
 local varieties than appears from the Edicts alone, and I pointed 
 out that this discovery greatly bettered the position of those 
 who, like myself, hold the art of writing to have been pract- 
 ised for many centuries before the times of the great Maurya 
 reformer. So far I have nothing to change or to add. 
 
 But the comparison of the Indian characters with Semitic 
 signs, which 1 have instituted since, has greatly altered my 
 opinion regarding the palaeographic value of the independent 
 signs. I no longer believe in the possibility to regard the gha 
 of the Edicts as a derivative from ga, and admit now that the 
 Bhattiprolu gh (below plate I, No 3, Col. VI) is an indepen- 
 dent form, the framer or framers of the alphabet having dis- 
 carded one of the old Semitic radicals, which the common 
 Brahma alphabet retains. I further must admit that the BhaJ- 
 tiprolu j (below plate I, No 7, Col. V, a) and 9 (plate I, 
 No 15, Col. V) are older forms than the corresponding ones 
 of the Edicts, the former being a tolerably faithful representa- 
 tion of the oldest form of Zain and the latter being a simpli- 
 fication of the Semitic Samech, turned topsy-turry. The I and 
 I (below plate I, No 12, Col. V, 2, Col. VI), regarding which I 
 did not say anything definite, I must now declare to be indepen- 
 dent evolutions from the ancient Semitic Lamed, and even in 
 the tailed c (below plate I, No 18, Col. V, 3) I now recognise an 
 archaic form. On the other hand, for m fl, which I took to 
 be a possibly older form of the ordinary Brahma sign, I can 
 no longer claim this distinction. It is merely the oldest ma of the 
 Edicts, placed top downwards. This change of opinion regard-
 
 III. The Antiquity of the Brahml. 47 
 
 ing- the details, the necessity for which will become more fully 
 apparent in the next part of this paper, naturally forces me to 
 modify the general proposition that the Bhat^iprolu alphabet 
 does not teach us much regarding the history of the Brahma 
 [45] writing and regarding the conversion of the Semitic letters 
 into Indian characters. It certainly furnishes us with valuable 
 intermediate "forms for four radical signs, c, j, I and s, which 
 latter appears to be the parent of the ordinary Brahma sa and 
 sa, and with one entirely independent derivative sign gh. Its 
 separation from the ordinary Brahma alphabet must fall not 
 only before the third century B. C. 7 but also before the time, 
 when the Eran coin was struck, and cannot have happened at 
 a later period than the fifth century B. C., though it may fall 
 much earlier. 
 
 This estimate carries us back to the period, for which the 
 passages of the Jatakas, the Pacittiya and the Mahavagga, 
 quoted above, assert the common use of writing, though they 
 do not give the name of the characters employed nor any de- 
 tails regarding them, by which they might be identified. The 
 coincidence makes it of course tempting to assume that the 
 writing, referred to in the Buddhist Canon, is the Brahml lipi. 
 And the correctness, or at least the great probability of this 
 assumption, I think, is made apparent by the recent discoveries 
 regarding the relative position of the Brahml and the Kharo- 
 sthi the only other script which could come into question 
 as well as by the facts bearing on the origin of the Kha- 
 rosthl. 
 
 The late finds of very ancient inscribed coins in North- 
 western India leave no doubt that according to the epigraphic 
 evidence the Brahml lipi was since the beginning of the his- 
 torical period the paramount Indian alphabet, used from the 
 Himalayas to Cape Komorin and from the Khyber Pass to the 
 Bay of Bengal, while the Kharosfhi held always a secondary 
 place only in a very confined territory. Again, the clear evid- 
 ence of the forms of the Kharosthl letters, the original stock 
 of which is doubtlessly derived from the Aramaic alphabet of 
 500 400 B. C., shows that this alphabet cannot have been 
 developed, much less have penetrated into Eastern India at 
 the early period to which the Buddhist works refer.
 
 48 Indian Studies. No III. 
 
 The first point, which is of considerable general import- 
 ance for Indian palaeography, will best become intelligible by 
 a brief review of the epigraphy of those districts where Kha- 
 rosthl inscriptions occur. From the third century we have the 
 two Kharosthl versions of the Asoka Edicts, incised in the 
 [46] northwestern corner of the Panjab, at Shahbazgarhi and Man- 
 sehra. To the same or possibly a somewhat earlier period belong 
 Sir A. Cunningham's coins from the site of the Taxila or Tak- 
 sasila, which prove the contemporaneous popular use of the 
 Brahma characters in Gandhara. 
 
 These coins have been figured by Sir A. Cunningham in 
 his Coins of Ancient India, plates II and III. He has pointed 
 out, op. cit. p. 61, that they are partly punch-marked silver 
 pieces and partly single or double die copper pieces, all of the 
 standard peculiar to India, and he takes them for this reason 
 and on account of the very archaic forms of the letters of the 
 legends, "to be anterior to the Greek conquest of Alexander". 
 It will perhaps be safer to say "anterior to the Greek conquest 
 of Demetrius". From their inscriptions, which are partly in 
 Brahma characters of the Kalsl type 1 and partly both in 
 Brahma and in Karosthi letters, Sir A. Cunningham has al- 
 ready drawn the obvious inference, that both alphabets were 
 used in Northern India during the third century B. C. Some- 
 thing more, it seems to me, may be elicited from an analysis of 
 the legends. 
 
 On the coin, plate II, 17, the Brahma legend Vatasvaka 
 corresponds to Sanskrit Vatasvakdh and probably means "the 
 Vata - Asvakas" or "the Asvakas of the Vata or 'fig-tree' di- 
 vision". It is well known that there was an Asvaka tribe in 
 Northwestern India, whom the Greeks call Assakenoi and state 
 (Arrian, Indica, I. 1) to have inhabited the country west of the 
 Indus as far as the Kophen. It may further be mentioned that 
 some old Indian tribes, like the Yaudheyas 2 were actually di- 
 vided into sections or ganas, as well as that, as the case of 
 the Audumbaras shows, tribes were occasionally named after 
 
 1 The Kalsl type is visible in the looped ja, 
 
 * See, Sir A. Cunningham, Arch. Surv. Reports, Vol. XIV, p. 141 and plate 
 XXXI, where coins of their second and third ganaa are described.
 
 III. The Antiquity of the BrahmT. 49 
 
 trees. With this explanation the coin appears to have been 
 issued by one of the subdivisions of a tribe, which occupied just 
 those districts, from which so many Kharosthl inscriptions come, 
 and a Brahma inscription on a tribal coin would certainly in- 
 dicate that the alphabet was in popular use. 
 
 [47] The latter point comes out still more strongly through 
 some other coins, figured on plate III, 1 viz. 
 
 Obverse Reverse 
 
 No 9 Dujaka (Kharosthl) Negama (Brahmi) 
 
 No 10 Dojaka (Brahmi) Negama (Brahmi) 
 
 No 8 [T]alima[ta] (Brahmi) [NJegafmJ (Brahmi) 
 
 No 11 A[taka?Jtaka (Brahmi) NegamfaJ (BrahmT) 
 
 The word negama is common enough in Pali and in the 
 epigraphic Prakrits, and means always 'the traders'. It shows 
 here that the coins are mercantile money-tokens, issued by 
 traders, and the words on the obverse may be either names 
 of towns or of guilds. 2 The latter explanation is perhaps the 
 more probable one. But however that may be, the use of both 
 alphabets by traders proves indeed that both were in popular 
 use in the heart of Gandhara. 
 
 The other coins of the same period, struck outside the 
 Panjab, show legends in Brahmi, even that from Mathura, 
 op. cit. plate VIII, 1, in which town also a votive inscription 
 in Brahma characters of the third century (Reports Arch. Sur- 
 vey, Vol. XX, plate VI) has been found. Only in one case 
 Kharosthl letters have turned up further south, but under cir- 
 cumstances, which do not allow the inference that the alpha- 
 
 1 Though the letters are perfectly distinct, Sir A. Cunningham gives er- 
 roneously Nekama as the reading of the Kharosthl legend of No 9. On 
 the obverse of No 8 he reads Rdlimata, and he takes negama as the 
 equivalent of the Greek No'[Ata[ia, for which explanation there is no au- 
 thority (see also Ind. Ant., vol. XXI, p. 346). He correctly points out 
 that, on plate II, Nos 21 & 22 have the inscription Kadasa, i. e. Kalasa, 
 in Brahma letters of the type of the Edicts, and asserts that No 13 on 
 plate III bears the Kharosthl legend Pamca Nekamma. According to the 
 autotype the third sign is not ne but a and the fourth looks like kra. 
 
 2 I am unable to explain Dujaka- Dojaka. Talimata seems to be connected 
 with the Sanskrit Tali "Corypha Taliera" or "Flacourtia cataphracta". 
 
 Biihler, Indian Studies. III. 4
 
 50 Indian Studies. No III. 
 
 bet was generally used or known. This case occurs in the 
 Siddapur Edicts where the writer Pa<Ja has added at the end 
 his qualification lipikarena 'the scribe' in Kharosthl characters. 
 This looks like a joke or a boast, as if Pada, proud of his 
 accomplishments, had been anxious to make it apparent that 
 he knew more than the ordinary characters. And, as he was 
 [48] in the royal service, it is not unlikely that he may have ac- 
 quired a knowledge of the Kharosthl during a stay in a north- 
 ern office, or may have been a Northerner by birth. 1 
 
 From the second and first centuries B. C. we have chiefly 
 legends on coins, which were struck in the Panjab or in non- 
 Indian countries further west. The Indo-Grecian kings gener- 
 ally use Kharosthl letters, but Agathokles and Pantaleon em- 
 ploy also Brahma characters, showing thereby that this alphabet 
 likewise continued to be used in the Northwest, side by side 
 with the Kharosthl. The same fact is proved by the double 
 legends in Kharosthl and Brahma letters on Sir A. Cunning- 
 ham's Audumbara and Kuninda coins, op. cit. pi. IV 2 and 
 V, 1 6, which come from the same districts and probably 
 belong to the same time. Further east in Kosambl, Ayodhya 
 (op. cit., pi. V, 718 and IX) and Pancala (pi. VII) as well 
 as further south in Ujjain (pi. X) none but Brahma letters 
 occur. And it is curious that even the Yaudheyas (pi. VI), 
 who were settled on the lower Satledge, use only the latter 
 characters. A few single letters on the gateway of the Bharahut 
 Stupa, among which there is a sa of the second century, (Cun- 
 ningham, the Stupa of Bharahut, pi. VIII), are probably marks 
 of northern masons, who erected this additional portion of the 
 monument for Dhanabhuti. 
 
 During the period of the 6aka and Kusana kings, from 
 the first century B. C. to the end of the second century A. D., 
 
 1 See the Wiener Zeitschrift f. d. Kunde d. Morgenlandes, Vol. VII, p. 30 f. 
 Dr. Grierson, Ind. Ant., Vol. XXIV, p. 247, points out an analogous 
 practice of the modern Kayasths of Bihar who sign the documents, 
 written in the Kaithl character, with their names in Arabic letters. 
 
 8 The Vrfni-coin No 15 is according to the evidence of the characters 
 later and probably belongs to the period when the Sakas had carried 
 the Kharosthl as far as Mathura. A few votive inscriptions of the second 
 and first centuries B. C. from the same town are in Brahma characters, 
 Epigraphia Indica, Vol. II, p. 195.
 
 III. The Antiquity of the Brahinl. 51 
 
 the numerous inscriptions, incised in the Western Panjab, are 
 in Kharosthl and those on Dr. Bhagvanlal's Lion Capital prove 
 that the Kharosthl penetrated during the reign of the Saka 
 Satrap and king 6u<Jasa or o<Jasa as far as Mathura, where 
 however, as the enormous quantity of Jaina and Bauddha votive 
 inscriptions shows, the Brahma alphabet was at that time and 
 later the usual one. [49] The coins of the earlier &aka kings from 
 the North like those of Mauos and Azes and of their Satraps, 
 as well as those of Gondopherres and his brothers, have Greek 
 and Kharosthl legends, while the later Kusana kings, Kaniska, 
 Huviska and Bazodeo or Vasudeva discard the latter, and their 
 successors finally adopt the Brahma alphabet. Further east and 
 south the king and Satrap Rajubula and his son &ug!asa, who 
 ruled over Mathura and perhaps over other portions of the eastern 
 Panjab, either follow the same practice as Mauos or use Brahma 
 letters (Cunningham, op. eit, pi. VIII, 2 5), which occur also 
 on the coins of the Satraps Hagamasha and Hagamana (ibidem, 
 Nos 6 7). Further two foreign (Saka?) Satraps and kings of 
 Ujjain, Nahapana and Castana employ on their coins both the 
 Indian Alphabets simultaneously, while the inscriptions of Aya- 
 ma, the minister of Nahapana, of his daughter Daksamitra 
 and his son-in-law, the Saka Usavadata or Usabhadata, as well 
 as of the immediate descendants of Castana show exclusively 
 Brahma characters. 
 
 The epigraphic evidence shows therefore that in the third 
 century the popular use of the Kharosthl was strictly confined 
 to the Panjab and that it was nothing more than a secondary 
 script, running along by the side of the Brahmi, which pre- 
 vailed all over India. This state of things continued during 
 the next two centuries. During the period of the Saka rule 
 the Kharosthl spread further south, without however losing 
 its character as an accessory alphabet. A very strong ar- 
 gument for the assumption that its position was the same even 
 in the fourth century B. C. is furnished by Mr. Rapson's 
 discovery of Persian Sigloi with countermarks in Brahma and 
 Kharosthl letters (Jour. Roy. As. Soc., 1895, p. 865 ff.). For 
 this reason and because the general prevalence of the Brahma 
 alphabet has now become more clearly apparent, it seems very 
 probable that the Buddhist Canon can only refer to the latter. 
 
 4*
 
 52 Indian Studies. No III. 
 
 The second point, the improbability that the Kharosthl 
 was already developed or in general use even in its home as 
 early as say 500 B. C. requires here only a few remarks as 
 the details have been discussed fully below in Appendix I. 
 Its derivation from the Aramaic alphabet has been generally 
 accepted, ever since Mr. E. Thomas pointed it out, and the 
 shape of its &, na, ra and va makes a doubt impossible. Ac- 
 cording to Dr. I. Taylor's suggestion, The Alphabet, Vol. II, 
 p. 261, which [50] is most probable on historical and palaeogra- 
 phical grounds, its immediate source is the Aramaic alphabet 
 of the Akhaemenian period, and the introduction of the Aramaic 
 letters into India dates from the time, when the Akhaemenian 
 kings of Persia had conquered the Panjab, just those districts 
 where the Kharosthl has its real home. 1 As the Persian con- 
 
 1 M. Halevy, op. cit,., p. 250 ff. and particularly p. 267, tries to establish 
 a still later date for the elaboration of the Kharosthl. He seeks the 
 prototypes of its characters in the Aramaic alphabet of the fourth cent- 
 ury B. C. He believes that they were developed in Ariana not earlier 
 than 330 B. C. after the appointment of Macedonian Satraps by Alexan- 
 der, who, as he thinks, may have fostered the use of the already pre- 
 valent Aramaic writing 1 and thereby may have induced their subjects to 
 adapt it to their special wants. The obvious weakness of M. Halevy's 
 arguments prevents my accepting his theory, which would be more fa- 
 vourable for my point. 
 
 It is a matter of course that I am likewise unable to agree with 
 M. Halevy's theory (op. cit., p. 280286 and Plate II, A) according to 
 which the Brahmi lipi has borrowed from the Kharosthl six radical signs, 
 !<a, jha, da, na, u and ra, the Anunasika, the system of vowel-notation, 
 and the numeral signs 49. The reasons, why I do not consider the 
 six Matrkas and the vowel-notation of the Brahma alphabet as loans 
 from the Kharosthl, will become apparent in the next section of this 
 Essay. With respect to the Anunasika, I must point out that the sign, 
 ,, which M. Halevy derives from the Kharosthl ma V of the Indo- 
 Grecian coins, appears in no Indian document written in India before 
 1200 A. D., while the Kharosthl went out of use about 200 A. D. As 
 regards the assertion that the Brahma numeral signs for 49 are the 
 initial Kharosthl letters for the words catur, pancan, sat (old Prakrit, sa, 
 chha or sad), saptan (Prakrit satta), affan (old Prakrit asta, attha) and 
 navan (old Prakrit nava), which is also put forward by Dr. Taylor (The 
 Alphabet, II, p. 266), it is unfortunate that in four cases the facts are 
 absolutely against it. The Kharosthl cha % no doubt looks like the 
 Ksatrapa sign for four. But the Indian words for four, all begin with the 
 unaspirated ca. Again, the Asoka signs for six (see below plate IH) have on
 
 IV. The Derivation of the Brahml. 53 
 
 [51] quest probably happened shortly before 500 B. C., it is impos- 
 sible that the Kharosthi can have been developed before 450 B. 
 C., and it is not to be thought of that it could have penetrated into 
 Eastern India, where the Buddhist Canon was composed, during 
 the fifth century, much less could it have been there in so general 
 use, as the alphabet mentioned by the ancient Buddhists cer- 
 tainly was. Under the circumstances just discussed, the as- 
 sumption that the alphabet, referred to in the Jatakas, the 
 Mahavagga and so forth is the Brahml which, I repeat, the 
 palaeographic facts contained in Asoka's Edicts, the Indian 
 letters on the Persian Sigloi, the legend of the Eran coin, the 
 Bhattiprolu inscriptions strongly suggest undeniably gains a 
 very high degree of probability. 
 
 IV. 
 
 As the literary evidence points to the common use of 
 writing in India during the fifth and perhaps in the sixth cen- 
 tury B. C., and as the palaeographic evidence proves the Brahml 
 lipi to be the oldest Indian alphabet and to be probably identical 
 with the script referred to in the Buddhist Canon, it is a matter 
 of course that its source must be found in the more ancient Se- 
 mitic characters. A short time ago such a result would have 
 precluded the possibility of all atternps to make Southern Ara- 
 bia the country from which the parents of the Brahma letters 
 came. But at present, pretensions to a high antiquity are put 
 
 resemblance to the Kbarosthi aa *p. Further the Ksatrapa sign for 
 seven ^ or Gupta ft cannot be the Kharosthi palatal Sa |~| , because 
 the word for seven begins in all Indian dialects with a dental *a. Again, 
 the Ksatrapa and -Andhra l>j cannot represent the Kharosthi AS, because 
 no such ligature is ever formed in any Indian writing. Besides, if it 
 had been formed, it would be wrong. The second part ought to be the 
 lingual sa *P, because the word is astan not aSfan. Finally, the 
 nought of the ninth and later centuries O cannot be derived from a 
 sign for 10 expressed by a circle, because no such figure exists, nor 
 from the entirely dissimilar Kharosthi da j . The Hindus call the nought 
 explicitly sunyabindu "the dot marking a blank", and about A. D. 500 
 they marked it by a simple dot, which latter is commonly used in in- 
 scriptions and MSS in order to mark a blank (see the Gruudriss, I, 11, 
 pp. 78 and 86), and which was later converted into a small circle.
 
 54 Indian Studies. No III. 
 
 forward on behalf of various epigraphic documents from the 
 latter country. The theories, it is true, are still conflicting. 
 Some ascribe certain Sabaean inscriptions to the tenth century 
 B. C. or an earlier period, while others declare those of the 
 Minaeans to belong to very ancient times and deny the anti- 
 quity of the Sabaean documents. More light on these questions 
 is required, but even as matters stand at present, it seems to 
 me hazardous to make use of the old argument, that the de- 
 rivation of the Brahma letters from a South-Arabian source 
 is a priori impossible, because the inscriptions, found there, 
 are of too modern dates. 
 
 [52] Nevertheless the theory of a South -Semitic origin of 
 the Brahma alphabet appears to me untenable. What has been 
 brought forward in its favour by the two chief advocates, is to 
 a great extent far from convincing. The ocular evidence, it 
 seems to me, speaks against many of their identifications. 1 
 Further, in some cases late or at least secondary forms of 
 Indian letters have been used for comparison. 2 Again in other 
 cases the most extraordinary and absolutely unnecessary changes 
 in the phonetic value of the signs are assumed to be quite nat- 
 ural and hardly worthy of an explanatory remark. When one 
 is told that the Hindus utilised the Semitic Qoph to express 
 the sound chu (Deecke) or cha and ka (Taylor), that they 
 employed the Semitic Shin for ja, sa, sa and sa (Deecke) or 
 at least for ja and .sa (Taylor), that the Semitic Waw furnished 
 both the signs for ya and va, or that Lamed became ra and 
 Resh was turned into la, one can only say that such assertions 
 are hard to believe. As the Hindus are always very particular, 
 even pedantic, in matters connected with phonetics, and as 
 the franiers of the Brahmi lipi evidently have been careful 
 with respect to the formation of many derivative signs, duly 
 deriving <]ha from da, pha from pa, bha from ba and so forth, 
 it seems incredible that they should have had no regard for 
 phonetic affinities in utilising the signs, which they borrowed. 
 And the assumed vagaries appear absolutely inexplicable if 
 
 1 That is also Sir A. Cunningham's statement, Coins of Ancient India, 
 pp. 3940. 
 
 s To the former class belong Dr. Deecke's initial A and Dr. Taylor's broad- 
 backed to, to the latter Dr. Taylor's initial A, kha and ra.
 
 IV. The Derivation of the Brahml. 55 
 
 one considers that the Semitic alphabet has the Tsade, the 
 phonetic value of which comes close to ca and cha, and that 
 it possesses separate signs for Waw and Yod, which might be 
 used to express va and ya. 
 
 As long as the theories regarding the derivation of the 
 Brahma alphabet contain equations like those just mentioned, 
 and as long as these theories do not take into account all the 
 oldest forms of the Indian letters, the problem, it seems to me, 
 has not been solved and the task has to be redone. Really trust- 
 worthy results can only be attained under the following conditions: 
 
 [53] (1) The comparison must be based on the oldest forms 
 of the Indian alphabet and actually occurring Semitic signs of 
 one and the same period. 
 
 (2) The comparison may include only such irregular equa- 
 tions, as can be supported by analogies from other cases, where 
 nations have borrowed foreign alphabets. 
 
 (3) The comparison must show that there are fixed prin- 
 ciples of derivation. 
 
 In resuming the inquiry on these principles I soon found 
 that the Southern Semitic characters would not serve my pur- 
 pose. Though the Himyaritic or Sabaean alphabet on account 
 of its stiff, monumental appearance bears a general resemblance 
 to the Brahml lipi, and though two or tree of its letters come 
 close to the corresponding Indian signs, it includes so many 
 dissimilar characters, which evidently are more strongly modi- 
 fied than the Brahma Matrkas, that the real resemblances can 
 only be considered as accidental or as due to an analogous de- 
 velopment. But on trying the oldest characters, which are al- 
 most identical throughout all the countries, occupied by the 
 Northern Semites, it became possible to identify in the Brahma 
 alphabet all the twenty two Semitic letters, and to explain the 
 formation of the numerous derivative signs, which the Indians 
 were compelled to add. As stated already, the identifications 
 agree for the greater part with Professor Weber's, whose im- 
 portant essay in the Zeitschrift der Deutschen Morgenlandischen 
 Gesellschaft of 1855 very nearly solved the problem of the 
 origin of the Brahma alphabet. If Professor Weber had had 
 at his disposal the materials which are accessible at present, 
 his acumen would no doubt have removed every difficulty.
 
 56 Indian Studies. No III. 
 
 The identifications are shown in the accompanying Table, 
 kindly drawn for me by Mr. S. Pepper and etched by Messrs 
 Angerer and Goschl of Vienna. The signs of the first two co- 
 lumns have been taken from M. Ph. Berger's Histoire de 
 PEcriture dans PAntiquite, pp. 185 and 202 and those of 
 Col. Ill with the exception of the Taw from Sinjirli (22. III. b) 
 from Professor Euting's Tabula Scripturae Aramaicae (1890). 
 Among the forms of column IV without asterisks, those opposite 
 the twelfth and the seventeenth Semitic characters [54] have 
 been taken from a plaster cast of Sir A. Cunningham's Eran 
 coin (see also Coins of Ancient India, plate XI, 18) and that 
 opposite the sixteenth character from the word esa in the Kalsi 
 version, Ed. XIII, 1, 1. 37. 
 
 Among the signs in column V and VI a few are from Sanchi, 
 viz. the lingual la, the fifth of Col. VI opposite the fourth 
 Semitic character, and the first of Col. V opposite the sixteenth 
 Semitic character. Further, in Col. VI the fourth sign opposite the 
 sixth Semitic character, U, comes from Bharahut, the first 
 opposite No 15 from the Eran coin the third opposite the fifteenth 
 Semitic character in Col. VI has been taken from the Ghasundi 
 slab, and two signs of the two rows in the same Col. opposite the 
 sixteenth character, viz. the second in the first row, AI, and the last 
 in the second row, I, belong to the Hathigumpha and Nanaghat in- 
 scriptions. All the remaining letters have been drawn according 
 to Dr. Burgess' facsimiles of the Rock and Pillar Edicts, 1 and of 
 the Bhattiprolu inscriptions. 
 
 A. The Borrowed Signs. 
 
 Before proceeding to the details, it will be desirable to 
 state the principles, on which the twenty-two signs of the Se- 
 mitic alphabet seem to have been turned into Brahma letters. 
 Even a superficial examination of the Brahma alphabet shows 
 the following chief characteristics: 
 
 (1) The letters are set up as straight as possible, and 
 they are with few exceptions made equal in height. 
 
 1 The facsimiles of the Girnar and Kalsi versions, on which I have chiefly 
 drawn, have been published, Epigr. Ind. vol. II, p. 447 if. The Pillar 
 Edicts have appeared op. cit., p. '245 ff. and the Bhattiprolu inscriptions 
 op. cit. p. 323 S.
 
 IV. The Derivation of the Brahml. 57 
 
 (2) The majority consists of vertical lines with append- 
 ages attached mostly at the foot, occasionally at the foot and 
 at the top, or rarely in the middle; but there is no case where 
 an appendage has been added to the top alone. 
 
 [55] (3) At the top of the characters appear mostly the ends 
 of vertical lines, less frequently straight horizontal lines, still 
 more rarely curves or the points of angles opening down- 
 wards, and quite exceptionally in ma (No 13, Col. V) two lines 
 rising upwards. In no case does the top show several angles, 
 placed side by side, with a vertical or slanting line hanging 
 down from it, or a triangle or circle with a pendant line. 
 
 The principles or tendencies, which produced these cha- 
 racteristics, seem to be a certain pedantic formalism, a desire 
 to have signs well suited for the formation of regular lines, 
 and a strong aversion against all top-heavy characters. 1 The 
 natural result was that a number of the Semitic signs had to 
 
 The appearance of a pedantic formalism in the alphabet is no more than 
 what might be expected. For it is a characteristic of most Indian cre- 
 ations in literature, science and art. The aversion, shown by the Hindus 
 against letters with heavy tops, has also a good and sufficient reason, 
 and, I think, it is chiefly due to their making the letters hang down from 
 a top-line instead of placing them on a base-line (see, Beruni, India I, 1 72). 
 The modern Karkun or clerk of Western, arid, as Dr. Grierson informs 
 me, also of Eastern India, as a rule, actually draws this line across the 
 whole breadth of the paper, and the modern Lekhak or copyist places 
 a small board with thin threads fastened at equal distances below his 
 sheet of paper, or squeezes the paper between the strings and the board. 
 The general appearance of the most ancient MSS., like the Bower birch 
 bark volume and the Horiuzi palmleaves, makes it highly probable that 
 their writers used the same or a similar contrivance. Even the look of 
 the Asoka inscriptions, especially of the Pillar Edicts and of the Girnar 
 and Jaugada Rock Edicts, indicates that their eng'ravers, or the writers 
 of the original copies, were accustomed to the use of top lines. Though 
 they are rarely quite exact, it is clearly visible that they at least tried 
 to equalise the height of the tops. If the custom, as is not improbable, 
 dates from still earlier times, the aversion against heavy tops is easily 
 explained. For signs beginning with vertical lines or with short hori- 
 zontal strokes are most suitable for such writing, and the later modifi- 
 cations of the Brahma letters are to a great extent the results of a 
 variety of attempts to obtain such forms. Another reason may have 
 been the introduction of the vowel-signs, which mostly are added to the 
 tops of the consonants.
 
 58 Indian Studies. No III. 
 
 be turned topsy-turvy or to be laid on their sides, while the 
 triangles or double angles, occurring at the tops of others had 
 to be got rid of by some contrivance or other. A further 
 change in the position of the signs had to be made, when the 
 Hindus began to write from the left to the right. They had, 
 of course, to be turned from the right to the left, as in Greek. 
 [56] Instances, where the old position has been preserved, are 
 however met with both in borrowed and derivative signs. 
 
 With respect to the single borrowed letters I have to 
 add the following remarks. 
 
 No 1. That the Hindus, like the Greeks used the Se- 
 mitic Aleph in order to express the vowel a has been suspected 
 by Professor Weber. His successors have asserted the identity 
 of the first letters of the two alphabets as a fact, though the 
 Indian forms of the initial A with two curves, which have 
 been mostly used for comparison, 1 do not agree very well. 
 The doubtlessly oldest form with the angle to the left of the 
 vertical (Col. V) which, as stated above, is peculiar to the 
 Southern versions of the Edicts, agrees almost exactly with 
 the oldest Semitic signs. Only the vertical has been placed at 
 the point of the angle in order to make the appearance of the 
 letter more regular, and the whole sign has, of course, been 
 turned from the right to the left. On a Patna seal it occurs, 
 however, once in its original position, see the Grundriss, 
 Taf. II. 1. I, and p. 8. 
 
 No 2. I can only agree to the identification of the Brahma 
 la with the Semitic Beth, which has been asserted on the strength 
 of the Sabaean arid Ethiopian forms 2 by Professor Weber and 
 his successors, with the exception of M. Halevy, who derives 
 it from the dissimilar Greek Beta. But I think that it is 
 a development, derived by the Hindus immediately from the 
 
 1 M. Halevy, however, correctly compares the angular form. His attempt 
 to derive it from the Greek Alpha is not intelligible to me. The two 
 signs possess a minimum of resemblance to each other. 
 
 2 The Sabaean form is no doubt very similar to the Brahma sign but, I 
 think, due to a different modification of the ancient North Semitic letter. 
 It would seem that the triangle at the head of the letter has been turned 
 into a square and the stroke on the right has been made straight. Hence 
 arose first the pj and by a simplification [~|-
 
 IV. The Derivation of the Brahml. 59 
 
 old Phoenician and Mesa signs. As the Hindus did not toler- 
 ate heavy tops, they opened the triangle and made the central 
 stroke hang down from the end of the angle in the manner 
 shown in the hypothetical form in Col. III. 1 The next result was 
 a rhomboid form (Col. V, a), a few examples of which are found 
 occasionally in various versions of the Edicts, where we have e. g. 
 -O- bo in apalibodhaye, Kalsi Edict V, 1. 15. The liking of the 
 [57] Hindus for straight top -lines produced further the square and 
 the oblong ba, which latter is the commonest form in Kalsi 
 and the regular one in the Bhattiprolu inscriptions. It may 
 be noted, that this letter is frequently, though not regularly, 
 made smaller than ' those with verticals. Regarding the want 
 of the lower line in the derivative bha (Col. VI) see below. 
 
 No 3. The identity of the Indian ga with the Semitic 
 signs is so evident that it has been always recognised. I will 
 only add that there are numerous instances in the Edicts where 
 one of the sides of the angle is shorter than the other. The 
 formal looking very pointed angles with perfectly straight equal 
 sides are characteristic of the Western and Southern versions. 
 
 No 4. The identity of the Brahma dha with the Semitic 
 Daleth has "also been generally admitted, Dr. Taylor, who de- 
 rives it from tha (No 9, Col. V) alone dissenting. To me it 
 seems that the first form, given in Col. V, with the vertical on 
 the right, is the immediate offspring of the oldest Semitic forms, 
 the letter having been made higher and the right hand stroke 
 straightened in accordance with the general principles of forma- 
 tion, stated, above. The round back is probably due to a cur- 
 sive development. In the Edicts this form is rare, and the 
 second, given in Col. V, which is younger and due to the change 
 in the position required by the change in the direction of the 
 writing, is found in the great majority of cases. The older 
 one (Col. V, a) is used exclusively in the Bhattiprolu inscriptions 
 and usually in the Western cave inscriptions of the Nanaghat, 
 Nasik, Karle, Junnar and so forth, which range from about 
 150 B. C. to 200 A. D. It is also the parent of all the later 
 
 An apparently analogous modification of the old Semitic Beth which one 
 might be inclined to see iu the b 2 of the Orkhou inscriptions (V. Thomson, 
 Insr. de TOrkhou dechiffrees, p. 24) is disposed of by Prof. Doniior's re- 
 marks on the sign in his essay, sur 1'origine de 1'Alphabet Turc, p. 57 f.
 
 60 Indian Studies. No III. 
 
 northern forms of the letter, including the present Devanagarl 
 dha. In the caves mentioned we find also in the Andhra and 
 other inscriptions of the first and second centuries A. D. numer- 
 ous triangular forms 1 like ^ or ^. They may be possibly an- 
 cient survivals, especially as the Edicts too occasionally offer 
 forms with half angular backs like ^. 
 
 The difficulty with respect to the phonetic value of the 
 Brahma sign, which does not exactly correspond to that of 
 [58] the Dalvth, may perhaps be explained by the suggestion that 
 the letter was used for a long time both for the unaspirated 
 and the aspirated dental, and that the da was separated at a 
 time, when the real value of the borrowed sign had been for- 
 gotten. 
 
 No 5. Professor Weber denies the connexion of the 
 Brahma ha with the Semitic He, while Drs. Deecke and Tay- 
 lor derive it from the Sabaean form y. 2 As long as only 
 the round form, given as the third sign in Col. V, and similar 
 ones were known to exist in the Edicts, the case was a diffi- 
 cult one. Dr. Burgess' new impressions and facsimiles show, 
 however, that the angular ha, Col. V, b, which is constant in 
 the cave inscriptions of the next centuries, occurs here and 
 there in Asoka's inscriptions. 3 The new Siddapur version 
 shows throughout the little horizontal bar (on the right) attached 
 very close to the foot of the letter and offers in the word ma- 
 hatpane No. I, 1. 4) the sign, placed first in Col. V. This is al- 
 most exactly a turned He of the Assyrian weights (Col. Ill, b) 
 which according to M. de Vogue" 4 appears from about 700 B. C. 
 Professor Euting kindly points out to me that the very similar 
 He of Col. Ill, 1 is found on a Mina (C. I. S., P. H,_No &>\ 
 which according to the accompanying cuneiform inscription dates 
 from the reign of Salnianassar, and therefore, as also M. de 
 Vogue has stated in his remarks on the inscription, is older 
 
 1 See Burgess, Arch. Surv. Reports West-Ind., Vol. IV, Plates XL1V, 17; 
 XLVIII, 16, 17; and the Grundriss, I, 11, Tafel III. 24. VII, XI. 
 
 2 This sign is due to a very much stronger modification of the ancient 
 North-Semitic He than the Brahma letter. 
 
 3 See also above p. 41, and the Grundriss, Tafel II. 40. VI. 
 
 4 Corpus Inscr. Semit., P. II, Vol. I, p. III. Mr. Berger enters this sign 
 (Table, op. cit., p. 300) in the column for the Persian period.
 
 IV. The Derivation of the Brahmi. 61 
 
 than the year 725 B. C. As this sign belongs to the eighth 
 century B. C., when the very archaic forms of the Aleph, Daleth, 
 Cheth, Theth, Wav and Qoph, represented by the corresponding 
 BiTihma letters, still existed/ it may be considered the proto- 
 type of the Indian ha. The aversion of the Hindus against 
 heavy tops, of course, caused the sign to be turned round, 
 and their liking for regularity caused the bar to be attached 
 to the stroke which then became the base-line, and the whole 
 letter to be set up straight. All things considered, this explan- 
 ation seems [59] more probable than attempts to connect ha 
 with the later more exactly agreeing sign in Col. Ill, b, or 
 with the still older form with three parallel bars. 
 
 No 6. The Indian va, regarding which Professor Weber 
 felt uncertain, evidently corresponds with the Mesa form of Wdw. 
 Owing to the aversion against heavy tops the Semitic letter 
 has been turned topsy-turvy. At the same time the semicircle 
 has been closed. Among the two forms, given in Col. V, the 
 second is the regular one in the Edicts, the first, as well as 
 an angular variety not given, 2 occurs only rarely. 
 
 No 7. Professor Weber has already pronounced himself 
 for the identity of the Brahma ja with the Zain. On phonetic 
 grounds Professor Weber's conjecture is extremely probable. 
 For, ever since the Indians began to borrow Persian and Arabic 
 words, they have rendered the Zain of the latter by their ja, 
 and the dictionaries of the Indian vernaculars contain hundreds 
 of instances of this kind. Even in earlier times, about A. D. 
 500, they express, e. g. in jiva for Zsu; (Ziv-s) the Greek 
 Zeta, which then had the value of Zain, by ja. Both facts 
 prove that the two sounds had a close affinity for their ear. 
 The discovery of the Bhaftiprolu j? Col. V, a, considerably 
 facilitates the identification of the two signs, as it closely agrees 
 with the archaic Phoenician sign, differing only slightly in the 
 position of the bars at the top and the foot. The signs of the 
 
 1 See M. de Vogue, op. cit., p. II. 
 
 2 Compare also above p. 41, where a fourth cursive form has been figured. 
 
 3 Dr. Burgess' facsimiles in the Epigraphia Indica show occasionally an 
 incipient central bar in ju and ju, where it ought not to occur, 
 impressions prove clearly that in all these oases there is not a real third 
 line, but that the stone has a flaw.
 
 62 Indian Studies. No III. 
 
 Edicts, given in Col. V, b d, are all cursive, and represent 
 attempts to make the letter with a single stroke or at least 
 with two. The third form with a dot at the junction of the 
 two curves, is of course a development from the second with 
 the loop. And the substitution of a dot for a loop indicates 
 that the persons who invented it wrote with pen and ink. A 
 man working with a stylus would not think of such a substitution, 
 nay could hardly effect it. 1 
 
 No 8. The discovery of the angular gha (Col. V, a) which 
 occurs a few times in the Kalsi version 2 and is constant in the 
 cave inscriptions, makes its identification with the three-barred 
 [GO] Chefh of the archaic Phoenician, of the Sinjirli and other 
 ancient inscriptions, very probable. The Semitic sign, which 
 often slants towards the left, has been laid on its side (see the 
 hypothetical form in Col. IV) and the long upper bar, covering 
 the letter, has been raised upwards and converted into the long 
 vertical on the left. 3 The usual form of the Asoka Edicts with 
 the curve is of course cursive. In the Bhattiprolu alphabet this 
 letter has been discarded and a derivative from ga (No 3, 
 Col. VI) takes its place. The Sabaean Cheth ^, from which 
 Dr. Taylor derives the Indian gha, is no doubt an analogous 
 development from the old North-Semitic sign. 
 
 No 9. The identity of the Brahma tha with the Semitic 
 Theth has always been recognised and is evident enough. I 
 may mention as a curiosity that a small cross appears inside 
 the circle instead of a dot in the Pallava landgrant, Epigraphia 
 Indica, Vol. I, p. Iff. In this document all dots are replaced 
 
 1 Regarding the use of ink in India, see above p. 22. The ancient Indian 
 term for ink is masi, often spelt masi, and now pronounced malchi, which 
 is derived from the obsolete verb mas, hinisdyam, compare masa and 
 mo.yma.9a. Etymologically and originally it means 'powder 1 (of charcoal 
 and the like), used for the preparation of ink, see the larger Petersburg 
 Dictionary under masi, and the Grundriss, I, 11, p. 91. 
 
 2 The sign given in the Table has been taken from upaghate, Ed. XIII, 1, 
 1. 37. 
 
 8 It may be mentioned as a curiosity, that from the twelfth century 
 A. D. the Nagarl gha is again placed upright and has become almost ident- 
 ical with its Semitic prototype, see the Grundriss, Taf. V. 13. XX XXII. 
 This is, however, not due to a survival, but merely to the desire of 
 equalising the breadth of the Nagarl letters.
 
 IV. The Derivation of the Brahml. 03 
 
 by minute crosses. On the other hand, a Theth with a dot 
 occurs on an Assyrian weight, see Col. III. 
 
 No 10. The derivation of the Brahma ya from the an- 
 cient Semitic Yod has been asserted by Professor Weber, who 
 however could only compare the later round Phoenician form, 
 which opens downwards. 1 The oldest Indian form is probably 
 the notched one (Col. V, a), 2 which appears exclusively in the 
 Delhi Pillar Edicts and occasionally in most other versions and is 
 the parent of all the later developments. It is probably de- 
 rived from the oldest Semitic Yod, formed of a downstroke with 
 two bars, attached to the left, and one to the right. Its forma- 
 tion seems to have been effected by placing the top-heavy sign 
 on its side, in the manner shown in the hypothetical form of 
 [61] Col. Ill, by lengthening the stroke which now stood in 
 the middle of the line and equalising the direction of that on 
 the right. 
 
 No 11. The derivation of the Brahma ka was a difficult 
 problem, as long as the cross with four perfectly equal bars 
 (Col. V, b) was considered to be its most ancient form, and 
 consequently the earlier attempts to account for its existence 
 are not very satisfactory. Professor Weber, following Kopp, 
 and Dr. Taylor have tried to derive it from the Semitic Qoph 
 very much against the ocular evidence, and M. Halevy has 
 identified it with the Aramaic Kaph, which likewise, as far as 
 I can see, does not resemble the Indian letter. But now, when 
 the more accurate reproductions of the Edicts have proved the 
 frequent early occurrence of the dagger-shaped form (Col. V, a) 
 which is constant in Bhattiprolu and in all the various inscrip- 
 tions, incised between 200 B. C. and 550 A. D., the case of 
 the sign is by no means desperate nor even difficult. If, as now 
 may be done with perfect propriety, the dagger-shaped ka is 
 taken as the older form and the cross with equal bars as a 
 regularised or formalised derivative, it is easily recognisable 
 that the dagger-form probably has been derived from a Kaph, 
 like that of Mesa's inscription (Col. II), in which the down- 
 
 1 Is has also been admitted by M. Halevy, who compares a late form of 
 the Semitic letter. 
 
 2 That with the single curved stroke at the foot of the vertical is appa- 
 rently cursive.
 
 64 Indian Studies. No III. 
 
 stroke slants sharply to the left and the angle on the left is 
 attached very close to the top, compare also the signs of the 
 Siloah inscription, in Prof. Euting's Tabula Script. Aram. Col. 3. 
 In setting the sign up straight, the Hindus took the upper stroke 
 of the angle for the top of the down-stroke, and a cross-like 
 figure was the natural result, which, on being made more 
 regular, yielded the dagger-shape. 
 
 No 12. For the identity of the Brahma la (Col. V, a) 
 with the Semitic Lamed (Cols. I and II), which has been re- 
 cognised by Professor Weber and accepted by Dr. Deecke and 
 M. Halcvy, the la of the Eran coin (Col. IV), the Bhattiprolu 
 la (Col. V, b) and the Bhattiprolu I (Col. VI) furnish new 
 illustrations. The most ancient survival is found in the last 
 mentioned character, which, if the little bar denoting the ling- 
 ualisation is removed, almost exactly represents the Semitic 
 sign. Next comes the letter of the Eran coin, which has pre- 
 [62] served the original direction of the Lamed, but shows on the 
 right the little bar, no doubt intended to mark the end of the 
 line. This bar is wanting in the Bhattiprolu I, which how- 
 ever has been turned from the right to the left and has deve- 
 loped a curious long tail, hanging down from the top of the 
 main line. Originally this appendage no doubt was a flourish, 
 marking the end of the % vertical. 
 
 No 13. The identification of the Brahma ma with the 
 Semitic Mem is also due to Professor Weber. He, as well as 
 M. Halevy, derive it from the Aramaic sign of the seventh 
 century 1 B. C., V/. Though this derivation is no doubt tempting 
 it seems to me that it is equally possible to connect the Brahma 
 ma with a sign like that of the Mesa inscription, which likewise 
 shows a curve at the end of the down-stroke. A little stronger 
 development of this curve would necessarily lead to a sign, 
 like the hypothetical one, entered in Col. IV, which is almost 
 the same as the first Pehlevi Mem of Col. 58 in Professor 
 Euting's Tabula Script. Aram. And in support of this view I 
 may appeal to the development of the looped ka, Northern ^ 
 and Southern Jf from the dagger-shaped ka with a curve at 
 he foot ;f, as well as of the round ad Southern ra "\J from J, which 
 
 1 See M. de Vogue, loc. cit.
 
 IV. The Derivation of the Brahml. 65 
 
 show analogous changes on purely Indian ground. The super- 
 position of the angle, which originally remained at the side, 
 over the round or elliptical figure 1 is of course due to the 
 striving after regularity, which the Hindus show throughout. 
 The Brahma ma with the angle at the top (Col. V, a) which is 
 (see also above p. 39) particularly common in the Girnar ver- 
 sion and not found in the Northern versions, thus appears to 
 be the older form, and that with a curve (Col. V, b) the cursive 
 one. In the Bhattiprolu alphabet the ancient Brahma ma with the 
 angle appears turned topsy-turvy (see p. 46). The Sabaean, Beta- 
 like Mem, $ from which Dr. Deecke and Dr. Taylor derive the 
 Brahma ma, shows little resemblance and is probably the product 
 of a different process. 
 
 [63] No 14. Professor Weber, 2 who is followed by Dr. 
 Deecke, considers the Semitic Nun to be represented by the 
 Brahma na, Col. VI, b, and it is Dr. Taylor alone who connects 
 the dental na, Col. V, directly with the Sabaean form of Nun. 
 I do not think it probable that a Hindu would make such a 
 mistake as to take the phonetic value of Nun to be equivalent 
 to his lingual nasal, which sounds very differently, and I fail 
 to see, except on the impossible theory of Greek mediation, 
 how the double horizontal line of na can be explained by means 
 of a real Semitic form. It seems to me that the Hindus, owing 
 to their aversion to heavy tops, first turned the Nun topsy- 
 turvy and used for some time a figure like that drawn in 
 Col. IV. 3 As a proof for this assertion I would cite the palatal 
 
 1 The third Pehlevl Mem of Col. 59 of the same Table, which exactly re- 
 sembles the Brahma ma, may be pointed out as an interesting analogous 
 development. 
 
 2 I must add that Professor Weber admits the possibility of deriving the 
 dental na from Nun, but thinks that the na comes nearer. M. Halevy 
 connects na immediately with the Greek vu, which, as he believes, was 
 laid on its side. 
 
 3 A Nun with a small open square at the top, instead of an angle, occurs 
 in the Siloah inscription and again much later, see Prof. Euting's Tabula 
 Scr. Aram. Cols. 3a-b, 8b, and 13 a. As the change from an angle to 
 an open square is in keeping with the stiff formalism of the Indian al- 
 phabet, it is not necessary to assume that the Semitic prototype showed 
 the latter. The Indian form is probably nothing but an analogous de- 
 velopment. 
 
 B u h 1 e r , Indian Studies. Ill . 5
 
 66 Indian Studies. No III. 
 
 fia (Col. VI, a), which consists of the hypothetical form, turned 
 from right to left, and of a small bar at the top, indicating that 
 it belongs to a different class or Varga. 
 
 Later only, I think, the Hindus substituted as a cursive 
 development for the hook at the side, the single bar which 
 protrudes on either side of the foot of the vertical. 
 
 No 15. Professor Weber doubtingly proposes to connect 
 the Brahma sa (Col. VI, d) with the Semitic Samech, and his 
 idea has been taken up by Dr. Taylor who derives it from 
 the very dissimilar Sabaean form y. The new Bhattiprolu s, 
 Col. V (the origin of which has been wrongly explained in my 
 introductory remarks to the edition of the inscriptions) and the 
 sa of the Eran coin (Col. VI, a) as well as the sa of the 
 Ghasundi slab (see below), I think, permit me to say that 
 Professor Weber's conjecture is well founded. I take the Bhatti- 
 prolu s to be derived from a Samech like that in Col. I, b, 
 the two [64] upper bars of which were replaced by a curve 
 (as in the hypothetical sign in Col. IV) and which was then 
 turned topsy-turvy in accordance with the Indian principle, 
 appealed to so often. This sign probably served in the begin- 
 ning to express both sa and sa. Later two separate signs were 
 developed out of this original representative for Samech, (1) 
 the sign for sa with the original cross bar placed at the outside 
 of the curve, which appears in its original position on the Eran 
 coin (Col. VI, a) and turned from the right to the left (Col. 
 VI, b) in the Southern versions of the Edicts (Grirnar and Sid- 
 dapur) and in Bhattiprolu; (2) the sign for sa (Col. VI, c), 
 with the original cross - bar shortened and placed inside the 
 curve, which hitherto is traceable only on the Ghasundi slab, 1 
 but must have been used extensively, as it is evidently the 
 parent of the square sa of the later Sanskrit inscriptions. 2 It 
 occurs neither in the Bhattiprolu alphabet, which retained the 
 old sign, in order to denote s, nor in the ordinary Brahml lipi 
 
 1 See Journ. As. Soc. Beng., Vol. LVI, Plate at p. 79, where it occurs in 
 the name Samkar$ana. 
 
 2 The square form occurs first in the Mora Well inscription from the times 
 of Rajuvula's son (Sodasa), Arch. Surv. Rep., Vol. XX, Plate V, 4, which 
 possibly belongs to the first century A. D.
 
 IV. The Derivation of the Brahml. 67 
 
 of the Edicts. The northern variety of the latter developed a 
 new sa (Col. VI, d) with 'a curve instead of the straight limb 
 at the side, and out of this the sign with two curves, one below 
 the other (Col. VI, e), which occurs in the Kalsl version, in 
 Dasaratha's Nagarjum cave inscriptions and so forth, and which 
 probably is also intended for sa. 
 
 It would, of course, be possible to connect the Bhatti- 
 prolu s with the cursive Aramaic Samech (Col. Ill) of the sixth 
 century B. C. But then the cross bar of the Bhattiprolu sign 
 would remain unexplained. Moreover, there would be the in- 
 superable difficulty that the Aramaic Beth, Daleth, Waw, Cheth, 
 Ain, Qoph and Eesh of the sixth century are so strongly mo- 
 dified that they cannot be considered the prototypes of the corre- 
 sponding Brahma ba, dha, va, gha, E, kha and ra. It seems, 
 therefore, advisable to assume, as must be done also in the case 
 of sa, that the Hindus independently changed the form of the 
 ancient sign, but in a manner analogous to that adopted by the 
 Aramaeans. 
 
 [65] The derivation of the signs for sa and sa from the same 
 original letter points to the influence of the Sanskrit grammar- 
 ians or phoneticists, who are well aware of the intimate con- 
 nection of the two sounds and duly teach that Sanskrit sa 
 becomes sa through the influence of a preceding i, u, r, e, ai, 
 o, au, k } T or I. 
 
 No 16. Professor Weber's ingenious identification of the an- 
 cient Ain (Col. I and II) with the Brahma E, which is supported 
 by the analogous proceeding of the Greeks, who also used it to 
 express the gutturo-labial vowel, and of the modern Jews, 1 who 
 express the e of foreign words by Ain, receives a strong 
 confirmation through the round and half-round signs for E in 
 Col. IV and Col. V, a b. The first occurs once in the word esa, 
 Kalsl, Edict XIII, 1, 1. 37. The same version offers also several 
 times the very similar form 2 of Col. V, b. The first sign in Col. V, 
 
 1 See Grierson, Irid. Ant., vol. XXIV, p. 248, Note 5. The cause of this 
 substitution is probably the resemblance of the Hebrew p to a cursive 
 e, torned towards the left. 
 
 2 It seems probable, that this or a similar half round form is the parent 
 of the southern E, which in the Pallava and Vengi inscriptions looks 
 like ca, td and later becomes I/, closely resembling the ancient la. 
 
 5*
 
 68 Indian Studies. No III. 
 
 which looks like a dha is found in the word etina of the 
 Hathigumpha inscription of Kharavela, 1. 8, and in the Sanchi 
 inscription No 173 (Epigraphia Indica Vol. II, p. 375) where, 
 not thinking of the Hathigumpha letter, I unfortunately have 
 read Dharakina for Erakind. Professor Weber's conjecture 
 has been accepted by Dr. Taylor alone, Dr. Deecke and M. 
 Halevy, trying to connect the triangular E with Aleph. There 
 was however a very good reason for giving up the round E. 
 For it could not have been distinguished from the lingual tha, 
 which the Hindus developed out of the corresponding dental. 
 In my opinion the triangular E is a development, formed by 
 the Hindus independently, and the angular forms for Ain in 
 the later Semitic alphabets, Euting op. cit., Col. 8b, 14 ff., are 
 merely analogous, showing how easily a circular letter may be 
 converted into a triangle or rhombus. It is just possible that 
 phonetic considerations may have induced the Indians to select 
 Ain for expressing their e. For according to a description of 
 the Indian e-sound, which seems to go back to Sakatayana 
 and has been generally accepted by the later native gram- 
 marians, 1 it is half guttural and half palatal (kanthya-talavya). 
 As the Ain no doubt was a guttural, it may have appeared 
 more suitable for E than for / or U. 
 
 No 17. The fact that the Brahma pa is the old Semitic 
 Phe turned topsy-turvy, has been acknowledged by everybody. 
 The new form of the Eran coin, Col. IV, shows it in its orig- 
 inal position. 
 
 No 18. Regarding the Brahma equivalent of the Semitic 
 Tsade I differ from all my predecessors. I believe that it was 
 [66] used, as one would expect from its phonetic value, for the 
 formation of the letter ca, the sound of which is and always 
 has been, not the English tsha, but tsa almost like tya. The Se- 
 mitic sign, of course, had to be turned topsy-turvy on account 
 of its heavy top, and the small bar running to the right was 
 turned towards the vertical. This process gave first a tailed 
 \s and later with the direction to the left, if. The angular 
 shape of the letter has been well preserved in the ca of the 
 Edicts, Col. V, a, which is not uncommon, and the tail appears 
 
 1 See Weber, Indische Studien, vol. IV, p. 119 f.
 
 IV. The Derivation of the Brahml. 69 
 
 in the Bhattiprolu form, Col. V, c. 1 The round ca (Col. V, b) 
 is purely cursive, and not the parent of the later Indian letters, 
 which mostly go back to angular or pointed forms. 
 
 No 19. With respect to the Semitic Qoph I must like- 
 wise differ from all my predecessors, who have compared it to 
 the Brahma cha, which is clearly an Indian derivative from ca 
 and has a very different phonetic value. In my opinion the 
 Semitic Qoph has its counterpart in the Brahma kha, the oldest 
 form of which (Col. V, a) consists of a circle with a super- 
 imposed vertical line ending in a curve. This is the oldest 
 Semitic sign (Cols. I and II), 2 turned topsy-turvy on account 
 of its heavy top. The curve at the end of the vertical has 
 no doubt been added in order to distinguish the letter from 
 va. The kha with the circle at the base occurs sometimes in 
 Jaugada and is used frequently in the Kals! version 3 of the 
 Edicts, where not rarely the circle is replaced by a somewhat 
 irregular loop, sometimes attached to the right of the vertical 
 line (see the kha figured above p. 41). It is the parent of 
 most of the later Indian signs for kha, including the modern 
 Devanagarl form, which all show a loop or a triangle at 
 the base of the vertical. In several versions of the Pillar 
 [67] Edicts and elsewhere a well developed dot takes the place of 
 the circle. This change is analogous to that pointed out above 
 in connexion with the second and third forms of the Brahma 
 ja, and likewise indicates the use of pen and ink for the time 
 when it was made. There is also a third form of the kha 
 which consists merely of a vertical with a hook at the top. It 
 is chiefly used in the Southern versions of the Edicts, parti- 
 cularly in Girnar, as well as in Bhattiprolu and later in the 
 cave inscriptions, and is evidently a cursive development. 
 
 1 The later Semitic alphabets furnish various analogous developments, see 
 Prof. Euting's Tabula, Col. 9 (third sign), Col. 15 (first sign), Col. 42 
 (second sign), differing from the Brahma letter only by the position of 
 the angular or round appendage, which of course remains at the top of 
 the character and to the right of the vertical stroke. 
 
 2 Still more closely agreeing Phoenician and Mesa forms, in which the 
 vertical is not drawn across the round head, are given in Prof. Euting's 
 Tabula, Col. Id, Col. 2 a. 
 
 3 The form given in the plate has been taken from the word likhite, Kalsl, 
 Edict IV, 1. 12 (end).
 
 70 Indian Studies. No III. 
 
 In favour of this identification speaks also the fact that 
 the modern Indians in loan words not rarely express the 
 Arabic Qoph by their Ma. Thus the Marathl has f^f, f^WT, 
 *?WT*r 7 IT^TW, *JHT and so forth for Arabic words with Qoph. 
 This substitution shows that the Indian ear detects an affinity 
 between the deep guttural Qoph and the harsh sounding kha. 
 
 No 20. I can only agree with Professor Weber's identi- 
 fication of Resh with the Brahma ra or repha. But the orig- 
 inal ra has not the form, consisting of a straight stroke, which 
 is given in his table and in that of Dr. Taylor. The straight 
 ra, evidently a late cursive development, is very rare in the 
 Edicts. I know only of one instance in the Rupnath Edict in which 
 a nearly straight stroke occurs. In the Girnar version, where ra 
 is very common, it has at the top invariably one or two little 
 angles, open to the right, or instead of the angle an irregular 
 bulge towards the left, see Col. V, a b. I take the form 
 with the single angle to be the oldest and consider it to be 
 derived from a Semitic form like that of the Mesa inscription, 
 the triangle of which the Hindus opened, in order to avoid 
 the heavy top, by attaching the vertical to the lower side of 
 the base i. e. by putting ^ for ^. The signs with two or 
 more angles, Col. V, d e, are no doubt artificial, ornamental 
 developments. In the new Siddapur Edicts this development has 
 been carried to an extreme, and the whole letter has been con- 
 verted into a wavy line, consisting of four or five little 
 angles. 
 
 No 21. Professor Weber's identification of Shin with the 
 palatal sibilant sa, which has been accepted by Dr. Deecke, 
 seems also to me self-evident. It is only necessary to substi- 
 tute for the late Indian form (given in Professor Weber's 
 Col. II) the formerly not accessible, real old Brahma signs, 
 Col. V, a c, which are found in the Kalsl, Rupnath and Sid- 
 dapur versions of the Edicts, on the Bhattiprolu prism, on the 
 Ghasundi slab and in the Pabhosa cave inscriptions. 1 I con- 
 
 1 I enumerate the occurrences so fully, because, when the sign first turned 
 up in Kalsi, it was considered to have been borrowed from the Kharo- 
 sthi, which has the remotely similar lingual .yo *P. I will add that in 
 Bhattiprolu where we have samanude&anam i. e. 6ramanoddeh/anam, in 
 the Ghasundi inscription where we have Para&ariputa and Slid, and in
 
 IV. The Derivation of the Brahml. 71 
 
 [68] sider as their prototype the oldest Semitic form with two 
 angles (Cols. I and II), not the very similar Aramaic Shin (Col. Ill) 
 of the sixth century. For in accordance with the principles 
 of the Brahml lipi, the Semitic letter had to be turned topsy- 
 turvy, and the double angle at the top had to be got rid of, 
 which latter change could be most easily effected by placing 
 the one angle inside the other. The Hindus may be well 
 credited with the independent invention of this modification, 
 as the later Phoenicians and the Ethiopians have likewise in- 
 troduced it independently. Moreover, the same considerations, 
 which make it impossible to accept the Aramaic Samech as 
 the immediate source of the Bhattiprolu s (see above p. 67), 
 speak in this case against the assumption that the Semitic al- 
 phabet, adopted by the Hindus, contained the Aramaic Shin. 
 Finally the Western letter-numeral su for 100, still shows the 
 two angles quite plainly, see App. II, pi. Ill and the Grundriss, 
 Taf. IX, Cols. IV V, XI. 
 
 No 22. The oldest form of the Brahma ta, which Professor 
 Weber has recognised already as a modification of the Semitic 
 Taw, seems to be that (Col. V, a) consisting of a long slightly 
 inclined downstroke with a short straight bar slanting off to 
 the right. Next comes the sign, given in Col. V, b, with the 
 bar slanting off to the left, and this is due to the change in 
 the direction of the writing. The ta consisting of a vertical 
 stroke with an angle at the foot, Col. V, c, which is frequent 
 in the southern and western versions of the Edicts and also on 
 the Sohgaura Plate, is probably [69] due to the formalism of the 
 
 Pabhosa which offers feonakayana i. e. ftaunakayana, this sign appears 
 only in words which have &a in Sanskrit, while in the Edicts it occurs 
 mostly and very irregularly in words, where the Sanskrit has tta or fa. 
 The confusion is owing to the negligent pronunciation and writing of 
 the clerk, who made the copy. His dialect probably contained two 
 sibilants, the dental and the palatal, and his alphabet had three, the 
 dental, the palatal and the lingual. His negligence in pronunciation 
 made him pronounce sa for Sa and vice versa, and his negligence in 
 writing made him use sa and So, indiscriminately. As stated above, 
 papers written in our times by half-educated clerks and others show ex- 
 actly the same confusion and for the same reasons. They pronounce 
 e. g. both sac and sac (satyam) and they write indiscriminately ^^, 
 H^ and ^^(.
 
 72 Indian Studies. No III. 
 
 Hindus, their desire to set the letter up straight and to make 
 it perfectly regular in appearance. The forms with a round 
 limb to the right or left of the vertical (see above p. 41) of 
 course are cursive. The oldest among these five forms for ta 
 closely resembles the Taw from the earliest Sinjirli inscription 
 (Col. Ill, b), which belongs to the beginning of the eighth 
 century B. C., and still more that of Col. Ill, a, which occurs 
 on an Assyrian weight of the reign of Salmanassar 1 and has 
 therefore been engraved before 725 B. C. As it thus appears 
 that Semitic forms, consisting of a long inclined downstroke 
 Avith a crossbar very slightly protruding on the left or with a 
 simple bar on the right, have been developed in very early 
 times, it is not improbable, that one of them occurred in the 
 alphabet which the Hindus borrowed. But the possibility that 
 the Indian sign is an independent development from the straight 
 Phoenician cross (Col. I), is not altogether excluded. For the 
 cross could not remain in the Brahma alphabet, because it 
 would have been undistinguishable from ka. 
 
 B. The Derivative Consonants and Initial Vowels. 
 
 The contrivances, by which the derivative signs, both 
 primary and secondary, for consonants and initial vowels have 
 been formed, are: 
 
 (1) the transposition of one of the elements of a phonetic- 
 ally cognate borrowed sign, 
 
 (2) the mutilation of a borrowed letter or of another 
 derivative sign of similar phonetic value, 
 
 (3) the addition of straight lines, curves or hooks to bor- 
 rowed or derivative signs. If a hook is added the original sign 
 suffers a slight mutilation. 
 
 Two cases of transposition have already been mentioned 
 above under No 15 ; where it has been pointed out that the 
 sa and the Ghasundi sa come from the Bhattiprolu s. A 
 second case occurs, as Professor Weber has been the first to 
 recognise, in the dental da (No 4, Col. VI, a b, f g). The 
 
 As Professor Euting kindly points out to me, the inscription is found 
 Corp. Inscr. Sem., P. II, No 2c.
 
 IV. The Derivation of the Brahml. 73 
 
 [70] two first forms, one of which occurs in Bhattiprolu as 
 well as on a Patna seal l and the other in the majority of the 
 Edicts, are derived from the two dha in Col. V, by dividing 
 the straight line and pushing the halves back to the ends of 
 the semicircle which remained. In the third actual form 
 (Col. VI, f) and the hypothetical one, Col. VI, g (required on 
 account of the next following sign) the semicircle has been 
 converted into a small square, left open on one side, and this 
 change is due to the liking for angular forms, mentioned 
 above p. 37. 
 
 The cases in which a borrowed or derivative sign has been 
 mutilated, are' those of the lingual tennis, tennis aspirata and 
 media, all of which Professor Weber has already explained cor- 
 rectly. The lingual da under No 4, Col. VI, c, which occurs 
 occasionally in Kalsl and commonly in the later inscriptions of the 
 Nanaghat and the other Western caves etc., is derived, by the 
 removal of the lower end, from the Bhattiprolu da (Col. VI, a), 
 which had not yet been turned from the right to the left. The 
 angular da (Col. VI, h) comes in like manner from an ang- 
 ular da (not yet turned to the left), of which the known in- 
 scriptions do not offer any example. The tha (No 9, Col. VI, a) 
 is of course the dental tha minus the central dot, and the ta 
 (No 9, Col. VI, b) has probably been obtained by halving the 
 tha, as Professor Weber conjectures. To a Hindu this process 
 probably appeared very natural. For he formed several aspi- 
 rates by adding curves. Hence he may be supposed to have 
 considered a round sign, denoting an aspirate, as equivalent to 
 an unaspirated letter plus a curve of aspiration. Thus the di- 
 vision of the sign would be quite legitimate. In the Edicts 
 both tha and ta are frequently made smaller than the other 
 letters. 
 
 Two other cases of the mutilation of borrowed letters 
 occur in the signs for initial /and U. It has been recognised 
 already by Mr. Prinsep that the three dots of / (No 16, Col. VI, 
 a c, lower row) indicate the three corners of the triangular 
 E\ and this view, which has been generally accepted, is con- 
 firmed by the fact, that in the Edicts the position of the dots 
 
 Cunningham, Arch. Survey Rep., vol. XV, pi. Ill (first figure).
 
 74 Indian Studies. No III. 
 
 of / generally agrees with the position of the angles of E. To 
 a Hindu phoneticist or grammarian the derivation of / from E 
 would appear a matter of course, because E is very commonly 
 [71] the representative of an Jin strong forms or its Guna. Hence 
 he expressed the latter by a lighter form of the former, just 
 marking the corners of the triangle. 
 
 The case of U (No 6, Col. VI, a) is somewhat different. It 
 has been customary to derive the sign directly from one of the 
 later forms of the Semitic Waic. Considering the facts, con- 
 nected with the linguals and with initial I, I would propose to 
 derive it either directly from the old Semitic Waw, turned topsy- 
 turvy, or from the Brahma va by a bisection of the circle at the 
 foot of the sign and the substitution of a straight line for the 
 irregular pendant, which remained. The several steps were, 
 therefore, ^ or J, ), J, and finally 1 L. It is probable that 
 phonetic considerations, the observation of the frequent substi- 
 tution of n for va in weak verbal and nominal forms, as in 
 uliyate, upyate, supyate, upta, supta, usita, from vah, vap, 
 svap and vas, may have led to the adoption of this pro- 
 ceeding. The framers of the Brahma alphabet were, as pointed 
 out already, Brahmans, acquainted with phonetics and with 
 grammatical theories. 
 
 The last case of mutilation, I believe, occurs in the Anu- 
 svara (No 13, Col. VI, b) which is represented by a simple dot. 
 This is no doubt a substitute for the small circle (Col. VI, a), 
 which reappears very commonly in certain later epigraphic and 
 literary alphabets. I consider the circle to be a mutilated minute 
 ma, the upper angle of which was omitted, 2 and I believe that 
 this conjecture is supported by the following facts. In the Edicts 
 and all other ancient inscriptions the Anusvara is placed rarely 
 at the top of the letter after which it is pronounced. Usually 
 
 1 The former actual existence of the third form is proved by the Jaugada 
 (No 6, Col. VI, f). In the second century A. D. and later, the U fre- 
 quently shows a curve at the foot instead of the straight line, ^ and 
 it is possible that this may be an ancient survival, not, as is usually 
 assumed, a modern development. 
 
 2 I must acknowledge that M. Halevy's ingenious, but erroneous, conjec- 
 ture, according to which the Brahma Anusvara is derived from the Kha- 
 rosthi ma, induced me to frame my theory.
 
 IV. The Derivation of the Brahml. 75 
 
 and regularly it stands opposite the middle, but in a number 
 of cases it is found also to the right of the foot. Now vowelless 
 consonants always stand at the foot of the preceding sign in 
 the oldest Sanskrit or Mixed inscriptions from the times of the 
 [72] Kusana kings down to the fifth century A. D., when the Vi- 
 rama makes first its appearance. Moreover all such vowelless 
 consonants are made exceedingly small, even after the inven- 
 tion of the Virama, and they are very commonly mutilated at 
 the top. 1 These facts would fully explain the use of a small 
 circle for a vowelless ma of the Asoka type, which then be- 
 came the general sign of nasalisation in the Brahma alphabet, 
 just as the Kharosthl ma was turned into the Kharo.sthl Anus- 
 vara. My theory, of course, rests on the assumption that the 
 Brahma alphabet was used from the beginning, not for Pra- 
 krit, but for Sanskrit, and this is made more than probable 
 by the occurrence of na in the Mahabodhi Gaya alphabet of 
 the masons, as well as by the arrangement of its letters on 
 phonetic principles, see above p. 33, likewise by the numerous 
 indications that the alphabet was elaborated by phoneticists or 
 grammarians, i. e., by Brahman schoolmen. 
 
 Short straight strokes marking the additional matra, are 
 added (originally on the left 2 ) to the vowel-signs for A and U 
 in order to frame the signs for the long vowels A (No 1, Col. VI) 
 and U (No 6, Col. VI, d). In long / an additional dot appears 
 instead of the stroke which would not have agreed with the 
 character of the sign. 3 
 
 Added to vowel-signs (originally on the right), short strokes 
 indicate a change of the quality of the sound. This is the way 
 
 1 See e. g. Epigr. Ind. II, p. 208, Mathura Inscr., New Ser., No 27, Dr. Fleet's 
 Gupta Inscr., Nos 3, 6, 11, etc., Dr. Hoernle's Bower MS. passim. In the 
 first mentioned inscription the m of siddkam is not much more than a 
 triangle, in the Gupta inscriptions and the Bower MS. m is regul- 
 arly U. 
 
 2 In the actually existing signs they appear on the right, because the 
 signs were turned on the change in the direction of the writing. 
 
 3 Dots appear for short horizontal bars also in other cases, e. g. in the 
 hyphens at the end of verses, which often look like, and have been 
 misread, as Visargas. Vice versa small horizontal strokes are substituted 
 for dots e. g. in the letter /, which in the inscriptions of the Nasik and 
 Karle caves sometimes consist of three short horizontal bars.
 
 76 Indian Studies. No III. 
 
 in which O, 1 No 6, Col. VI, f g, has been formed out of U, 
 and in the first sign (Col. VI, f), the stroke stands in its 
 [73] original position, because the letter (Jaugada form) has not 
 been turned round. The AI (No 16, Col. VI, b, upper row) 
 appears to have been derived from E in the same manner, but 
 the letter has been turned from the right to the left. The 
 sounds u and o, as well as e and ai, appear to a Hindu and to 
 a Sanskritist closely connected, because in numberless cases o 
 is the Guna or representative of u in strong forms, and because 
 e and ai both appear in the strong forms of roots with i and in 
 derivatives from nouns with i and e. These phonetic or gram- 
 matical affinities no doubt influenced the formation of the signs. 
 
 Added to signs for consonants, either on the right or 
 across the top, a straight stroke likewise denotes a change of 
 quality viz. that the sign expresses the corresponding sound of 
 a different class or Varga. The stroke has its original position 
 in the Bhattiprolu I (No 12, Col. VI), which has not been 
 turned round. It appears on the left in the palatal ?1a, 2 because 
 this has been turned. Its position is again the original one in 
 the guttural iia, Avhich in the Gaya alphabet and the Kusana 
 (see above p. 31) inscriptions is P*, and in which the foot of the 
 na has also been modified. Finally in na (No 14, Col. VI, b) 
 the bar goes across the vertical. The peculiarities of the last 
 two letters are probably due to a desire to avoid collisions 
 with na, ne and 0. 
 
 Aspiration is expressed by the addition of a curve in the 
 Bhattiprolu gha (No 3, Col. VI), and the ordinary Brahma signs 
 for dha (No 4, Col. VI, d), pha (No 17, Col. VI) and cha (No 18, 
 Col. VI, a b) are derived in the same manner from da, pa and 
 CM. In the sign for cha both ends of the curve have been con- 
 nected with the vertical line of ca. There are numerous in- 
 stances, like that given in Col. VI, a, in which a difference 
 between the two halves of this letter is clearly discernible, and 
 the one half is angular, the other round. These, I think, are the 
 older forms. The second sign for cha (Col. VI, b) which consists 
 
 1 The sign -ft-, which is sometimes given for O in palaeographic works, 
 does not exist. 
 
 2 See above the remarks under No 14, p. 65.
 
 IV. The Derivation of the Brahml. 77 
 
 of a circle bisected by a vertical line is in my opinion 
 cursive. 
 
 In the two signs for bha (No 2, Col. VI) and for jha 
 (No 1, Col. VI) an angle or hook serves the same purpose as 
 the curve of the other four aspirates, and in both cases the 
 original sign is mutilated in order to make the new form less 
 cumbersome. The ba has lost its base line and the ja its two 
 [74] bars. It seems not improbable, as has already been suggested 
 by others, that the angle or hook of aspiration may be a cur- 
 sive development from the letter ha. 
 
 In the lingual la, derived from the round da (No 4, 
 Col. VI, e) a small semicircle * has been added to the foot of the 
 original sign in order to indicate the change of the phonetic 
 value. Here also, I believe, we may recognise the influence 
 of the grammarians or phoneticists. For the sounds da and la 
 are frequently interchanged in the same word. Thus we find 
 already in the Vedas regularly a la for a da between two 
 vowels, as in lie for ide. In the later Sanskrit and in the Pra- 
 krits there are numerous variants like nadi and nail, nadikera 
 and nalikera and so forth, where it is often difficult to decide, 
 which is the original form. As the principle, on which the ordi- 
 nary la has been formed, differs from the more general one, 
 applied in the case of the Bhattiprolu letter, I consider it to be 
 of later origin. 
 
 C. The Medial Vowels. 
 
 Hitherto two systems for the notation of the medial vo- 
 wels have become known, that of the Edicts and all the later 
 Sanskrit and Prakrit inscriptions and that of Bhattiprolu. The 
 first, which is by far the older one, shows clearly the influence 
 of the grammarians and their ingenuity. As the vowel, ex- 
 pressed in the beginning of words by the representative of the 
 Semitic Aleph, occurs in Sanskrit nearly as frequently as all the 
 
 1 Possibly a sign which occurs in Jaugada and in the Pillar Edicts |J f. e. 
 da with a dot a the foot, may have the same value, as it is used in 
 words which have double forms with da and la or la e. g. dudl or duli, 
 edaka or elaka, Coda and Cola
 
 78 Indian Studies. No III. 
 
 other vowels taken together, 1 it was not expressed by any sign, 
 but considered to be inherent in all signs for consonants. This 
 device fully agrees with the system of neai'ly all the phonetic 
 and grammatical treatises, which, as Professor Max Mtiller has 
 shown so clearly, do not refer to written letters but only to 
 sounds. 2 They almost invariably speak of the kakara, gakara 
 [75] and so forth. The commentators no doubt are right, when 
 they assert, that the vowel has been added in order to make the 
 pronunciation of the consonants possible, and the vowel a was 
 selected for this purpose on account of the frequency of its 
 occurrence. It seems impossible to assume that there is no 
 connexion between the two facts, and, as the grammarians 
 base their theories on spoken words not on written texts, I 
 think that they are the men who also in this case influenced 
 the formation of the Brahma alphabet. 
 
 As regards the other vowels, medial a is expressed by 
 placing to the right of the consonant the same short horizontal 
 stroke, 3 used for the differentiation of the initial A from A, ap- 
 parently because the other portion, the short a, is already 
 contained in the consonant. The remaining ones are expressed 
 by the signs for the initial vowels or by modifications thereof, 
 
 1 See Professor Whitney's calculations in his Sanskrit Grammar, p. 73 
 (second edition). 
 
 2 History Anc. Sansk. Lit., p. 507 ff. This assertion has been hotly con- 
 tested by Professor Goldstiicker in his Introduction to the Manava Kalpa 
 Sutra, p. 13ff. But Professor Kielhorn, who has studied the Vyakarana 
 during so many years, informs me that he does not know of a single 
 passage even in the Bhasya, which indicates with certainty, that a 
 written text of Paniiii's grammar is referred to, or where the technical 
 terms of the grammarians and their theoretical speculations refer to 
 written signs; see also his remarks in Gurupujakaumudi p. 29 ff. It 
 seems, therefore, to be a fact that the grammatical and phonetic rese- 
 arches were begun either before the introduction of writing or indepen- 
 dently of writing, and that even those ancient authors, who like Panini, 
 mention alphabets and clerks, continued to work on in the old manner. 
 I must, however, add that Geheimrath von Bohtlingk dissents and believes 
 that the phrase yathanyasam in the Mahabhasya (e. g. I, 22, 1. 3) points 
 to a written text. 
 
 3 Originally the stroke, of course, stood on the left, and it is found in 
 this position on the Eran coin, where the letters run from the right to 
 the left, see the Grundriss, Tafel II. 28. I.
 
 IV. The Derivation of the Brahml. 79 
 
 placed above or below the consonants. A very clear case is that 
 of the medial o in the syllable ko, 1 given in two forms under 
 No 6, Col. VI, h i. If the k, i. e. the portion of the sign 
 below the second bar, is removed, there remains in the eighth 
 sign a minute initial o of the type in Col. VI, f, and in the 
 ninth one of the type in Col. VI, g. Now in the Jaugada 
 version of the Rock Edicts, where the initial has the top bar 
 to the right, the same is invariably the case with the medial o. 
 It, therefore, would seem that the writer was perfectly aware 
 of the connection of the two signs. But, in Asoka's time this 
 [76] feeling was dying out. For in Girnfir, where the initial 
 with the top-bar to the left alone is used, the medial o is made 
 in both ways, and in the second part of the Delhi Pillar Edict 
 VII, 2, 1. 2 we have once the cursive medial o in nigohani, where 
 both strokes are placed on the same level above the consonant. 
 
 Equally clear is the case of the medial u, which is the initial 
 U, put below the consonant. This is distinctly recognisable in the 
 sign dhu (No 6, Col. VI, b) which occurs repeatedly in the Kals! ver- 
 sion. 2 Cursively it assumes the form, given in Col. VI, c or of p, with 
 the omission either of the vertical stroke or of the horizontal bar at 
 the end. On the same cursive principle u, which in ku etc. retains 
 the shape of 7(Grundriss, 1, 11, p. 37), is expressed by two strokes, 
 placed either horizontally at the side of the consonant (No 6, Col. VI, 
 e) or below the consonant, where they frequently form an acute 
 angle but are also placed parallel side by side. These facts seem to 
 indicate that Asoka's clerks had lost the remembrance of the origin 
 of the signs for medial u and w, and that they considered the 
 old forms, which they occasionally used, merely as permissible 
 variants without any special significance. In later inscriptions, 3 
 however, reminiscences of the origin of the subscript u and ii are 
 found. Thus in the ancient arada characters of the Baijnath 
 Prasasti and elsewhere pu is expressed by ^. 
 
 As regards the medial i, the small angle to the left of 
 the top of the consonant by which it is marked, seems to be 
 
 1 Compare also the (jo ~fc in mago. Qiruar, Ed. I. 1. 11, where a distinct 
 stands above ga. 
 
 2 E. g. in Kaisi, Edict III, 1. 8 (twice). 
 
 8 Compare also the Grundriss, gu in Taf. VII. 10. 20; bhu in Taf. VII. 30. 
 XII; tu in Taf. VII. 3(5. XXI; and bhu in Taf. VII. 30. XX-XXI.
 
 80 Indian Studies. No III. 
 
 the result of a connexion of the three dots of the initial vowel 
 by means of two lines, see the ki No 16, Col. VI, d e, lower 
 row. As long as the Brahma alphabet was written from the 
 right to the left, the ^-strokes, as well as all other vowel signs 
 of course stood to the left of the consonant. In the Kals! ver- 
 sion, Ed. XIII, 2, 1. 10, there are two signs, at the end of the 
 words anuvidhiyamti and anuvidhiyisamti, where the vowel 
 has this position. They look like A, because the vowel strokes 
 have been added to the middle of the consonant. A better 
 formed ti with the vowel on the left occurs in Allahabad, 
 Ed. I (end), and a similar hi on the Sohgaura copperplate, 1. 4. 
 The medial l does not seem to be connected with the initial I, 
 but to have been formed by the addition of the [77] straight 
 stroke, denoting also in other cases the lengthening of vowels, 
 which for convenience's sake in this case was made vertical. 
 Cursively the angle of medial i is converted in Girnar into a 
 shallow curve and medial I is expressed by a curve bisected 
 by a vertical stroke. 
 
 The sign for medial e, a straight or slanting line to the 
 left of the consonant, I take to be the remnant of a triangular 
 initial E, the outlines of which have been indicated by dots 
 in the ke, figured under No 16, Col. VI, a, upper row. I may 
 add that in the Edicts forms like \ are occasionally used 
 for ge, where the vowel sign seems to consist not of a line, 
 but of a hook put on the top of the letter. Such forms are 
 perhaps ancient survivals, dating from a time when the vowel 
 was represented by an angle, to which the triangle probably 
 was reduced in the first instance. The position of the 
 e-stroke is abnormal, as it stands to the left of, or before, the 
 letter, after which it is pronounced. The cause is no doubt, 
 that, if it had been placed to the right, it would have been 
 undistinguishable from medial long a. 
 
 In accordance with the form of the initial AI consisting 
 of e and a horizontal bar to the left of the top, the medial ai 
 is expressed by two bars to the left of the consonant, see the 
 syllable thai from the Girnar version, given under No 16, 
 Col. VI, c, upper row. 
 
 The absence of a medial vowel between two consonants 
 is expressed by the formation of a ligature, in which ordinarily
 
 IV. The Derivation of the Brahml. 81 
 
 the second consonant or its most essential portion is attached 
 to the foot of the first. In the Girnar and Siddapur versions 
 however, the ligatures sta, tpa, vya 1 as well as those contain- 
 ing a ra, like tra, pro, and vra show the inverse order and 
 are spelt tsa, pta, yva, rta and so forth, while all the others 
 like sta, mha etc. are formed regularly. Moreover, in the words 
 bramhana (Girnar IV, 1. 2) and drahyitavyarp. (Siddapur, 1, 1. 9) the 
 first vertical of ba and the two vertical strokes of da have 
 been converted into wavy ra-lines. These irregularities are no 
 [78] doubt due to an artistic feeling and the desire of the clerks to 
 produce regular, shapely signs. The combinations of the conson- 
 ants with medial vowels furnish analogies. Properly all the 
 vowels ought to stand to the right of the foot of the consonants. 
 If the majority is nevertheless placed at the top, that has been 
 done merely for the sake of convenience. Later inscriptions 
 also furnish a few isolated cases of an inversion of the order 
 of the elements of ligatures. Thus the name of Castana's fa- 
 ther is spelt on the coins Ysamotika instead of Syamotika and 
 Harsa's Banskhera Plate of A. D. 628/9 offers vannra for 
 varnna. 
 
 The cause of the formation of ligatures in order to ex- 
 press the absence of vowels must again be sought for in the 
 influence of the Sanskrit phoneticists on the development of 
 the Brahma alphabet. The Pratisakhyas and the later works 
 on phonetics and grammar all use the expression samyukta- 
 ksara "a conjunct syllable" for groups like kta, kra and so 
 forth. The combination of the signs in writing looks very 
 much like a practical illustration of the meaning of the term. 
 
 The manner in which the absence of a vowel after a final 
 consonant was probably expressed, has been stated in the re- 
 marks on the Anusvara, above p. 74 f. 
 
 The Bhattiprolu system of vowel-notation differs from the 
 ordinary one merely by marking the short a by the bar, which 
 denotes a in the Edicts, and the long a by the same bar plus 
 a vertical or slanting stroke, hanging down from it, e. g. Jt 
 
 1 At least in mjamjanato, Ed. Ill, 1. 6. The cases of the passive future 
 participles in tayva are doubtful, as they may have been pronounced as 
 they are written, compare Pali mayham and so forth. The Siddapur 
 version has tavya in drahyitavyam. 
 
 Buhler, Indian Studies. III.
 
 Indian Studies. No III. 
 
 na and _fc na. This system, according to which the conso- 
 nants have no inherent a, seems to have been invented in 
 order to avoid the necessity of forming the ligatures, which 
 make the ordinary Brahma alphabet cumbersome and difficult 
 to read in its later developments, and in order to express final 
 consonants more conveniently. I believe it, therefore, to be of 
 later origin, especially as the other Bhattiprolu vowel-signs do 
 not differ from the ordinary ones. The invention must, of course, 
 have been made for writing Sanskrit, as the Prakrits have 
 few groups of dissimilar medial consonants and no final ones. 
 To sum up the forty four letters 1 of the ordinary Brah- 
 ma alphabet, traceable in the oldest inscriptions as well as the 
 variants of the Bhattiprolu in scrip- [79] tions, contain represen- 
 tatives of, and derivatives from, all the twenty two Semitic 
 characters, viz: 
 
 Semitic 
 letters 
 AUph 
 Beth 
 Gimel 
 
 Daleth 
 
 He 
 
 Waw 
 
 Zain 
 
 Cheth 
 
 Theth 
 
 Yod 
 
 Kaph 
 
 Brahma letters 
 
 a (initial) 
 
 ba 
 
 ga 
 
 dha 
 ha 
 
 ja 
 
 gha 
 
 tha 
 
 ya 
 
 ka 
 
 Derivatives 
 
 a (initial and medial) 
 
 bha 
 
 gh (Bhattiprolu) 
 
 da, da 
 
 I dha 
 \la 
 
 u { (initial and medial) 
 jha 
 
 tha, ta 
 
 With respect to the missing A U and (A)fy it may be noted, that the 
 forms of AUin the Gupta and Pallava inscriptions, as well as those in 
 the Bower MS , show that it was derived from by the addition of a 
 bar to the right of the vertical. But the medial au in the Bharahut Torana 
 inscription (Grundriss, I, 11, Taf. II. 28. XVIII) makes it probable that there 
 was a second form with the additional bar to the left. Regarding the 
 origin of the Visarga, which first occurs in the Kusana inscriptions, of the 
 Jihvamullya and Upadhmanlya, first traceable in the Gupta inscriptions, 
 I am not able to suggest anything.
 
 Semitic 
 letters 
 Lamed 
 Mem 
 
 Nun 
 Samech 
 
 Ain 
 
 Phe 
 
 Tsade 
 
 Qoph 
 
 Resh 
 
 Shin 
 
 Taw 
 
 IV. The Derivation of the Brahml. 
 Brahma letters Derivatives 
 
 83 
 
 la 
 ma 
 
 I (Bhattiprolu) 
 m (Anusvara) 
 
 s (Bhattiprolu) 
 e (initial) 
 
 pa 
 
 ca 
 
 kha 
 
 ra 
 
 sa 
 
 ta 
 
 na 
 
 e (medial) ai (init. & med.) 
 i, I (initial & medial) 
 pha 
 cha 
 
 With the exception of the signs for the sibilants sa and 
 sa-sa, which in consequence of modifications, introduced, it 
 [80] would seem, independently by the Hindus, 1 resemble later 
 Aramaic characters, the Brahma letters closely agree with or 
 are most easily derivable from the old types of the North-Semitic 
 alphabet. And the Brahma initial vowels A and E as well as 
 the consonants kha, ga, gha, tha, dha, ba and va point to par- 
 ticularly archaic prototypes, while ha and ta appear to be con- 
 nected with somewhat modified forms. It would, therefore, seem 
 that the Semitic alphabet became known to the Hindus at a 
 period when the angle of its Aleph opened wide and the ver- 
 tical crossbar protruded about equally on the two sides, when 
 the top of Beth was still closed, when ga consisted of an angle 
 open below, when Daleth had not yet developed a tail, when 
 Waw consisted of a semicircular head with a vertical depending 
 from the middle, when Cheth had three bars, when Theth and 
 Ain were quite or nearly circular and Qoph had a round head 
 with a vertical hanging dowii from it, but when the simplified 
 He consisting only of three strokes had been developed and 
 the left half of the original crossbar of Taw had nearly or 
 quite disappeared. 
 
 1 See the remarks made above p. 67 under No 15 and p. 71 under No 21. 
 
 6*
 
 84 Indian Studies. No III. 
 
 According to the dates of the Semitic inscriptions, which 
 can come into question, those of Mesa's stone and the Assyrian 
 weights, this period must fall somewhere between circiter 890 
 and 750 B. C., probably more towards the lower than the re- 
 moter of the two limits. Hence the terminus a quo for the 
 introduction of the prototypes of the Brahma letters lies between 
 the beginning of the ninth century and the middle of the 
 eighth, or about 800 B. C. And it seems to me that some fur- 
 ther considerations make it probable that their actual impor- 
 tation took place at this early time. 
 
 As the Brahma ha goes back to a form of He, which is 
 not found in any Phoenician alphabet, but occurs on the As- 
 syrian weights, where also a Taw very similar to the Brahma 
 ta is found, the conjecture seems not altogether improbable 
 that the Semitic alphabet may have come to India through 
 Mesopotamia. And it would agree with such an assumption 
 that passages in ancient Indian works prove the early exist- 
 ence of a navigation of the Indian Ocean and the somewhat 
 [81] later occurrence of trading voyages, undertaken by Hindu 
 merchants to the shores of the Persian Gulf and its rivers. 
 
 The now well known Baveru Jataka, 1 to which Professor 
 Minayeff first drew attention, narrates that Hindu merchants 
 exported peacocks to Baveru. The identification of Baveru 
 with Babiru or Babylon is not doubtful, and according to what 
 has been said, above p. 16ff., regarding the age of the mate- 
 rials of the Jatakas, the story indicates that the Vanias of 
 Western India undertook trading voyages to the shores of the 
 Persian Gulf and of its rivers in the fifth, perhaps even in the 
 sixth century B. C., just as in our days. This trade very pro- 
 bably existed already in much earlier times. For the Jatakas 
 contain several other stories, describing voyages to distant 
 lands and perilous adventures by sea, in which the names 
 of the very ancient Western ports of urparaka-Supara and 
 Bharukacha-Broach are occasionally mentioned. References to 
 sea-voyages are also found in two of the most ancient Dhar- 
 
 1 No 339, see Fausboll, Jatakas, Vol. Ill, p. 126 ff. It has been trans- 
 lated by Professor Rhys Davids in the Babylonian and Oriental Record, 
 Vol. Ill, p. 7ff.
 
 IV. The Derivation of the Brahml. 85 
 
 masutras. Baudhayana, 1 Dh. S. II, 2, 2, forbids them to the 
 orthodox Brahmans and prescribes a severe penance for a 
 transgression of the prohibition. But he admits, Dh. S. I, 2, 4, 
 that such transgressions were common among the "Northern- 
 ers", or strictly speaking, the Aryans living north of the 
 author's home, the Dravidian districts. The other forbidden 
 practices, mentioned in the same Sutra as customary among 
 the Northerners, such as the traffic in wool and in animals Avith 
 two rows of teeth, (horses, mules, etc.), leave no doubt that the 
 inhabitants of Western and Northwestern India are meant. 
 It follows as a matter of course that their trade was carried 
 on with Western Asia. The same author, Dh. 6. I. 18. 14 and 
 Gautama, X. 33 fix also the duties, payable by shipowners to 
 the king. Even from still earlier times there is the story of 
 a shipwreck, the scene of which must have been the Indian 
 [82] Ocean. Numerous hymns of the Rgveda mention the mighty 
 deed of the twin brethren, the Asvins, who saved Bhujyu, the 
 son of Tugra, from the sea, "where, as one account says, 2 
 there is no support, no rest for the foot or the hand, after 
 he had ascended the hundred-oared galley," of the two deities. 
 -The later Vedic literature contains also a few evidently 
 Semitic legends, among which that of the Deluge and Manu's 
 preservation in a ship, built by the advice of a miraculous 
 fish, is the most noteworthy, 3 and it is possible, that they may 
 have been brought over from Mesopotamia by the early Ind- 
 ian navigators and traders. But this is of course a mere pos- 
 
 1 Sacred Books of the East, Vol. XIV, pp. 146, 200, 217. Later Smrtis e. g. 
 Manu VIII, 157, give rules regarding marine insurance and other matters 
 referring to sea-borne trade. Moreover, Manu III, 156 declares a Brahman, 
 who has gone to sea, to be unworthy of entertainment at a Sraddha. 
 
 2 R. V. I, 116. 5, see also the larger St. Petersburg Dictionary sub voce 
 Bhujyu, and Prof. Oldenberg, Vedische Religion, p. 214. I quite agree 
 with Professor Oldenberg regarding the interpretation of the myth, but 
 I would not venture to infer from the deeds of the Asvins and of their 
 Greek representatives, the Dioskouroi, that the Indo-European race ori- 
 ginally dwelt near the sea. 
 
 3 I am glad to see that Professor Oldenberg, Vedische Religion, p. 276, 
 also declares this myth to be borrowed from a Semitic source, in spite 
 of Prof. M. Muller's and Prof. Lindner's attempts to make it an Indo- 
 Aryan invention.
 
 86 Indian Studies. No III. 
 
 sibility, and other explanations of their occurrence in the Brah- 
 manical literature may be and have been, suggested. The pas- 
 sages, adduced above, are however sufficient to prove that the 
 Indo-Aryans began to navigate the Indian Ocean in very re- 
 mote times, and it is, therefore, quite imaginable that they 
 themselves imported the Semitic letters from Mesopotamia. l 
 
 Between this importation and the complete elaboration of 
 the Brahma alphabet there lay, however, in all probability a 
 prolonged period. This, I think, appears from the following 
 considerations. One of the undeniable results of the pre- 
 ceeding enquiry is that the Brahma alphabet must be con- 
 sidered the work of Brahmans, acquainted with phonetic and 
 grammatical theories. The Pandit's hand is clearly visible in 
 the arrangement of the letters, used by Asoka's masons at 
 Mahabodhi Gaya, according to their organic value as vowels, 
 diphthongs, nasalised vowel, vowel with the spirant, gutturals, 
 palatals and linguals. And it is also visible at a much earlier 
 stage, in the very formation of the alphabet. Nobody but a 
 [83] grammarian or phoneticist would have thought of deriving five 
 nasals, one for each class of the Indian consonants, from the two 
 Semitic prototypes and of inventing in addition a sign to denote 
 the nasalisation of vowels, the Anusvara, or of forming two 
 spirants, ha and the Visarga. Nobody but a Sanskrit grammar- 
 ian would express the initial U by half the sign for va, z and 
 the phonetically very different, but etymologically allied sa 
 and sa by modifications of one sign, or derive the initial from 
 U, /from E s and la from c/a. 4 And only a grammarian would 
 invent the peculiar system of notation for medial vowels, 
 which throughout marks the distinction between short and long- 
 ones 5 omits the short a, and expresses the long a by adding to the 
 consonants the mark, used for differentiating A from -4, and the 
 remaining medial vowels by combinations of the initial vowel- 
 
 1 Compare the case of the Tibetan alphabet, the elements of which the 
 Tibetan Thon - mi or Sambhota is said to have imported from Magadha 
 between A. D. 630660, Jour. As. Soc. Bengal, vol. LVII, p. 41 f. 
 
 2 See above p. 74. 
 8 See above p. 66 f. 
 
 4 See above p. 77. 
 
 5 See Wackernagel, Altindische Grammatik, vol. I, p. LVII, Note 3.
 
 IV. The Derivation of the Brahml. 87 
 
 signs, or of modifications thereof, with the consonants. This 
 is so complicated and so highly artificial that only a Brahman's 
 or Pandit's ingenuity can have worked it out. 
 
 There are also very good reasons for alleging that an Ind- 
 ian alphabet, elaborated by traders or other men of business, 
 clerks and accountants, would never have possessed a single 
 medial vowel. For until a very recent period, within the last fifty 
 years, the Indian traders never used any medial vowels in their 
 books or in their correspondence. Almost every child in Gujarat 
 knows the story of the letter, 1 sent by a Vanio to his relatives, 
 which caused great grief and lamentation owing to the want of 
 the vowel signs. The letter, it is said, contained the following pas- 
 sage: oficfi ^n?T *re ^^R ^efi ere; W H On seeing it, the reci- 
 pients at once began to lament loudly. They interpreted it to 
 mean: cfrptfr ^TT5T jpft *T%T ^% ^!TWt f% % "Uncle died to- 
 day and aunt beats her breast." A sympathising neighbour 
 inquired for the cause of the wailing. On being shown the 
 letter, he remarked that the outside did not bear the usual 
 superscription of announcements of deaths, viz. "Strip and 
 [84] read", 2 and he suggested that the missive might be read: 
 SRTSfit ^r5T*fK *nft ^ ^iTWt ^te % "Uncle has gone to Aj- 
 mlr and aunt is at Kot." On further enquiry this proved to 
 be correct. 
 
 The progress of vernacular education and the action of 
 the English lawcourts and of the native princes has of late 
 changed these habits of the mercantile classes. As the law- 
 courts refuse to take ledgers written without vowel-signs as' 
 legal evidence, as the native princes follow suit, 3 and as the 
 schools now teach composition, the Vanias and accountants write 
 
 1 The story is a regular Indian "Joe Miller", and is commonly told by 
 the masters to the schoolboys. Dr. G. A. Grierson informs me that a 
 similar story is current in the Bengal Presidency. 
 
 2 A Hindu becomes impure on hearing of the death of a relative and is 
 obliged to throw away the clothes, which he wears when the news of 
 such an event comes. In order to obviate unnecessary loss, the announ- 
 cements of deaths bear on the outside the words, given above. 
 
 3 In 1875 Maharaja Ranbirsingh of Kasmlr told me that he had weaned 
 his clerks from the bad habit of writing their Takkarl or Dogra cha- 
 racters without vowels by refusing to pass accounts written in this 
 manner.
 
 88 Indian Studies. No III. 
 
 at present more frequently in the same manner as the Brah- 
 mans do and have done always and almost without exception. 1 
 With such propensities prevailing among the business 
 people of modern times, it is difficult to believe that those be- 
 longing to a very remote antiquity would have acted differently 
 and would have framed for their writing a vowel-system which 
 their descendants discarded. Nevertheless though the Brah- 
 man schoolmen undoubtedly have framed the Brahma alpha- 
 bet, the introduction of its elements, the Semitic signs, into 
 India is presumably due to the merchant class. 2 For the Va- 
 nias naturally came most into contact with foreign nations. 
 Moreover, they were the men who most urgently wanted a 
 means for perpetuating the record of their daily transactions, 
 while the Brahmans possessed since very early times the sys- 
 tem of oral instruction for preserving their literary composi- 
 [85] tions and for teaching them to their pupils, to which they 
 have always adhered. Traces of the existence of this system are 
 found, as Professor Max Miiller has already pointed out, even 
 in the Rgveda, in the famous Frog - hymn, R.V., VII, 103, 
 5, where it is said of the bull-frogs, that the one imitates the 
 cry of the other, "just as a pupil repeats the words of his 
 teacher". Its full development, which is found in the later 
 Vedic works, both the Brahmanas and the Vedangas and has 
 been described repeatedly, 3 must certainly be as early as 
 the period when the Semitic letters can have been imported, 
 or even more ancient. With this system the Brahmans can- 
 not have felt the necessity for writing so strongly as the men 
 of business, and it is also for this reason improbable that they 
 should have been the first Hindus who practised writing. Nay, 
 
 1 I have seen one Gujaratl inscription without vowels, which may be due 
 to a Brahman. It is incised on the right hand gate post of the temple 
 of Dharanldhar at Dehema in Northwestern Gujarat and omitting the 
 date, runs thus: rf^ T TffST ^cf T *R T T^ f T II Its meaning 
 was interpreted to me, as follows: rf^^T'H' '''fit^T ^Tpf^ % ^T^'fl' 
 TT^h" TTrWl " The Musalman army^came and the Rao of Vav was 
 defeated." 
 
 2 See also Westergaard, Zwei Abhandlungen, p. 37 f. 
 
 8 M. Miiller, History of Ancient Sanskrit Literature, p. 503 ff., compare 
 also A. Weber, Indische Studien, X, p. 128ff.
 
 IV. The Derivation of the Brahml. 89 
 
 it may be even doubted whether they cared to undertake the 
 adaptation of the foreign invention very soon after it had be- 
 come known to the mercantile class. 
 
 Further, when they undertook it, the evolution of the 46 
 signs of the Brahml lipi from the 22 Semitic characters cannot 
 have been accomplished very quickly. The evidence of the 
 Bhattiprolu alphabet shows that in the case of several letters 
 more attempts than one were made, and the alphabet of the 
 Edicts, the ordinary Brahml lipi, likewise bears witness that 
 the signs were invented gradually. As has been pointed out 
 above, the dental dha yielded, by a slight transposition of one 
 of its lines, the dental da, from this the lingual da was derived 
 by the omission of the lower vertical stroke, and from the 
 lingual da came, by the addition of a curve, the lingual dha, 
 as well as, by the addition of the semicircle, the lingual la. 
 The series of the derivatives from va, from the dental tha, 
 the dental na, the Bhattiprolu s and from the initial E are 
 similarly complicated. It is incredible that in these cases the 
 whole series of derivatives should have been invented at one 
 time or even in quick succession, though no doubt the Brah- 
 mans had their system of phonetics, based on spoken works, 
 to guide them and to help them on. 
 
 [86] It would thus seem that a not inconsiderable interval 
 must lie between the first importation of the Semitic letters and 
 the complete elaboration of the Brahma alphabet. First the im- 
 ported characters remained in the hands of the traders and 
 the men of business, during which period, as perhaps may be 
 inferred from the treatment of the Arabic letters during the 
 Mahommedan period, none or very slight modifications were 
 introduced. Next came the transference of the foreign invention 
 to the Brahmans and finally its adaptation to the wants of the 
 Sanskrit language. 
 
 These considerations certainly show that the introduction 
 of the Semitic letters must fall centuries before the period when 
 the Brahma alphabet was in general use, i. e. 500 B. C. or 
 thereabouts. -And they thus confirm the approximate date, 
 deduced above from the age of the Semitic signs, which ap- 
 pear to have been the prototypes of the Indian modifications. 
 This date, I repeat, is merely approximate and I may add,
 
 90 Indian Studies. No III. 
 
 provisional. New finds of ancient Semitic or Indian inscriptions 
 may, and very likely will, alter it. And I should not be sur- 
 prised, if such discoveries forced us to put the terminus a quo 
 for the introduction of the Semitic characters into India earlier 
 than is advisable on the evidence, available at present. 
 
 Finally, the fact that the Brahma alphabet is the work of 
 the Brahmans has also, it seems to me, a certain bearing on the 
 vexed question whether in ancient times writing was used for liter- 
 ary purposes. I believe that it enables us to answer this question 
 in the affirmative, of course with certain reservations. For the 
 Brahmans, though often considered mere dreamers, are in re- 
 ality very practical people, who, as far as my observation goes, 
 do not take and never have taken trouble with anything that 
 does not serve their purposes. As they adapted the Semitic 
 letters to the wants of their sacred language in a very thorough 
 manner, I consider it certain that they also utilised their in- 
 vention at once for their special aims, the cultivation of learn- 
 ing, and that they committed at least their scientific composi- 
 tions to writing. It is not necessary, nor even probable, that 
 in early times the MSS. were used otherwise than esoterically, 
 as auxiliaries for composition and for the preservation of the 
 texts, much in the manner suggested by Geheimrath von Bb'ht- 
 lingk, Professor Whitney and Dr. Burnell. 1 
 
 Their Mantras and other sacred compositions may have 
 remained unwritten somewhat longer. That is no more than 
 might be expected, as the Brahmans had a great interest in 
 [87] keeping their "great medicine" secret. And there are also 
 several indications to this effect, such as the imprecations 
 against copyists of the Veda and the general feeling, even 
 among the heterodox sectarians, that sacred books ought to be 
 preserved only orally. 
 
 The Southern Bauddhas and the Jainas give expression 
 to this feeling by asserting that their sacred books were written 
 only many centuries after their composition. But it may be doubted, 
 whether their dates are always quite correct. The occurrence 
 of a petaki (literally "a Pitaka-possessor") a monk who knows 
 one or more Pitakas (see above p. 18), certainly proves that the 
 
 1 See Elements of South-Indian Palaeography, p. 10 and compare also the 
 Grundriss, I, 11, p. 3 f.
 
 IV. The Derivation of the Brahml. 91 
 
 Buddhist scriptures were written, when the Bharahut Stupa was 
 built. Pitaka is only c a basket' or 'a box' and corresponds to 
 the modern dabado of cardboard or wood in which the Jainas 
 usually keep the MSS. of their parish libraries. As soon as the 
 Buddhists divided their scriptures into Pitakas, 1 they must of 
 course have been written. If, as I believe, the inscription of the 
 petaki, which shows the same characters as the Edicts, be- 
 longs approximately to the same time as the latter, the traditional 
 date of the Ceylonese Buddhists, who say that their Canon was 
 first committed to writing about 88 B. C., must be considerably 
 wrong. The oldest MSS., actually found, are probably the 
 birch bark leaves, inscribed with Kharosthi letters, from the 
 topes of the Panjab. 3 Next comes the Bower MSS. with the 
 characters of the oldest Gupta inscriptions, then follow the 
 Horiuzi palmleaf, the Bakhshali MS., the Cambridge Collection 
 from Nepal and the Bombay Collections of Jaina MSS. Older 
 documents than are accessible at present, both on birchbark 
 or palmleaves and on stone or metal, will no doubt be found, 
 as soon as the old historical sites of India are excavated in a 
 thorough and rational manner. 
 
 1 The term pitaka occurs in one of the latest portions of the Vinaya, the 
 Parivara, I, 1 (Oldenberg Vinaya pitaka, vol. V, p. 3), where also a tipe- 
 takl is mentioned. 
 
 2 Though often appealed to in proof of the late use of writing for literary 
 purposes (see e. g. Sacred Books of the East, vol. XIII, p. XXX Vff.), this 
 assertion of the Dipavamsa and Mahavamsa is, I fear, not worth much 
 more than the contradictory story of the Chinese Buddhists. The Chinese 
 boldly assert that the Canon was written on palm leaves immediately 
 after Buddha's death (Grundriss, I, 11, p. 89) and that in A. D. 489 
 a MS. of the Vinaya was extant, which bore 975 dots, marking the number 
 of the readings at the annual Pavarana since the time of Upali, who is 
 said to have entered the first dot (Jour. R. As. Soc., 1896, p. 436 f.). Like 
 similar, certainly incorrect, stories of the Brahmans and others regarding 
 their scriptures, the tradition of the two Pali chronicles may be taken 
 merely as the product of the well founded belief that the Buddhists, 
 imitating the Brahmans, preserved their scriptures orally for a prolonged 
 period. But even if it is considered to be historical, it refers only tc 
 Ceylon not to India. 
 
 8 One small fragment with the letters mi, dha and ya (?) is figured in 
 H. H. Wilson's Ariana Antiqua, Plate III, No 11.
 
 Appendix I. 
 The Origin of the Kharosthl Alphabet. 
 
 [44] * Though the origin of the Kharosthl alphabet is much 
 easier to explain than the derivation of the Brahml and though 
 the general lines for the enquiry have already been settled by 
 others, yet a somewhat fuller review of the whole question, 
 than the narrow compass of my Grundriss der indischen Pa- 
 laeographie permits, will perhaps not be superfluous. The very 
 considerable progress, which has been achieved, is chiefly due 
 to the discussions of the Kharosthl by Mr. E. Thomas in his 
 edition of Prinsep's Essays, vol. II, p. 147 ff., by Dr. Isaac 
 Taylor in The Alphabet, vol. II, p. 256 ff., and by Sir A. Cun- 
 ningham, who has also settled the value of many of its signs, 
 in his book on The Coins of Ancient India, p. 3 Iff. 
 
 Sir A. Cunningham's remarks refer to the first point which 
 requires consideration in all questions of this kind, viz. the 
 true character of the script, the origin of which is to be deter- 
 mined. He has emphatically recalled to the memory of the 
 palaeographists that the Kharosthl is an Indian alphabet, and 
 by an ingenious utilisation of his finds of ancient coins in the 
 ruins of Taxila he has shown that the Kharosthl held always, 
 during the whole period for which epigraphic evidence is avail- 
 able, only a secondary position by the side of the Brahma al- 
 phabet even in Northwestern India. It is rather curious that 
 the reminder regarding the essentially Indian character of the 
 alphabet should have been necessary, as even a superficial 
 
 1 The bracketed figures refer to the pages of the first edition of this 
 article in the Wiener Zeitschrift fiir die Kunde des Morgenlandes, (Vienna 
 Oriental Journal) vol. IX.
 
 The Origin of the Kharosthl Alphabet. 93 
 
 considera- [45] tion of its letters teaches that lesson. Its full 
 system of palatals and linguals cannot be designed for any 
 other language than Sanskrit or an ancient Prakrit, the only 
 forms of speech which possess five sounds of each of the two 
 classes mentioned. If this has been sometimes forgotten and 
 even Bactria has been considered as the cradle of the Kharosthl, 
 the cause is no doubt the loose way in which it used to be 
 called the "Bactrian, Bactro-Pali or Indo-Bactrian" alphabet, 
 which appellations are due to its occurrence on the coins of 
 Greek kings, who, originally ruling over Bactria, conquered 
 portions of Northwestern India. Sir A. Cunningham very pro- 
 perly points out, op. cit, p. 35, that not a single Kharosthl 
 inscription has been found north of the Hindu Kush and that 
 in Bactria a different alphabet seems to have been used. He 
 further proposes to substitute for "Indo-Bactrian" the Indian 
 term "Gandharian", which would have been suitable in every 
 way, if in the mean time the old Indian name had not been 
 found. The districts, in which the largest number of Kharosthl 
 inscriptions have been found, are situated roughly speaking 
 between 69 73, 30' E. L. and 33 35 N. L., while single 
 inscriptions have turned up southwest near Multan, south at 
 Mathura and east at Kangra, and single letters or single words 
 even at Bharahut, in Ujjain and in Maisur. This tract, to 
 which the Kharosthl inscriptions of the third century B. C. 
 are exclusively confined, corresponds to the Gandhara country 
 of ancient India, the chief towns of which were Puskalavatl- 
 Hashtnagar to the west of the Indus and Taxila or Shah Deri to 
 the east of the river. And it is here, of course, that the 
 Kharosthl alphabet must have originated. 
 
 In addition, Sir A. Cunningham has shown that the Kha- 
 rosthl held always a secondary position and was used even in 
 the earliest times side by side with the Brahmi. This is proved 
 by the evidence of his coins from Taxila, several of which 
 bear only Brahma inscriptions or Kharosthl and Brahma in- 
 scriptions, with letters of the type of Asoka's Edicts. The an- 
 alysis of the legends, which I have given ante p. 48 f., shows 
 that those of four types have been issued by traders' guilds 
 and that one is probably a tribal [46] coin, belonging to a sub- 
 division of the Asvakas or Assakenoi, who occupied portions
 
 94 Appendix I. 
 
 of the western bank of the Indus at the time of Alexander's 
 invasion. This result considerably strengthens Sir A. Cun- 
 ningham's position, as it indicates a popular use of the Brahma 
 alphabet in the very home of the Kharosthl. 
 
 The next step which is required, is to find the class of 
 alphabets, to which the prototypes of the Kharosthl belonged. 
 This problem is settled, as Mr. Thomas has first pointed out, 
 by the close resemblance of the signs for na, lia, va and ra 
 to, or identity with, the Nun, Beth, Waw and Resh of the 
 transitional Aramaic alphabet, and requires no further discussion. 
 
 Then comes the question, how the Hindus of Northwestern 
 India can have become acquainted with the Aramaic characters 
 and which circumstances may have induced them to utilise 
 these signs for the formation of a new alphabet. Dr. Taylor, 
 The Alphabet, vol. II, p. 261 f., answers this by the suggestion 
 that the Akhaemenian conquest of Northwestern India, which 
 occurred not very long before B. C. 500 and led to a pro- 
 longed occupation, probably carried the Aramaic or, as he 
 calls it, the Iranian, Persian or Bactrian, alphabet into the 
 Panjab and caused its naturalisation in that province. Though 
 it seems to me, just as to Sir A. Cunningham, impossible to 
 accept Dr. Taylor's reasoning in all its details, I believe with 
 Sir A. Cunningham that he has found the true solution of this 
 part of the problem. For even if the doubtful and often disputed 1 
 conquest of western Gandhara by Cyrus about B. C. 530 and 
 the equally doubtful and disputed continuance of its occupation 
 during the reign of Cambyses were historical, the introduction 
 of the Aramaean letters into the Indian province could only 
 date from the reign of Darius. According to all accounts Darius 
 first organised the administration of the Akhaemenian empire 
 and first sent Satraps with their staff of subordinate officials 
 into the provinces. It is obvious that only such an administra- 
 tive organisation could cause the spread of the official alphabet 
 in the remoter portions of the empire. 
 
 One argument in his favour is the occurrence of the Old 
 Persian word dipi "writing, edict" in the Northwestern versions 
 
 1 Disputed e. g. by Lassen, Niese, J. Oppert and G. Rawlinson, assumed to 
 be historical by Droysen, Dunker and others.
 
 The Origin of the Kharosthl Alphabet. 95 
 
 of the Edicts and of its derivatives dipati "he writes" and 
 dipapati "he causes to write", which are not found in any 
 other Indian language. Dipi is undoubtedly as Dr. Taylor 
 himself has stated an Old Persian loanword, and all the three 
 words mentioned point to a Persian influence, dating from the 
 Akhaemenian period. And the Sanskrit and Pali lipi or libi 
 "writing, written document", which does not occur in the Vedic 
 and Epic literature, nor in the ancient works of the Buddhist 
 Canon of Ceylon, but appears first in Sutras of Panini, a na- 
 tive of Gandhara (traditional date circiter 350 B. C.), furnishes the 
 same [47] indication, since in all probability, as Dr. Burnell con- 
 jectured, it is a corruption of di-pi, favoured by a fancied connexion 
 with the verb lip, limpati "he smears" (see above, 21 f.). Equally 
 valuable is a second point, the fact that the territory of the 
 Kharosthl corresponds very closely with the extent of the por- 
 tion of India, presumably held by the Persians. Dr. Taylor 
 and Sir A. Cunningham very justly lay stress on the statement 
 of Herodotus (III, 94, 96) who asserts that the Persian satrapy 
 of India paid a tribute of 360 talents of gold dust. They na- 
 turally infer that the Indian possessions of the Akhaemenians 
 must have been of considerable extent, as well as that they must 
 have included the greater portion of the Panjab. 
 
 But there remain still two gaps which must needs be 
 filled up. The Akhaemenian theory requires it to be shown 
 that the ancient Persians actually used the Aramaic letters and 
 that peculiar circumstances existed which compelled the Hindus 
 to use these letters. The second point is at present particularly 
 important, because the literary evidence regarding the use of 
 writing in India 1 (with which the epigraphic evidence fully 
 agrees) proves that the Hindus were by no means unlettered 
 in the fifth and sixth centuries B. C., but possessed and exten- 
 sively used an alphabet, which probably was a form of the 
 Brahmi lipi. As long as it was possible to maintain that the 
 Hindus became acquainted with the art of writing not earlier 
 than 400 B. C., it was, of course, easy to understand, that the 
 use of the Aramaic letters by the conquerors of Northwestern 
 India should have acted as a natural incentive for their Hindu 
 
 1 Ante p. 5ff.
 
 96 Appendix I. 
 
 subjects to form out of these characters an alphabet suited 
 for their own language. But the case becomes different, if it 
 must be admitted that the Hindus possessed already a script 
 of their own before the Persian conquest. With this admission 
 it becomes neccessary to show that there were special circum- 
 stances which forced them to use the alphabet of their conquerors. 
 Both the points just discussed are explained, it seems to 
 me, by certain discoveries, made of late years in Semitic pa- 
 laeography. M. Clermont-Ganneau's important articles in the 
 Revue archeologique [48] of 1878 and 1879 have shown that 
 the Aramaic language and writing, which already in the times 
 of the Assyrian empire occur in contracts and on the official 
 standard weights, were frequently employed for official cor- 
 respondence, accounts and other official purposes during the 
 rule of the Akhaemenian kings in many different provinces 
 of their empire. Egypt has furnished Aramaic inscriptions on 
 stones and potsherds as well as Aramaic Papyri, addressed to 
 Persian governors; in western Asia and in Arabia both inscrip- 
 tions and numerous Satrap coins with Aramaic legends have 
 been found, and even Persia has yielded an Aramaic inscription 
 (of which unfortunately no trustworthy facsimile exists) at Senq- 
 Qaleh, midway between Tabriz and Teheran. 1 And, I may 
 add, there is also a scrap of literary evidence to the same 
 effect. A statement in the Book of Ezra IV. 7, points to the 
 conclusion that the Aramaic language and writing was well- 
 known in the Imperial chancellerie at Susa. For it is said that 
 a letter, addressed by the Samaritans to Artaxerxes, "was 
 written", as the Revised Version of the Bible has it, "in the 
 Syrian (character) and in the Syrian tongue". The Samaritans 
 would hardly have adopted the "Aramlt" in addressing their 
 liege lord, if it had not been commonly used in official cor- 
 respondence, sent out from, or into, the Imperial Secretariat. 2 
 
 1 See Ph. Berger, Histoire de VEcriture dans VAntiquite, p. 218ff., where 
 M. Berger pertinently remarks with respect to the last inscription, that 
 it puts us on the road to India. 
 
 2 As Prof. Euting kindly points out to me, a similar inference has already 
 been drawn from the above passage by the authors of the Kurzgef. Com- 
 mentar z. d. heil. Schriften d. N. u. A. Test., hg. v. H. Strack und O. 
 Zockler; Alt. Test, Abth. 8, p. 159.
 
 The Origin of the Kharosthi Alphabet. 97 
 
 The custom itself, no doubt, has to be explained by a strong 
 infusion of Aramaeans, or of men trained in the learning of 
 the Aramaeans, in the lower grades of the Persian Civil Ser- 
 vice, among the scribes, accountants, treasurers and mint-masters, 
 and this is no more than might be expected, when a race like 
 the Persian suddenly comes into the possession of a very large 
 empire and becomes the heir of an older civilisation. 
 
 Under these circumstances it appears natural to assume 
 that the Persian Satraps carried with them also into India their 
 staff of [49] subordinates, who were accustomed to the use of 
 the Aramaean letters and language. And this would fully ex- 
 plain, how the Hindus of the Indo-Persian provinces were 
 driven to utilise the characters, commonly employed by the 
 scribes and accountants of their conquerors, though they al- 
 ready possessed a script of their own. The Kharosthi alphabet 
 would appear to be the result of the intercourse between the 
 offices of the Satraps and of the native authorities, the Indian 
 chiefs and the heads of towns and villages, whom, as the ac- 
 counts of the state of the Panjab at the time of Alexander's 
 invasion show, the Persians left in possession in consideration 
 of the payment of their tribute. The Hindus probably used at 
 first the pure Aramaic characters, just as in much later times 
 they adopted the Arabic writing for a number of their dialects, 
 and they introduced in the course of time the modifications, 
 observable in the Kharosthi alphabet, for which process the 
 additions to the Arabic alphabet, employed for writing Hindi, 
 furnish an analogy, perhaps not perfect but nevertheless worthy 
 of notice. 
 
 In support of these conjectural combinations three further 
 points may be adduced. First, the Kharosthi alphabet is not 
 a Pandit's, but a clerk's, alphabet. This appears to me evident 
 from the cursive appearance of the signs, which has been fre- 
 quently noticed by others, from its (according to Indian views) 
 imperfect vowel-system, which includes no long vowels, from 
 the employment of the Anusvara for the notation of all nasals 
 before consonants and from the almost constant substitution of 
 single consonants for double ones. The expression of all long 
 vowels by separate signs, which occurs in no other ancient 
 alphabet but the Brahml lipi, was no doubt natural and de- 
 Buhl er, Indian Studies. III. 7
 
 98 Appendix I. 
 
 sirable for the phoneticists or grammarians, who developed that 
 alphabet. 1 But it is a useless encumbrance for men of business, 
 whose aim is rather the expeditious despatch of work than 
 philological or phonetic accuracy. Hence, even the Indian 
 clerks and men of business using the Brahml, have never paid 
 much attention to their correct use, though they were in- [50] 
 structed by Brahmans in the principles of their peculiar alphabet. 2 
 If, therefore, these signs, which have only a value for school- 
 men, do not occur in the Kharosthl, the natural inference is 
 that this alphabet was framed by persons who paid regard 
 only to the requirements of ordinary life. The other two pecu- 
 liarities mentioned, the substitution of the Anusvara for all na- 
 sals, standing before consonants, and the substitution of ka for 
 kka, of ta for fta and so forth and of kha for likha, of dha 
 for ddha and so forth, are clearly the devices of clerks, who 
 wished to get quickly through their work. If thus the Kha- 
 rosthl appears to be an alphabet, framed with particular regard 
 to the wants of clerks, that agrees with and confirms the as- 
 sumption, put forward above, according to which it arose out 
 of the official intercourse between the scribes of the Satraps 
 and those of the native chiefs or other authorities. 
 
 More important, however, is the second point, which is 
 intimately connected with the details of the derivation of the 
 Kharosthl. The originals of the Kharosthl letters are, it seems 
 to me, to be found in the Aramaic inscriptions, incised during 
 the rule of the earlier Akhaemenian kings. The whole ductus 
 of the Kharosthl with its long verticals or slanting downstrokes 
 is that of the Saqqarah inscription of 482 B. C. and the pro- 
 bably contemporaneous larger Teima inscription, which Pro- 
 fessor Euting assigns to circiter'bQO B. C. It is also in these 
 inscriptions that most of the forms occur, which apparently 
 have served as models for the corresponding letters of the 
 Kharosthl. One or perhaps two seem to rest on forms found 
 in the somewhat later Lesser Teima, Serapeum and Stele Vati- 
 cana inscriptions, while three are connected with .older letters 
 on the Assyrian Weights and the Seals and Gems from Babylon. 
 
 1 Ante p. 86 f. 
 
 2 Ante p. 43 f., note 3.
 
 The Origin of the Kharosthi Alphabet. 99 
 
 The accompanying Comparative Table, Plate II, 1 illustrates 
 the details of the derivation, as I understand it. Cols. I and II have 
 been reproduced by photozincography from Professor Euting's 
 Tabula Scripturae [51] Aramaicae, Argentorati 1892, and give 
 the twenty Aramaic signs, which, as I believe, have been uti- 
 lised by the Hindus, Theth and Ain being rejected by them. 2 
 In Col. I the fat signs belong to the Teinia inscription (Euting, 
 Col. 9) with the exception of No 1, I, b and No 9, I, b c, 
 which come from the Stele Vaticana (Euting, Col. 12). The 
 thin signs have been taken from the Saqqarah inscription 
 (Euting, Col. 11) with the exception of No 4, I, a; No 9, I, a; 
 No 10, 1, b and No 20, I, a, which are from the Assyrian 
 Weights and the Babylonian Seals and Gems (Euting, Cols. 6, 
 8), as well as of No 17, I, a b, which are from the Serapemn 
 inscription (Euting, Col. 12), and of No 10, I, a, which Pro- 
 fessor Euting has kindly added on .once more looking over 
 the Babylonian Aramaic inscriptions. 3 
 
 The signs of Col. II have all been taken from Professor 
 Euting's Cols. 14 17, and represent the chief types on the 
 Aramaic Papyri, which M. J. Halevy 4 and others believe to 
 be the prototypes of the Kharosthi. They have been given in 
 my Table chiefly in order to show that they are not suited 
 for the derivation. Col. Ill gives the oldest forms of the bor- 
 rowed Kharosthi letters according to Tafel I of my Grundriss 
 der Indischen Palaographie, and Col. IV with the signs, which 
 I consider to be derivatives invented by the Hindus, comes 
 from the same source. 
 
 1 Arranged by Dr. W. Cartellieri and etched by Messrs. Angerer & Goschl 
 of Vienna. 
 
 2 According to Dr. Taylor these two characters are also, reflected in the 
 Kharosthi. But the sign opposite Theth in his Table, The Alphabet, 
 vol. II, p. 236, is a late vra, and Ain, cannot be 0, as he doubtfully 
 suggests. M. Halevy identifies Theth with the letter, which used to be 
 read tha, but is in reality tha and a derivative from fa, see below. 
 
 3 In this as well as in other respects I have to acknowledge Professor 
 Euting's kind assistance, who sacrificed a good deal of time in order to 
 verify the Semitic signs, which I had selected for comparison, in the 
 Plates of the Corp. Inscr. Sem. and carefully went with me through my 
 Table during a personal interview in Strassburg. 
 
 4 Journ. Asiatique 1885, p. 251 ff. 
 
 7*
 
 100 Appendix I. 
 
 Before I proceed to give my remarks on the details of 
 the derivation, I will restate the general principles which have 
 to be kept in mind for this and all other similar researches. 
 
 [52] (1) The oldest actually occurring signs of the alphabet 
 to be derived (in this case the Kharosthi) have to be compared 
 with the supposed prototypes (in this case actually occurring 
 Aramaic signs) of the same period (in this case of ca. 500 400 
 B. C.). 
 
 (2) Only such irregular equations of signs are admissible 
 as can be supported by analogies from other cases, where 
 nations are known to have borrowed foreign alphabets. Thus 
 it is not permissible to identify the Kharosthi sign for ja with 
 the Aramaic ga on account of a rather remote resemblance 
 between what the modern researches have shown to be a se- 
 condary form of the Kharosthi palatal media and the guttural 
 media of the Aramaeans. 
 
 (3) The comparison must show that there are fixed prin- 
 ciples of derivation. 
 
 The latter are given chiefly by the unmistakable tenden- 
 cies underlying the formation of the Kharosthi signs, 
 
 (1) A very decided predilection for forms, consisting of 
 long vertical or slanting lines with appendages added do the 
 upper portion, 
 
 (2) An antipathy against such with appendages at the 
 foot of the verticals, which in no case allows a letter to con- 
 sist of a vertical with an appendage at the foot alone; 
 
 (3) An aversion against heads of letters, consisting of 
 more than two lines rising upwards, as well as against top 
 lines encumbered with transverse or pendant strokes. 
 
 These tendencies required two Aramaic letters, Lamed, 
 (No 11, I and III) and Shin (No 19, I and III) to be turned 
 topsy-turvy, and caused in the Shin the development of a long 
 vertical out of the short central stroke, as well as mutilations 
 of some other signs. And it would seem that the aversion 
 against appendages at the foot is probably due to the desire 
 to keep the lower ends of the matrkds free for the addition 
 of the medial u, the Anusvara and the ra-strokes which are 
 ordinarily added here, while the top lines were kept free from 
 transverse or pendant lines on account of the signs for medial
 
 The Origin of the Kharosthl Alphabet. 101 
 
 i, e and o. Some other changes, such as turnings from the 
 right to the left, have been made in order to avoid collisions 
 with other signs, while again other rnodifi- [53] cations are purely 
 cursive or due to considerations of convenience in writing. 
 
 As regards the details, I have to offer the following re- 
 marks regarding the Borrowed Signs. 
 
 No 1. The identity of A with AUpli is evident enough 
 (Thomas, Taylor, Halevy). The long stretched shape of the 
 Kharosthl letter, which leans to the right, makes it in my 
 opinion more probable that it is a simplification of a sign like 
 that from the Saqqarah inscription in Col. I, a, than that it 
 should be connected with the diminutive letters in Col. I, b 
 and in Col. II, which are inclined the other way. 
 
 No 2. Ba is, of course, a slightly modified form of the 
 Beth in Col. I, a b (Thomas, Taylor and Halevy). The up- 
 ward bulge next to the vertical has been introduced in order 
 to make the letter with one stroke of the pen, and the bent 
 line at the foot is represented by a prolongation of the vertical 
 in accordance with the principle stated above. The Beth of 
 the Papyri (when cursive forms are used as in Col. II, b c 
 and in Professor Euting's Col. 15 b c, 16 b d) is more ad- 
 vanced than the Kharosthl ba. 
 
 No 3. The identity of ga (Col. Ill) with Gimel (Cols. I 
 and II) has been recognised by Dr. Taylor alone. The loop 
 on the right has been caused by the desire to make the letter 
 with one stroke of the pen. It may be pointed out as an an- 
 alogy that in the late Kharosthl of the first and second cen- 
 turies A. D. cursive loops are common in ligatures with ra 
 and ya and that there is a looped /a, exactly resembling a ga, 
 on the Bimaran vase in the word Mumjavata. The Aramaic 
 prototype may possibly have been set up straighter than the 
 forms given in Cols. I and II, and it may be noted that such 
 forms occur already on the Mesa stone and in other old in- 
 scriptions, see Euting, Cols. 1 and 3. 
 
 No 4. Da (Col. Ill) comes, as has been asserted by all 
 my predecessor's, from a Daleth like that in Col. I, a, which is 
 found, as Professor Euting informs me, already on an Assyrian 
 Weight of circiter 600 B. C. The cursive simplification of this 
 letter was therefore ancient in Mesopotamia. It re-occurs in
 
 102 Appendix I. 
 
 the Papyri, with a slight [54] modification, compare especially 
 Euting, Col. 14 b. The hook at the foot of the da, (Col. Ill, b), 
 which occurs twice in the Asoka Edicts and survives in the 
 later inscriptions, seems to have been added in order to distinguish 
 the letter from na (No 13, III, a). 
 
 No 5. The identity of ha (Col. Ill) with He has not been 
 recognised hitherto. But it seems to me derived from a round 
 He, like the Teima form in Col. I, a, with the transposition 
 of the central vertical to the lower right end of the curve, 
 which is particularly clear in the letter, given in Col. Ill, b, 
 a not uncommon form in the Asoka Edicts. A similar trans- 
 position of an inconvenient pendant, due to the consideration 
 for the medial i, e and o, will be noticed below under No 17. 
 The He of the Papyri, though not rarely round at the top, 
 shows nearly always a continuation of the central bar on the 
 outside of the topline, and hence is less suitable for comparison. 
 
 No 6. Va has preserved, as all previous writers have 
 acknowledged, exactly the form of the Waw in the Teima in- 
 scription, which re-occurs on various later documents as the 
 Ostraka from Elephantine and the Cilician Satrap coins, and 
 which is foreshadowed by the letter of the ancient Assyrian 
 Weights in Euting's Tabula Col. 6. The Papyri again offer a more 
 advanced round form, which is common in the Kharosthl in- 
 scriptions, incised during the first and second centuries of 
 our era. 
 
 No 7. Dr. Taylor alone derives ja. (Col. Ill, a b) from 
 Zain, apparently relying on the similar Pehlevi letter. The 
 form in Col. Ill, a, which is found repeatedly 1 in the Mansehra 
 version and survives in the legends of the Indo-Grecian and 
 Saka coins, is, however, without doubt the oldest, and derived 
 from a Zain, like those of the Teima inscription (Col. I, a b) 
 in which the upper bar has been turned into a bent stroke 
 with a hook rising upwards at the left end. In the second ja 
 (Col. Ill, b) the lower bar has been dropped in order to keep 
 the foot of the sign free. The Pehlevi letter probably is an 
 analogous development. The Zain of the Papyri (Col. II) is 
 
 *' Edict IK, 9 in raja, IV, 16 in raja, V, 19 in raja, V, 24 in praja, 
 VIII, 35 in raja, XII, 1 in raja.
 
 The Origin of the Kharosthi Alphabet. 103 
 
 again [55] much more advanced and unfit to be considered 
 the original of the Kharosthi sign. 
 
 No 8. With respect to the representative of the Cheth I 
 differ from all my predecessors. The Aramaic letter, such as 
 it is found in the Saqqarah inscription (Col. I, a c), in Teima 
 and various other documents 1 is exactly the same as. the Kha- 
 rosthi palatal sibilant sa. The pronunciation of the Indian sa 
 comes very close to the German ch in ich, lick etc., 2 and 
 hence the utilisation of the otherwise redundant Cheth for the 
 expression of sa appears to me perfectly regular and normal. 
 
 No 9. The derivation of ya (Col. Ill) from the Aramaic 
 Yod has been generally assumed, and it has been noticed that 
 the Kharosthi sign is identical with the late Palmyrenian and 
 Pehlevi forms (Euting, Cols. 2125, 3032, 3539, 58), which 
 of course are independent analogous developments, as well as 
 that it resembles the Yod of the Papyri (Col. II, c and Euting, 
 Cols. 14 17), where however the centre of the letter is mostly 
 filled in with ink. Still closer comes the first sign (Col. II, b) 
 from the Stele Vaticana, and it may be that a form like the 
 latter is the real prototype. But I think the possibility is not 
 precluded, that the Kharosthi ya may be an Indian modification 
 of a form like the more ancient Assyrian Aramaic sign in 
 Col. I, a, which differs only by the retention of the second bar 
 at the right lower end. The rejection of this bar was necessary 
 in accordance with the principles of the Kharosthi, stated above, 
 and may therefore be put down as an Indian modification. 
 The height of the Kharosthi ya seems to indicate that its pro- 
 totype had not yet been reduced to the diminutive size, which 
 it usually has in the Papyri, but which is not yet observable 
 in the otherwise differing letters of the Teima and Saqqfirah 
 inscriptions. 
 
 No 10. The connexion of ka (Col. Ill) with the Aramaic 
 Kaph is asserted by M. J. Halevy, but he compares the sign 
 of the Papyri [56] (Col. II), which is very dissimilar. I think, 
 there can be no doubt that the Kharo.sthT letter is a modification 
 
 1 It occurs even in the Papyri though these offer mostly more advanced, 
 rounded forms. 
 
 2 Prof. A. Kuhn long ago expressed his belief that etymologic-ally An is 
 derived from ka through a palatal %a.
 
 104 Appendix I. 
 
 of the Babylonian Kaph in Col. I, b, which was turned round 
 in order to avoid a collision with la and further received the 
 little bar at the top for the sake of clearer distinction from pa. 
 The sign in Col. I, a, which likewise comes from Babylon, has 
 been added in order to show the development of that in Col. I, b, 
 from the oldest form. 
 
 No 11. Lamed, consisting of a vertical with an appendage 
 at the foot had, as stated above, to be turned topsy-turvy in 
 order to yield the Kharosthi la, with which Dr. Taylor and 
 M. Halevy have identified it. Moreover, the curve which then 
 stood at the top was converted into a broken line 1 and at- 
 tached a little below the top of the vertical, in order to avoid 
 a collision with A. The signs of the Papyri (Col. II) are mostly 
 far advanced and cursive, so that they can not be considered 
 the prototypes of the Kharosthi la. 
 
 No 12. The Kharosthi ma (Col. Ill, a c) is, as has been 
 generally recognised, not much more than the head of the 
 Aramaic Mem in Col. I. The first two forms, which are common 
 in Asoka's Edicts and the second of which occurs also on the 
 Lido-Grecian coins, still show remnants of the side-stroke and 
 of the central vertical or slanting stroke. But they have been 
 placed on the left instead of on the right. The mutilation of 
 the letter is no doubt due, as has been suggested by others, 
 to the introduction of the vowel signs, which would have given 
 awkward forms, and the fact of the mutilation is indicated 
 by its size, as it is always much smaller than the other 
 Kharosthi signs. The curved head appears in the Saqqarah 
 Mem, which I have chosen for comparison, as well as on Ba- 
 bylonian Seals and Gems (Euting, Col. 8, e) and in the Car- 
 pentras inscription (Euting, Col. 13, c), and the later forms 
 from Palmyra prove that it must have been common. The 
 Mem of the Papyri are again much more cursive and unsuited 
 for comparison. 
 
 [57] No 13. Regarding na (Col. Ill, a), which is clearly the 
 Nun of the Saqqarah (Col. I, a b) Teima, Assyrian and Ba- 
 bylonian inscriptions, it need be only pointed out that the 
 
 1 The la of the Edicts almost invariably shows the broken line. The 
 later inscriptions offer instead a curve open below.
 
 The Origin of the Kharosthi Alphabet. 105 
 
 forms of the Papyri .are also in this case further advanced 
 than those of the Kharosthl. The na, given in Col. HI, b, is 
 a peculiar Indian development, rather rare in the Asoka Edicts. 
 
 No 14. The identity of sa with the Aramaic Samech (Col. I) 
 has hitherto not been recognised. Nevertheless the not uncom- 
 mon form of sa with the polygonal or angular head, given in 
 Col. Ill, permits us to assert that also in this case the Gan- 
 dharians used for the notation of their dental sibilant the sign 
 which one would expect to be employed for the purpose. The 
 top stroke and the upper portion of the right side of the Kha- 
 rosthl sa correspond very closely to the upper hook of the 
 Samech of Teima, being only made a little broader. The little 
 slanting bar in the centre of the Samech may be identified 
 with the downward stroke, attached to the left of the top line 
 of sa } and the lower left side of sa appears to be the cor- 
 responding portion of the Samech, turned round towards the 
 left in order to effect a connexion with the downward stroke. 
 These remarks will become most easily intelligible, if the com- 
 ponent parts of the two letters are separated. Then we have 
 for Samech ^ and for sa ^. The forms, in which the right 
 portion of the head of sa is rounded, are of course cursive. 
 The Teima form of the Samech with the little horn at the left 
 end of the top stroke is unique in the older inscriptions. But 
 the Palmyrenian letters (Euting, Cols. 2429, 32-33, 37, 
 39 40), though otherwise considerably modified, prove that 
 the Samech with an upward twist must have been common. 
 Finally, the corresponding Nabataean characters (Euting, Cols. 
 46 47), are almost exactly the same as the Kharosthi sa and 
 show that the changes, assumed above, are easy and have 
 actually been made again in much later times. The signs of 
 the Papyri are again far advanced and unsuited for com- 
 parison. 
 
 No 15. The identity of pa with Phe is plain enough 
 (Thomas, Taylor, Halevy). The Semitic letter (Col. I) has been 
 turned round [58] in order to avoid the resemblance to A. The 
 form with a hook, attached to the right top of the vertical 
 (Col. Ill, a) occurs still a few times in the Mansehra version 
 of the Edicts. Usually the hook or curve is placed lower, as 
 in Col. Ill, b, and it may be noted that in the Mansehra pa
 
 106 Appendix I. 
 
 it is attached nearly always very high up, in the Shahbazgarhl 
 letter not rarely lower. 
 
 No 16. On phonetic grounds it may, of course, be ex- 
 pected that Tsade should have been used for the Indian ca. 
 But the recognition of the real Kharosthi representative has 
 been impeded by the circumstance that the earlier tables of 
 the alphabet neglect to give the form of ca, which comes clos- 
 est to the Semitic letter, viz. that with the angular head (Col. III). 
 The tables give only the ca with the semicircular top, though 
 the other form is by no means rare in the Edicts and is used 
 also in the cha (Col. IV) of the same documents and even 
 survives in the late Kharosthi inscriptions of the first and se- 
 cond centuries of our era. If the angular ca is chosen for 
 comparison, it is not difficult to explain how the Kharosthi 
 sign was developed. The Hindus made the top of the Tsade 
 (Col. I, a b) by itself, separating it from the remainder of 
 the vertical, and omitted in accordance with the principles of 
 their writing, which do not admit more than two strokes at 
 the tops of letters (see above p. 52) the small hook on the 
 right of the angle. Next, they placed the lower part of the 
 vertical under the point of the angle and in doing so added 
 a small flourish to the top of this line, which in course of 
 time became an important element of their sign. The Tsades 
 of the Papyri (Col. II) come very close to the Kharosthi ca and 
 the second even shows the small projection on the left, just 
 below the top. Nevertheless they are only independent an- 
 alogous developments. For in both, the long line on the left 
 has been made continuous with one stroke of the pen and the 
 hook or curve on the right has been added afterwards. More- 
 over, in the sign of Col. II, b, it is very plain that the small 
 projection on the left of the main line, which makes the letter 
 so very like the Kharosthi ca, has been caused by a careless 
 continuation of the right hand hook across the vertical. 
 
 [59] No 17. The utilisation of the ancient Qoph for the 
 expression of kha in the Brahma alphabet suggests the con- 
 jecture that the curious Kharosthi sign for kha may be derived 
 from the corresponding Aramaic character. And in the Serapeum 
 inscription the Qoph (Col. I) has a form which conies very 
 close to the Kharosthi kha. Only the upward stroke on the
 
 The Origin of the Kbarosthi Alphabet. 107 
 
 left is shorter and there is still a small remnant of the original 
 central line of the ancient North-Semitic character. The smaller 
 Teima inscription 1 (Euting, Col. 10) has a Qoph, in which the 
 central pendant has been attached to the lower end of the 
 curve (compare above the case of the Kharosthi ha). These 
 two forms, it seems to me, furnish sufficient grounds, for the 
 assumption, that in the earlier Aramaic writing the component 
 parts of the looped Qoph (Col. II, c) were disconnected and 
 arranged in a manner, which might lead to the still simpler 
 Kharosthi sign, where the central pendant seems to have been 
 added to the upstroke on the left in order to gain room for 
 the vowel-signs. To this conclusion points also the first cor- 
 responding sign of the Saqqarah inscription (Euting, Col. 11, a) 
 though the top has been less fully developed and the ancient 
 central pendant has been preserved much better. 2 
 
 No 18. Ra (Col. Ill) has been recognised as the repre- 
 sentative of Resh by all previous writers. But it deserves to 
 be noted that the sign, which comes nearest to the Kharosthi 
 letter is the character from Saqqarah, given in Col. I, b. 3 The 
 Papyri offer mostly more advanced forms with top lines sloping 
 downwards towards the right. 
 
 No 19. Regarding Shin (Col. I) and its Kharosthi counter- 
 part, the sign for the lingual sibilant *a (Col. Ill), see above 
 p. 100. I may add that round forms of Shin appear already 
 on the Babylonian Seals and Gems (Euting, Col. 8). 
 
 No 20. The oldest representatives of the Semitic Taw ap- 
 pear in the dental tha (Col. IV, a), which consists of the old 
 Assyrian [60] Aramaic Taw (Col. I, a) of the 8 th century B. C., 4 
 or of a slight modification of the very similar Saqqarah letter 
 (Col. I, b) (turned round from the right to the left) plus 
 the bar of aspiration on the right, about which more will be 
 said below, and in the lingual ta (Col. IV, b c), where the 
 second stroke on the right in b and on the left in c denotes 
 the organic difference or, as the Hindus would say, the diffcr- 
 
 1 Compare the end of 1. 1 of the facsimile in M. Ph. Berger's Histoire de 
 TEcriture, p. 217. 
 
 2 Compare also the sign from the Lion of Abydos, Euting, Col. 7. 
 
 3 Compare also Euting, Col. 7, b. 
 
 4 Ante p. 72.
 
 108 Appendix I. 
 
 ence in the Varga. In the second form of ta (Col. IV, c) the 
 bar, which originally stood at the side, has been added at the 
 top, and out of such a form the dental ta (Col. Ill) appears 
 to have been developed. Its top line has been lengthened con- 
 siderably and the downstroke has been shortened and bent in 
 order to avoid a collision with va and ra. The steps, which 
 led to its formation, are therefore (1) /" or -J*, (2) "^, (3) *J. 
 
 With respect to the Derivative Signs, my views are as follows. 
 
 (1) The aspiration is expressed by a curve, by a hook 
 or by a straight stroke, which latter, as the case of bha (No 2, Col. IV, 
 a b) shows, is a cursive substitute for the curve. At the same time 
 the original form of the unaspirated letters is sometimes slightly 
 modified. The curve appears on the right of the ga in gha 
 (No 3, Col. IV) at the top of da in dha (No 4, Col. IV, a) with- 
 out any change in the original forms. In bha (No 2, Col. IV, a) 
 it is attached to the right of ba ; the wavy top of which is 
 converted into a simple straight stroke, from the middle of 
 which the vertical line hangs down. The same sign shows also 
 frequently in the Asoka Edicts a hook for the curve and as 
 frequently a cursive straight stroke (No 2, Col. IV, b), slanting 
 downwards towards the right. The hook alone is found in tha 
 (No 20, Col. IV, d), 1 which has been derived from the preceding 
 form of ta (No 20, Col. IV, c) by the addition of a hook open- 
 ing upwards. The straight stroke alone is found, on the left 
 of the original letter and slanting downwards, in jha (No 7, 
 Col. IV), and likewise on the left but rising upwards, 2 in pha 
 (No 15, Col. IV). In tha (No 20, Col. IV, a) [61] the stroke of 
 aspiration appears on the right. It has the same position in 
 chha (No 16, Col. IV) and in dha (No 4, Col. IV, c). But in 
 the former sign the small slanting stroke at the top of the 
 vertical on the left has been straightened and combined with 
 the sign of aspiration into a bar across the vertical. In dha 
 the whole head of the unaspirated letter (No 4, Col. IV, b) has 
 been flattened down and reduced to a single stroke, which 
 together with the sign of aspiration forms the bar across the 
 top of the vertical. 
 
 1 The sign in the table is really tho. 
 
 2 There are also examples, in which the stroke is made straight.
 
 The origin of the Kharosthl Alphabet. 109 
 
 With respect to the origin of the mark of aspiration I 
 can only agree with J)r. Taylor, who explains it as a cursive 
 form of ha, The, Alphabet, vol. II, p. 260, note 1. The manner, 
 in which it was attached in each particular case, seems to have 
 been regulated merely by considerations of convenience and 
 the desire to produce easily distinguishable signs. The way in 
 which the hook or curve of aspiration has been used in the 
 Brahma alphabet is analogous. There too, it is added very ir- 
 regularly sometimes to the top, sometimes to the middle and more 
 frequently to the foot of the letters, where properly it ought 
 to stand. 1 If the Kharosthl characters never show it at the foot, 
 the cause is no doubt the desire to keep the lower ends of the 
 signs free from encumbrances, as has been noticed above p. 100. 
 
 The device for expressing the lingualisation in ta (No 20, 
 Col. IV, b c) and na (No 13, Col. IV, a) is very similar to that 
 sometimes used in the Brahma alphabet, in order to indicate 
 the change of the Varga or class of the letter. A straight 
 stroke, added originally on the right, serves this purpose in 
 the Bhattiprolu la, in the Brahma na, na and ha. 2 The case 
 of the Kharosthl ta has been stated above in the remarks on 
 the representatives of Taw. With respect to na it is sufficient 
 to point out that it has been developed from the na No 13, 
 Col. Ill, b, by a slight prolongation of the right hand stroke. 
 The case of the lingual da (No 4, Col. IV, b) is doubtful. Pos- 
 sibly it may be derived from an older dental da, like that 
 [62] in No 4, Col. I, a, by the addition of a short vertical straight 
 line on the right, which coalesced with the vertical of the da 
 and thus formed the sign with the open square at the head. 
 But it is also possible that the Aramaic alphabet, imported into 
 India, possessed several variants for Daleth, and that the heavier 
 one (No 4, Col. I, b) was chosen by the Hindus to express the 
 heavier lingual da, while the lighter or more cursive one was 
 utilised for the dental da. 
 
 The origin of the remaining two Kharosthl consonantic 
 signs, the palatal na (No 13, Col. IV, b, c) and of the Anusvara 
 in mam (No 12, Col. IV) has been already settled by Mr. E. 
 
 1 Ante p. 76 f. 
 a Ante p. 76.
 
 110 Appendix I. 
 
 Thomas. He has recognised that the palatal na consists of two den- 
 tal no,, joined together., and it may be added that in the Asoka 
 Edicts sometimes the right half and sometimes the left half is 
 only rudimentary, as shown by the two specimens given in 
 the Table. He has also asserted that the Anusvara is nothing 
 but a subscript small ma, which proposition is perfectly evident 
 in the form given in the table, less apparent, but not less true 
 in other cases, for which I must refer to Tafel I. 29. IV of 
 my Grundriss der indischen Palaeographie. 
 
 As regards, finally, the Kharosthl vowel system, and the 
 compound consonants (not given in the accompanying table) I 
 can only agree with Mr. E. Thomas, Professor A. Weber and 
 Sir A. Cunningham, that they have been elaborated with the 
 help of the Brahma alphabet. Among the vowel signs the 
 medial ones have been framed first and afterwards only the 
 initial /, U, E, (No 1, Col. IV, a d). They consist merely 
 of straight strokes, which (1) in the case of i go across the 
 left side of the upper or uppermost lines of the consonant, (2) 
 in the case of it slant away from the left side of the foot, (3) 
 in the case of e stand, slanting from the right to the left, on 
 the top line of the consonant (mostly on the left side) and (4) 
 in the case of o stand below the top line (compare iho, No 20, 
 Col. IV, d) or slant away from the upper half of the vertical 
 as in 0. The position of the four medial vowels thus closely 
 agrees with that of the corresponding signs of the Brahma al- 
 phabet, where i, e and o stand at the top of the consonants 
 and u at the foot. This circumstance [63] alone is sufficient 
 to raise the suspicion that there is a direct connexion between 
 the two systems of vowel-notation; see add. p. 124. And the 
 suspicion becomes stronger, if some further facts are taken into 
 consideration. In the Brahma alphabet of the Asoka Edicts 
 the medial e and u are mostly expressed by straight strokes. 
 The medial o, too, consists at least in one case, Delhi Sivalik 
 Pillar Edict, VII. 2, 1. 2 (nigohani) of a straight bar across the 
 top of the consonant, and has the same form frequently, on the 
 Persian sigloi (see below p. 113) and in the Bhatjjiprolu in- 
 scriptions. Again the medial i of the Girnar version is expressed 
 by a shallow curve, which in many instances is not distinguishable 
 from the medial a. Thus even the oldest Brahma documents
 
 The Origin of the Kharosthi Alphabet. Ill 
 
 furnish instances, in which all the four vowels, expressed in 
 the Kharosthi by straight strokes, have exactly the same form, 
 and it is very probable that in the ordinary writing of every 
 day life these cursive forms were in the case of o and i much 
 more frequent than the Edicts show, as well as that they go 
 back to earlier times than the third century B. C. If, finally, 
 the fact is added, that the Kharosthi, like the Brahml con- 
 siders the short a to be inherent in all consonants and does 
 not express it by any sign, it becomes difficult to avoid the 
 inference, drawn already by Professor Weber, that the Kha- 
 rosthi system of medial vowels has been borrowed from the 
 older alphabet. 
 
 The marking of the initial 7, U, E, (No 1, Col. IV, a d) 
 by A plus the corresponding medial vowel-sign is, of course, 
 an independent invention of the fraraer or framers of the Kha- 
 rosthi and probably due to a desire to simplify the more cum- 
 bersome system of the Brahml, which first developed the initial 
 vowels, next used them in combination with the consonants 
 and finally reduced their shapes in such combinations to simple 
 strokes and curves. 1 Similar attempts have been repeatedly 
 made on Indian ground. The modern Devanagarl has its ^ft 
 and ^ since the thirteenth or fourteenth century, the modern 
 Gujarat! has its E, AI, and AU, consisting of A plus the 
 medial vowel-signs, and [64] the Tibetan alphabet, framed out 
 of the Vartu letters of the seventh century A. D., expresses 
 even I and U by A plus i and u. These examples show 
 that the idea at all events came naturally to the Hindus and 
 that it is unnecessary to look for a foreign source of its origin. 
 
 The rules for the treatment of the compound consonants 
 again agree so fully with those of the Brahml, especially with 
 those adopted in the Girnar version, that they can only be 
 considered as copies of the latter. 
 
 (1) Double consonants like kka, tta, and groups of un- 
 aspirated consonants like kkha, ttha etc., are expressed by the 
 second element alone, except in the case of two nasals of the 
 same class, where "the first may be optionally expressed by the 
 Anusvara as in amna or ana. Three times, however, a double 
 
 Ante p. 77ft'.
 
 112 Appendix I. 
 
 ma is used in the word samma (samyak-pratipatti), Shahbaz- 
 garhi Ed. IX. 19, XL 23 and XIII. 5. 1 
 
 (2) Groups of dissimilar consonants are expressed by 
 ligatures of the signs, except if the first is a nasal, for which 
 the Anusvara is used throughout. 
 
 (3) In the ligatures the sign for the consonants, to be 
 pronounced first, stands above and the next is interlaced with 
 the lower end of the first, except in the case of groups with 
 ra, where ra is almost invariably placed below. 2 The forms 
 of the Kharosthl ligatures are shaped exactly like those of the 
 B rah mi and, like these, illustrations of the grammatical term 
 samyuktaksara "a conjunct syllable". The neglect of non- 
 aspirates, preceding aspirates, and of the double consonants, 
 with the exception of the nasals, which can be marked with- 
 out trouble by the Anusvara, is, as already pointed out, a 
 clerks' trick and the same as that used in the Brahmi lipi 3 
 of the Prakrit inscriptions. The treatment of ra in groups is 
 closely analogous to that adopted in Girnar, where this letter 
 or its cursive representative always occupies the same position, 
 whether it must be pronounced before or after the consonant 
 with which it is combined. There is, however, this [65] dif- 
 ference that in the Girnar Brahmi ra stand always at the top 
 and in the Kharosthl invariably at the foot. The one writes 
 e. g. via for rta and tva, and the other tra both for rta 
 and tra. 
 
 These remarks at all events suffice to show that a rational 
 derivation of the Kharosthl from the Aramaic of the Akhae- 
 menian Period, based on fixed principles, is perfectly possible 
 and the attempt has this advantage that it shows some letters, 
 as da, ka and ta, to be closely connected with Mesopotamian 
 forms, which a priori might be expected to have been used 
 by the writers of the Satraps, ruling over the extreme east 
 of the Persian empire. If the ruins of the eastern Persian provin- 
 ces are ever scientifically explored and ancient Aramaic inscrip- 
 
 1 In the third case Sir A. Cunningham's copy of the Shahbazgarln Edicts 
 in the Corpus Inscr. Ind., vol. I, PI. II, gives the mma correctly. 
 
 2 There is only one exception in the Mansehra version Ed. V, 24, kar^a- 
 bhikare. 
 
 3 Compare also the Grundriss, I, 11, 14.
 
 The Origin of the Kharosthi Alphabet. 113 
 
 tions are found there, forms much closer to the Kharosthi will 
 no doubt turn up. 
 
 The third and last point, the existence of which has been 
 indicated above, furnishes perhaps the most convincing proof 
 for Dr. Taylor's theory. It is simply this, that Mr. E. J. Rapson 
 has discovered of late on Persian silver sigloi, coming from 
 the Panjab, both Kharosthi and Brahma letters. Mr. Rapson 
 was good enough to show me specimens, belonging to the 
 British Museum, during my late visit to England, and I can 
 vouch for the correctness of his observation. I think, I can 
 do not better than quote his paragraph on the Persian coins 
 in India from the MS. of his contribution to Mr. Trubner's 
 Grundriss der Indo -Arischen Philologie und Altertumskunde, 
 which will appear in Vol. II, Section 3 B: 
 
 7. "During the period of the Achaemenid rule (c. 510331 
 B. C.) Persian coins circulated in the Panjab. Gold double 
 staters were actually struck in India, probably in the latter 
 half of the 4 th cent. B. C. [Babelon, LRS Perses Achemenides, 
 pp. IX, XX, 16, PI. II, 16-19; 27]. Many of the silver sigloi, 
 moreover, bear countermarks so similar to the native punch 
 marks 1 as to make it seem probable that the two classes of 
 coins were in circulation together; and this probability is in- 
 creased by the occurrence on sigloi, recently acquired by the 
 British Museum, of Brahma and Kharosthi letters." 2 
 
 [66] This appears to me sufficient to establish the con- 
 clusion that the Kharosthi did exist in India during the Akhae- 
 menian times and did not originate after the fall of the empire. 
 At the same time we learn that before 331 B. C. the Kha- 
 rosthi and the Brahma letters were used together in the Pan- 
 jab, just as was the case in the 3 rd and 2 nd centuries B. C. 
 
 In conclusion I may offer a suggestion regarding the name 
 of the script of Gandhara. The Buddhist tradition derives the 
 term Kharosthi from the name of its inventor who is said to 
 have been called Kharostha "Ass'-lip". I am ready to accept 
 this as true and historical, because the ancient Hindus have 
 
 1 Babelon, op. cit., p. XI attributes these countermarks to other provinces 
 of Asia. 
 
 2 Compare now his article in the Journal Royal Asiatic Society 1895, 
 p. 865 ff. 
 
 Buhler, Indian Studies. III. 8
 
 114 Appendix I. 
 
 very curious names apparently nicknames. 1 Thus we find al- 
 ready in the Vedas three men, called unahsepa, Sunahpuccha 
 and /Sunolaiigula "Dog's-tail", and Sunaka "Little-Dog" is the 
 progenitor of a very numerous race. Again a Kharljahgha 
 "She-Ass'-Leg"' is according to a Gana in Panini's Grammar 
 likewise the father of a tribe or family. 
 
 1 But compare now the interesting article of Professor A. Ludwig in the 
 Gurupujakaumudl p. 68. Professor A. Ludwig takes kharot.tha to be the 
 original form of the name and kharostha a false transliteration into 
 Sanskrit. And be considers kharottha to be the Indian representative 
 of a possible, though not actually found, Aramaic haruttha Kfimin 
 "engraving, writing 1 '. In favour of Professor A. Ludwig's view that 
 kharostha may be due to the false transliteration of a foreign word 
 it may urged that its derivative kharostht, more usually spelt kharostt 
 and even kharosti or kharostri (Mahavastu p. 135, 1. 5), occurs only in 
 Northern Buddhist texts and that these are full of similar blunders in 
 the conversion of Pali words into Sanskrit. If the actual and fre- 
 quent occurrence of haruttha should be proved hereafter, I should be 
 quite ready to give up my conjecture in favour of his. Regarding other 
 derivations of the name Kharostha see, A. Weber, Indische Streifen, 
 vol. Ill, p. 8f.
 
 Appendix II. 
 Tlie Origin of the ancient Brahma Numerals. 
 
 In my Grundriss der Indischen Palaographie, p. 77 f., I 
 have briefly discussed the various theories regarding the origin 
 of the ancient system of numeral notation found in the older 
 inscriptions and MSS. which are written in Brahma characters. 
 And I have expressed the opinion that in spite of serious difficulties 
 Dr. Burnell's suggestion, according to which it has been derived 
 from an Egyptian source, seems for the present more probable than 
 any of the other attempts at an explanation of its origin. As 
 Dr. Burnell's remarks on this point in his Elements of South 
 Indian Palaeography, p. 65 f., are only general and very brief, 
 a more detailed comparison of the two Egyptian systems which 
 can come into question, the hieratic and its immediate derivative, 
 the demotic, will not be superfluous. In order to facilitate it 
 the accompanying Plate III 1 exhibits in Col. I the most im- 
 portant forms of the hieratic numerals according to Professor 
 Euting's table, given under No LXXIV in the Palaeographic So- 
 ciety's Oriental Series, in Col. II the demotic signs according to 
 the plates, appended to Professor Brugsch's Deinotische Gram- 
 matik, and in Cols III IV the Indian numerals according to 
 the originals from which Cols I IV of Tafel IX of my Grund- 
 riss have been compiled. Among the figures for 6, Col. Ill, c, 
 shows however an additional sign, that of the Sahasrain Edict, 
 which has been omitted in the Grundriss. 
 
 The resemblance of the principles of the Egyptian and 
 Indian systems is, if not complete, at least very great. Both 
 
 1 Drawn by Mr. Bo'hm and etched by Messrs Angerer and GOschl of 
 
 Vienna. 
 
 8*
 
 116 Appendix II. 
 
 have separate signs for 19, 1090, 100 and 1000. In both 
 the numerals 1 3 are expressed by strokes, either vertical or 
 horizontal 1 or by combinations thereof and the symbols begin 
 with 4. In both the signs for 200 and 300 consist of that for 
 100, united respectively with the signs for one and two. 2 The 
 not very important differences are (1) that the Indians form 
 2000 and 3000 on the same principle as 200 and 300, while 
 the Egyptians add in these cases the signs for two and three 
 to the symbol for 1000; (2) that the Indians form 400-900 
 and 40009000 by the combinations of 100 with 49 and of 
 1000 with 49, while the Egyptians have at least for 400 the 
 combination 100 + 3; (3) and that the Egyptians have a sepa- 
 rate sign for 10,000 which is wanting in the Indian system. 
 The first two discrepancies may possibly be due to the for- 
 malism of the Hindus, which may have induced them to reject 
 the vacillation in the Egyptian system and to rigorously ad- 
 here to one and the same principle. 
 
 If we turn to the comparison of the several symbols 
 beginning with 4, it is evident that the oldest Indian forms 
 of eight out of the total of seventeen come close to the 
 hieratic or to both the hieratic and the demotic. It is also 
 evident that those differences, which are observable, are due 
 either to the desire to simplify the cumbersome Egyptian 
 signs or to the tendency to transform them into Brahma letters 
 and syllables. 3 The five of Col. V, a, appears to be derived 
 from the hieratic sign of Col. I, which has been turned from 
 the left to the right, while the side-limb has been attached 
 lower in order to obtain an approach to the Brahma tra. The 
 six of Col. Ill, a, likewise comes close to that of Col. I, the 
 slight alteration being made in order to obtain the looped 
 Brahma ja, from which latter the signs in Col. Ill, b c, and in 
 Col. V, a looped sa and phra or phu y seem to have been de- 
 veloped in their turn. The seven of Cols IV V looks like 
 
 1 Horizontal strokes occur in the Egyptian signs for 2 and 3, referring 
 to the days of the month. 
 
 2 The strokes no doubt represent an original repetition of the symbols 
 for 100, and it is as if the Romans wrote CI and CH for CC and CCC. 
 
 3 Regarding this tendency and its ultimate effects, see the Grtindriss, 
 p. 75 ff.
 
 The Origin of the ancient Brahma Numerals. 117 
 
 a modification of that in Col. I, a, made in order to obtain 
 a sign similar to gra or gu, the later Indian letter-symbol. 
 The nine of Cols IV V agrees almost exactly with the hier- 
 atic and demotic signs in Cols I II. The fifty of Col. Ill, b, 
 is produced by a slight simplification of the symbols in Cols 
 I II, while the fifty of Col. Ill, a, has been turned round from 
 the left to the right. More strongly modified and more re- 
 mote from the hieratic signs are the Indian symbols for 60 
 and 70 in Col. V, the second of which has been converted into 
 a Brahma p il. Finally the thousand of Cols IV V, especially 
 the second, look like derivatives from the corresponding Egyp- 
 tian signs in Cols I II, which however have been turned topsy- 
 turvy and slightly altered in order to produce the Brahma syl- 
 lables ro, cu and dhu. 
 
 The list of closely similar signs will also include the hun- 
 dred, if it may be assumed that the Indian signs of Col. V, 
 which look like modifications of the hieratic in Col. I, in reality 
 are older than those of the Afioka Edicts and of the Nanaghat 
 inscription in Cols III IV. And it seems to me possible to 
 defend such an assumption. For the Andhra and Ksatrapa 
 signs for 100, which are also found very frequently in the 
 Sanskrit inscriptions of the 5 th 8 th cent, A. D. from Western 
 India as well as in the Kalinga landgrants of the 8 th cent., 
 in all probability represent the syllable hi. 1 Its sa has a form 
 which corresponds to the oldest Semitic Shin 2 with two angles 
 but has been turned topsy-turvy, and which is older than the &* of 
 the third century B. C. 3 Further, with the assumption that the 
 hundred of Col. V is really the oldest, it becomes possible to 
 explain the origin of those in Col. Ill IV. As the Asoka 
 sign opposite 200 4 in Col. Ill, a, clearly shows, they represent the 
 Brahma syllable su, and su is in Prakrit the regular represcn- 
 
 1 See the Grundriss, p. 77 and note 20, where two inscriptions are referred to, 
 in which more modern forms of Su or Sa appear in the hundred. 
 
 2 See above Plate I, 21, Cols I II. 
 
 3 See above Plate I, 21, Col. V, as well as Tafel II, 37, I II, XI etc. of 
 the Grundriss, and compare the remark made above p. 71. 
 
 4 Equivalent to the ordinary su with an unusual elongation of the vertical, 
 which seems to do duty for the additional horizontal stroke, commonly 
 found in 200 on the right side.
 
 118 Appendix II. 
 
 tative of a Sanskrit u. The process of the development of the 
 Indian hundreds would therefore be as follows. The Hindus 
 first converted the hieratic hundred into an archaic su (Col. V). 
 Next they substituted its phonetic Prakrit representative su 
 for it (200, Col. HI, a) and finally they again modified the form of 
 su in order to indicate that it has not a phonetic but a nu- 
 meral value (200, Col. Ill, b c; 100700, Col. IV). 
 
 With the admission of another conjecture the cases of 
 close resemblance will extend to a tenth sign. It is undeniable 
 that the Indian sign for 40 in Col. V is almost exactly identical 
 with the Egyptian symbols for 30 in Cols I II. As the Indian 
 30 shows a remote similarity to the Egyptian 40 7 it is tempting 
 to assume that the Hindus made the signs for 30 and 40 ex- 
 change their places, be it through negligence or for some 
 other reason. 
 
 The remaining signs differ much more. But it is quite 
 imaginable that the four and eight of Col. I may have be con- 
 verted respectively into the ka of Col. Ill l and the hra or hu 
 of Col. V, that the ten of Col. I, being turned from the left to 
 the right, became the thu of Cols IV V, and that the twenty 
 of Col. I by a considerable simplification yielded the tha of 
 Col. V. 2 The most difficult signs are, as Dr. Burnell has also 
 recognised, those for 80 and 90. There is no actual resem- 
 blance between the Egyptian and the Indian forms. But it 
 deserves to be noted that in both the hieratic and the Indian 
 systems they apparently form a separate group, in which the 
 higher numeral is differentiated from the lower one by the ad- 
 dition of a second line, standing in the hieratic sign below 
 and running in the Indian ninety horizontally across the middle. 
 
 1 The four of Col. V. (compare also those of 400 and 4000 in Col. IV), 
 which may be read ki and pka, are in my opinion later developments 
 from the simple ka. The apparent vowel or pa -mark at the top is 
 probably merely formed out of a flourish, which originally indicated 
 that the letter had not a phonetic value; compare the late rnka, nkd 
 in Tafel IX, Col. XXIV -XXV of the Grundriss, where a similar striving 
 at differentiation is perceptible, and the remarks in the Grund- 
 riss, I, 11, p. 75. 
 
 2 On the supposition that this is the older form, the tha of Col. IV in 
 20,000 might be regarded as a phonetic substitute.
 
 The Origin of the ancient Brahma Numerals. 119 
 
 The results of this comparison, the nearly complete identity 
 of the principles of the two systems and the closer or remoter 
 resemblances between the great majority of their symbols, raise, 
 it seems to me, a strong presumption in favour of Dr. Bumell's 
 theory, which, however, has to be modified by the admission 
 that the hieratic signs rather than the demotic are the parents 
 of the Indian Brahma numerals and that the tendency to frame 
 out of them Brahma letters or syllables and again to differentiate 
 these from the ordinary letters is already visible in a number 
 of cases. And I believe that this tendency is due to the fact 
 that the Hindus since ancient times were accustomed to express 
 numerals by such words as rupa ='1, bhuta = 5 and so forth. 1 
 In order to convert Dr. Burnett's theory into an absolute cer- 
 tainty, the discovery of more ancient signs of the third and earlier 
 centuries B. C. is required as well as the discovery of historical or 
 traditional information regarding an ancient intercourse between 
 India and Egypt. For the present this is entirely wanting and 
 the only way, in which one could try to account for the in- 
 troduction of the Egyptian numerals into India, would be to 
 conjecture that the early Indian navigators and merchants might 
 have reached countries which stood under Egyptian influence 
 or might have met on their voyages with Egyptian traders. But 
 such a conjecture is, of course, precarious, as long as it is not 
 supported by collateral proof. 
 
 See the Grundriss, I, 11, p. 80 ff.
 
 Addend a. 
 
 P. 9, 1. 28. Another case of a letter of defiance (panna) 
 sent to a king by his adversaries is mentioned in Jataka, 
 No. 23, Fausboll, vol. I, p. 178 (Rhys Davids). 
 
 P. 10, 1. 12. Promissory notes (panna) and their redemption 
 by payment are mentioned in the introduction to Jataka No 40, 
 Fausboll, vol.1, p. 227, 1. 3f., and p. 230, 1. Iff. (Rhys Davids). 
 
 P. 11,1. 14. According to Jataka No 388 (Fausboll, vol. Ill, 
 p. 292) the future Buddha caused the judicial rules to be written in 
 a book (pot thaka) and recommended that, in future, cases should be 
 decided, after reference to it. A book (potthqka) is mentioned also in 
 the Uddalaka Jataka No 487, Fausboll, vol. IV, p. 487, 1. 17 f, where 
 Uddalaka's preparations, intended to captivate the king, are de- 
 scribed as follows: "But he himself took with him eight or ten 
 disputant Pandits, placed a lovely book (potthaka) on a char- 
 ming reading desk (adhdraka) and sat down, surrounded 
 by his pupils, on a prepared seat which was furnished with 
 a support for the back 1 '. Here the book can only be a MS. 
 of a literary work, and the reading desk is probably one of 
 those little tripods, consisting of three laquered pieces of wood, 
 which are commonly used by the Jaina Yatis at their vdcands. 
 The whole description is merely an accessory without any im- 
 portance for the story and may be an interpolation, as its 
 rather modern look suggests. 
 
 P. 31, 1. 5. From Mr. Halhed's Grammar of the Bengal 
 language, p. 17 ff. (Calcutta, 1778), to which Dr. B. Liebich 
 has drawn my attention, it appears that about the end of the 
 last century the Bengal schoolmasters still used a set of twelve 
 tables, including chiefly compound tetters, which may be re- 
 motely connected with Hiuen Tsiang's twelve chang and Huilin's
 
 Addenda. 
 
 121 
 
 twelve fan. Mr. Halhed calls them the twelve Phola, which 
 name according to the Jonesian transliteration would be Phala 
 and corresponds to the Sanskrit phalaka, 'board, tablet', and 
 he states that they severally are kya, kra, kna, kla, kva, kma, 
 kr, kl, rka, hka, ska and siddhi. He adds that siddhi is an 
 abbreviation of siddhir astu 1 and that this Phala consists of the 
 sixteen vowels. Mr. Halhed also gives the contents of two 
 other Phalas in full, which I copy here in transcription, as his 
 grammar is rather rare. 
 
 The Phala nka includes: 
 
 nka 
 
 nkha 
 
 nga 
 
 ngha 
 
 nna 
 
 nca 
 
 ncha 
 
 nja 
 
 njha 
 
 nna 
 
 nta 
 
 ntha 
 
 nda 
 
 ndha 
 
 nna 
 
 nta 
 
 ntha 
 
 nda 
 
 ndha 
 
 nna 
 
 mpa 
 
 mpha 
 
 mba 
 
 mbha 
 
 mma 
 
 nya 
 
 ra 
 
 iila 
 
 nva 
 
 
 
 Ma 
 
 nsa 
 
 nsa 
 
 hha 
 
 nksa } or 34 signs, 
 
 and Phala ska: 
 
 
 
 
 
 ska 
 
 skha 
 
 dga 
 
 dgha 
 
 Mr a 
 
 ca 
 
 scha 
 
 bja 
 
 bjha 
 
 jiia 
 
 sta 
 
 stha 
 
 Ida 
 
 bdha 
 
 hna 
 
 sta 
 
 stha 
 
 bda 
 
 bdha 
 
 hna 
 
 spa 
 
 spha 
 
 dba 
 
 dbha 
 
 hma 
 
 hya 
 
 ra 
 
 hla 
 
 hva 
 
 
 
 tsa 
 
 tsa 
 
 tsa 
 
 tha 
 
 tksa, 
 
 or again 34 signs. 
 
 Though it is evident that the twelve Phala are not identical 
 with Hiulin's twelve fan, Mr. Halhed's statements show at all 
 events that a syllabary in twelve sections was actually used 
 in India, at least in Bengal. Probably the division is an old 
 one and has been taken over from one compilation into the 
 other, though the tables underwent changes in other respects. 
 
 This is a common Mangala in Sanskrit books and inscriptions, and 
 stated by Itsing (A Record of the Buddhist Religion, p. 170, Takakusu) 
 to be a name of the whole Indian syllabary. An other work quoted by 
 Mr. Takakusu, (op. cit. p. 171 note) the Siddhakosha of A. D. 880, gives 
 Siddhdm (sic) as the name of the sixteen vowels.
 
 122 Addenda. 
 
 Full certainty on the exact character of the Indian syllabaries 
 and their history can only be obtained by an examination of the 
 various works, called lipiviveka, mat rkaviveka , varnamala 
 and so forth, mentioned in the publications, connected with 
 the Search for Sanskrit MSS., and in the Catalogues of European 
 libraries. The kindness of Dr. Grierson, who made enquiries 
 regarding the actual occurrence of Huilin's twelve tables through 
 Mr. Prothero, Educational Inspector of Patna, enables to 
 add, that the indigenous schools of Bengal proper still teach 
 the compound letters, but do not teach the combinations of 
 the vowels with the ligatures. But it appears that this is not 
 the case in Bihar, where only the common cursive Kaethl 
 (Kdyastha-lipi) is taught. There, as one of the Deputy In- 
 spectors, Mr. Soinnath Jharkhundi says, the instruction in the 
 varnamala stops, when the boys have mastered the vowels A, 
 A ... Ah, the consonants ka, kha . . . ha, and the Bartarii 
 ka, ka . . . kau, kam, kah and so forth, which latter corresponds 
 exactly to the Barakhadl of Western India. Further enquiries 
 in Gujarat have only had a negative result and have elicited 
 the answer that no tables for compound consonants are known. 
 
 Regarding another more extensive syllabary in 18 tables, 
 see Itsing, op. cit, p. LXff. and p. 170ff. (Takakusu). 
 
 P. 32, 1. 6. Read a round-topped ga on No 14 instead of 
 on No 13. 
 
 P. 38, 1. 16. For plate V, No 2 read plate X, No 2. 
 
 P. 91, 1. 13. The kindness of Dr. S. von Oldenburg has 
 put me in possession of a photograph of a portion of the Kha- 
 rosthl MS. brought by M. Petroffski from Khotan, and he has 
 generously given me permission to say here a few words about 
 it. The photograph contains a part of a new version of the 
 bramhanavagga of the Dhammapada in a dialect of North- 
 western India, in many respects closely connected with that of 
 the Shahbazgarhl rock edicts of Asoka, but in some respects 
 more advanced in the direction of the later Prakrits. With the 
 Asoka Edicts the language of the MS. shares the clear distinc- 
 tion between the three Sanskrit sibilants, the preservation of 
 ra in groups, the substitution of o for final atn and various 
 characteristic forms like bramana (brammana) for brdhmana 
 or bramhana. Advanced forms are e. g. niha'i (nihai) for
 
 Addenda. 123 
 
 nidhaya, dana (danna) for danda, china (chinna) for chinda, 
 with which latter Pali forms like punnarlka for pundarlka may 
 be compared. 
 
 The principles of its spelling in no way differ from those 
 of the epigraphic Kharosthl. The long vowels are not marked, 
 the Anusvara does duty for all nasals in compound consonants, 
 single non-aspirate consonants stand for double ones, and before 
 aspirates the corresponding non-aspirate is left aut. The Bud- 
 dhists of the Northwest, therefore, found it unnecessary to have 
 recourse to the more accurate spelling of the Pandits, which 
 their brethren in the South have adopted. 
 
 The characters show the higly cursive type of the Kusana 
 period and come very close to those on the Wardak vase, in- 
 cised during the reign of Huviska. The wavy lines in the 
 side-limbs of ka, pa and bha are however more fully developed. 
 On the other hand some very cursive forms of ka and bha, 
 in which the side-limbs of these letters are attached to the top- 
 lines are wanting in the thirty lines accessible to me. The 
 Padas of the Slokas carefully divided and at the end of each 
 Sloka, which almost invariably fills a line, stands a mark re- 
 sembling the circular Indian nought, the first sign of inter- 
 punctuation found in a Kharosthl document. 
 
 With respect to the exact probable date of the MS., it is 
 difficult to say anything certain, as the question when the fa- 
 mous three Kusana kings reigned is not yet settled. If M. S. 
 Levi (Journ. Asiatique, 1897, p. 1 ff.) is right in contending that 
 even Bazdeo -Vasudeva belongs to the middle of the first cen- 
 tury, the Petroffski MS. cannot be much later than the begin- 
 ning of our era. But, though I fully agree with M. Levi in 
 believing that Kaniska is not the founder of the feaka era, 1 
 I do not yet venture to give an unqualified adhesion to his 
 theory, the full proof for which has still to be furnished. For 
 the present I would only say this much that the Petroffski 
 MS. is certainly the oldest Indian MS., yet discovered, and that 
 it perhaps was written in the first century A. D. and certainly 
 
 See my remarks on this subject and on the reasons why for practical 
 purposes I have adopted the theory of Messrs Fergusson and Oldenberg, 
 in the WZKM. I, 169.
 
 124 Addenda. 
 
 not much later. I believe that, like the Bower MS., it was 
 written in India as the dialect and the characters suggest, in 
 Gandhara and that it was carried by some Buddhist monk 
 into Chinese Turkestan. 
 
 P. 110. And it ought to be noted, that while no good 
 reason is apparent why u should have been placed at the foot 
 in the Kharosthl, it is patent that the sprawling / and /_ of the 
 Brahmi was not fit to be placed at the top of the consonants, 
 but better suited the position at the foot, where, in letters not 
 ending in verticals, it could be abbreviated, without the risk of 
 confusion with e, a and ai.
 
 Plate I. 
 
 Comparative Table 
 
 of 
 
 the oldest Semitic and the Brahma Alphabets. 
 
 g: Is Asoka, Bkattiprolu and cognate Inscriptions 
 
 I 1 
 
 J3 Original letters Derivatives 
 
 
 I 
 
 D 
 
 m 
 
 IV 
 
 v 
 
 VI 
 
 1 
 
 < 
 
 < 
 
 
 
 
 
 >r 
 
 2 
 
 <9 
 
 5 
 
 
 *0 
 
 ODD 
 
 rf 
 
 3 
 
 A 
 
 ^ 
 
 
 
 A 
 
 t 
 
 4 
 
 ^ 
 
 ^ 
 
 
 
 a D 
 
 O C 1 d 4 > *< r" 
 
 5 
 
 3 
 
 5j 
 
 AA 
 
 
 L U U 
 
 
 6 
 
 
 Y 
 
 
 
 AA 
 
 L 9D LQ 11 f ? 
 
 7 
 
 I 
 
 1C 
 
 
 
 C P f 
 
 /^ 
 
 8 
 
 B 
 
 N 
 
 
 *m 
 
 L L 
 
 
 9 
 
 
 
 
 ^ 
 
 
 
 
 OC 
 
 10 
 
 f 
 
 ^ 
 
 
 *d 1 
 
 JLi J/ 
 
 
 11 
 
 y 
 
 y 
 
 
 
 t _f_ 
 
 
 12 
 
 J 
 
 ^T 
 
 
 ^ 
 
 JA 
 
 b 
 
 13 
 
 "1 
 
 tX 
 
 
 *t> 
 
 ^ ^ 
 
 *o . 
 
 14 
 
 9 
 
 3 
 
 
 *'\ 
 
 
 
 "h I 
 
 IS 
 
 ^ 
 
 $ 
 
 '^ 
 
 ^ 
 
 
 
 if 1-i Jb 
 
 16 
 
 o 
 
 
 
 
 
 
 DQ<0^{ 
 
 : : /.tff :: 
 
 17 
 
 j? 
 
 P 
 
 
 ^ 
 
 L 
 
 b 
 
 18 
 
 h- 
 
 f^ 
 
 
 *b 
 
 d d 4 
 
 cb d> 
 
 19 
 
 9 
 
 9 
 
 
 
 2 1 
 
 
 20 
 
 ^ 
 
 <1 
 
 
 
 s;vn 
 
 
 21 
 
 w 
 
 w 
 
 V^ 
 
 
 A^ X\/Js 
 
 
 22 
 
 t 
 
 X 
 
 / 
 
 
 AAA 

 
 Plate II. 
 
 Comparative Table 
 
 of 
 
 the Perso- Aramaic and the Kharosthi. 
 
 Kharosthl 
 
 Inscrip- Borrowed 
 tions Pa P? n Letters Derivatives 
 
 
 I 
 
 II 
 
 in 
 
 IV 
 
 1 
 
 f H 
 
 ^ 
 
 7 
 
 <n 71 
 
 2 
 
 J# 
 
 2?J 
 
 7 
 
 % x 
 
 3 
 
 A 
 
 ^ 
 
 f 
 
 ^ 
 
 4 
 
 > H 
 
 *i>v 
 
 5 ^ 
 
 7 ^ 7 
 
 5 
 
 1\ "A 
 
 JH\ 
 
 1 \ 
 
 
 6 
 
 7 
 
 7)7 
 
 ~] 
 
 
 7 
 
 fc ^ 
 
 w 
 
 !t T( 
 
 y 
 
 8 
 
 v^rl 
 
 Krw 
 
 n 
 
 
 9 
 
 a- ^ 
 
 A4>S 
 
 A 
 
 
 10 
 
 7 ? 
 
 w 
 
 7 
 
 
 11 
 
 at 
 
 u 
 
 7 
 
 
 12 
 
 ^ 
 
 ^ 
 
 *- V ^ 
 
 L 
 
 13 
 
 ;; 
 
 ^ij 
 
 5 T 
 
 P V V 
 
 14 
 
 *i 
 
 t^^ 
 
 *7 
 
 
 15 
 
 j 
 
 ?1^ 
 
 r ^ 
 
 ?i 
 
 16 
 
 
 
 ^ 
 
 T 
 
 f 
 
 17 
 
 "bV, 
 
 />>> 
 
 <^ 
 
 
 18 
 
 i-T 
 
 y 
 
 "7 
 
 
 19 
 
 1^ 
 
 ^^ 
 
 T 
 
 
 20 
 
 A / 
 
 /y 
 
 7 
 
 t f-=n
 
 Plate III. 
 
 Comparative Table 
 
 of 
 
 Egyptian Indian Numerals. 
 
 
 Hieratic 
 
 Demotic 
 
 %*t* 
 
 ~~ 
 
 l^3 
 <jj W 
 
 "S 
 
 CO 
 
 Naraghat 
 inscriptioi 
 2nd cent. B. 
 
 Jill 
 
 Wl 
 
 
 I 
 
 II 
 
 III 
 
 IV 
 
 v 
 
 1 
 
 J 
 
 1 
 
 
 
 
 
 
 2 
 
 *i 
 
 "I 
 
 
 : 
 
 II 
 
 3 
 
 K\ 
 
 Uj 
 
 
 
 J5 
 
 
 4 
 
 yiin 
 
 + 
 
 
 f t 
 
 
 1 
 
 i 
 
 
 
 h 'i 
 
 
 t 
 
 ^ 
 
 ^ * <r 
 
 
 *f 
 
 
 1 t\ 
 
 "^^j 
 
 
 1 
 
 1 
 
 
 **>*=> 
 
 -2- 
 
 
 
 h y 
 
 
 ^ 
 
 ^f 
 
 
 P 
 
 ^ 
 
 10 
 
 / 
 
 /x 
 
 
 c* 
 
 ex: o< 
 
 20 
 
 v\ 
 
 /-^ 
 
 
 
 
 
 30 
 
 X 
 
 j: 
 
 
 
 rv7 
 
 40 
 
 x-- 
 
 / 
 
 
 
 y, 
 
 SO 
 
 1 
 
 \ 
 
 G 3 
 
 
 
 60 
 
 UL 
 
 "/ 
 
 
 
 j 
 
 70 
 
 ^ 
 
 1 
 
 
 
 $ 
 
 .80 
 
 im 
 
 ^ 
 
 
 CD 
 
 
 
 90 
 
 A 
 
 \ 
 
 
 
 
 
 100 
 
 _^> 
 
 ^4 
 
 
 ^ 
 
 1 3 
 
 200 
 
 
 ^ 
 
 ^ T-P^ 
 
 K 
 
 T 
 
 300 
 
 _Ji* 
 
 ^J2 
 
 
 If 
 
 
 400 
 
 j}5 
 
 uu 
 
 
 Kf 
 
 
 500 
 
 J2 
 
 fl 
 
 
 
 !7V 
 
 700 
 
 
 ^3r 
 
 
 ?n 
 
 
 tooo 
 
 
 
 ^ 
 
 
 T 
 
 1 
 
 2000 
 
 >* 
 
 J 
 
 
 
 T 
 
 3000 
 
 jj^ 
 
 j*j 
 
 
 
 T 
 
 4000 
 
 -^ 
 
 - ^i 
 
 
 Tf 
 
 f 1 
 
 20000 
 
 
 
 
 lb 
 
 
 70000 
 
 
 
 
 
 ^
 
 A u s z u g 
 
 aus dem 
 
 Verlagskatalog von Karl J. Triibner in Strassburg. 
 IV. (fcrientattfrfje f 6itoto0te. 
 
 Andreas, F. K., Iranische Schriftkuncle siche: Grundriss der iranischen 
 rhilologie. 
 
 Bacher, Dr. Wilh. (Professor an der Landes-Rabbinerschule zu Budapest)* 
 Die Agada der Babylonischen Amoraer. Ein Beitrag zur Geschichte 
 der Agada und zur Einleitung in den Babylonischen Talmud, gr. 8. 
 XVI, 151 S. 1878. ^/4 4 
 
 Fur Alle, welche sich mit der Litteratur- und Culturgeschichte der Juden vom Beginn 
 des dritten nachchristliohen Jahrhunderts bis zum Ende des Alterthums bcschaftieen 
 wollen, ist Bachers Sc-hrift ein unentbchrliches Hulfsmittel. Theolog. Litteraturztg. 1879 3. 
 
 Man vergleiche auch die Rezension im Literarischen Centralblatt 1879, Nr. 15. 
 
 te 2lgaba bet Xannaiten. 
 
 I. SBanb: 35on -riCel bi 5lfi6a. SSon 30 tor fct 135 ttacfi b g. 3 8 
 
 457 @. 1884. (ji s ) Sergriffen. 
 
 II. SBonb: Son TOa ob Bt3 3"m 5lbf^IuB bet SWif^no. (135 bi 
 
 220 nati) ber getoSljttl. ^eitredjn.) 8. VIII, 578 . 1889. Ji 10 
 S)te 2Igaba bet potafttnenfifc^en 3lmotaet. 
 
 I. SBanb: Sont 3l6jc^tu bet 3Jlifdjna Bi3 sum lobe Sorfjartani. (220 6i8 
 
 279 narf) bet geto. ^eittet^nitng.) 8. XVI, 587 . 1892. Jt, 10 
 
 II. 33anb: ie ^iilet ^oii)anan. gt. 8. VI, 545 @. 1896. UZ 10 
 Abraham Ibn Esra als Grammatiker. Ein Beitrag zur Geschichte 
 
 der Hebraischen Sprachwissenschaft. 8. 192 S. 1882. Jl. 4 
 
 Die Bibelexegese der jiidischen Religionsphilosophen des 
 
 Mittelalters vor Maimuni. gr. 8. VIII, 156 S. 1892. M 4 
 Varianten zu Abraham Ibn Esra's Pentateuch-Commentar aus dem 
 
 Cod. Cambridge. Nr. 46. 8. IV, 108 S. 1894. (In hebr. Sprache). M. 3 - 
 
 DieBibelexegeseMosesMaimuni's. 8. XVI,176S. 1897. M. 4 
 
 siehe auch : Sa'di's Aphorismen. 
 Baines, A., Indian Ethnography (Grundriss der indo-arischen Philologie 
 
 und Altertumskunde II. Band, 5. Heft.) (In Vorbereitung.) 
 
 Bartholomae, Chr., Altiranisches Worterbuch. (In Vorbereitung.) 
 
 Vorgeschichte der iranischen Sprachen. Awestasprache und Altpersisch. 
 
 Siehe : Grundriss der iranischen Philologie. 
 Benfey, Theodor, Vedica und Verwandtes. kl. 8. 177 S. 1877. Jt 6 
 
 Inhalt: I. ri bezeichnet in den Veden sowol den kurzen als langen Vokal. II. Rig- 
 veda X. 170, = Atharvaveda XVIII. 1,8. III. Nediyams nedishtha. IV. 1st Rigveda VII. 
 44,3 ma^ccator oder ma^gcator in der Samhita zu lesen? V. Ist Rigveda III. 53,19 
 spandane oder syandane, Rigveda IV. 3,20 aspandamano oder asyandamano zu lesen? 
 VI. Wie kam der Verfasser der Isten Varttica zu Panini VII. 3,87 dazu, eine Wurzel 
 spa<; mit langem a anzunehmen? VII. <;vanin oder cjvani'. VIII. jajhjhhatis Rigveda. 
 V. 52,6. IX. Ze'j; TcX^wv. X. Karbara oder Karvara gefleckt, scheckig. Indoger- 
 manische Bezeichnung der dem Beherrscher der Todten gehorigen Hunde. IX. Wahrung 
 meines Rechtes (betreffend die Entdeckung der urspriinglichen Stelle des Accents im 
 Indogermanischen). Index. 
 
 Prof. Th. Benfey has just published, under the title Vedica und Verwandtes a series 
 }f a number of very nice and subtle question 
 
 t terms in the Vedas and exhibiting fully the av 
 learning and critical acumen. Academy No. 267, Juni 16, 1877. 
 
 of papers mainly of a number of very nice and subtle questions of verbal criticism and 
 explanation of different terms in the Vedas and exhibiting fully the authors profound 
 
 Vedica u. Linguistica. kl. 8. 278 S. 1880. (Ji 10 50) 
 
 (In den Verlag von Kegan Paul, Trench, Triibner & Co. in London ubergegangen.) 
 
 Bhandarkar, R. G., The Path of Devotion or Bhaktimarga, the 
 Vaisnava, Saiva, Sakta and Saura Sects. (Grundriss der indo- 
 arischen Philologie und Altertumskunde III. Band, 6. Heft). 
 
 (In Vorbereitung.)
 
 16 VERLAGSKATALOG von KARL J. TRUBNER in Strassburg. 
 
 Bloomfield, M., Vedic Literature (Sruti) : Atharvaveda (Grundriss der 
 indo-arischen Philologie und Altertumsknnde II. Band, Heft Ib). 
 
 (In Vorbereitung.) 
 
 Brockelmann, Carl, Das Verhaltniss von Ibn-El-Atirs Kamil Fit- 
 Ta'rihzuTabarisAhbarErrusulWalmuluk. (Diss.) 8. 58 S. 1890. 
 
 Jt 1 80 
 
 Das Buch von der Erkenntniss der Wahrheit oder der Ursache aller 
 Ursachen. Aus dem syrischen Grundtext ins Deutsche iibersetzt von 
 Karl Kayser, Licentiat der Theologie und Pastor. 8. XXIII, 367 S. 
 1893. (Nur in 200 Exemplaren gedruckt.) Jl. 15 
 
 Biihler, Georg, Indische Palaographie von circa 350 a. Chr. circa 
 1300 p. Chr. (Grundriss der indo-arischen Philologie und Altertumskunde 
 I. Band 11. Heft). Lex. 8. IV, 96 S. mit 17 lithogr. Tafeln in besonderer 
 Mappe. 1896. JK 18 50 
 
 s jetzt nicht zusammenhangend bearbeitet 
 
 - The Sources of Indian History : Literature and Inscriptions 
 (Grundriss der indo-arischen Philologie und Altertumskunde II. Band, 
 3. Heft a). (In Vorbereitung.) 
 
 - Political History of India from the earliest times to the Mahom- 
 medan conquest with a chapter on chronology (Grundriss der indo- 
 arischen Philologie und Altertumskunde II. Band, 9. Heft). 
 
 (In Vorbereitung.) 
 
 Indian Studies, No. III. On the Origin of the Indian Brahma 
 Alphabet. Second Edition, revised. Together with two Appendices on 
 the Origin of the Kharosthi Alphabet and of the so-called Letter-Nume- 
 rals of the Brahmi. With three plates. Gr. 8. XIII, 124 S. 1898. 
 
 < und M. A. Stein, Indische Geographic (Grundriss der indo-arischen 
 
 Philologie und Altertumskunde II. Band, 4. Heft). (In Vorbereitung.) 
 
 J. Jolly and Sir R. West, Sociology, clans, castes, consti- 
 tution of villages and towns, forms of government and ad- 
 ministration. Economics, tenures, commerce and banking, 
 handicrafts (Grundriss der indo-arischen Philologie und Altertumskunde 
 II. Band, 6. und 7. Heft). (In Vorbereitung.) 
 
 Burgess, J., Architecture, Sculpture and Painting in India (Grund- 
 riss der indo-arischen Philologie und Altertumskunde III. Band, 11. Heft). 
 
 (In Vorbereitung.) 
 
 Cappeller, Carl (Professor des Sanskrit an der Universitat Jena), Sanskrit- 
 Worterbuch, nach den Petersburger Worterbiichern bearbeitet. Lex.-8. 
 VIII, 541 S. 1887. Broschirt JL 15 , in Halbfranz geb. Jk 17 
 
 (Ankiindigung siehe: III. Indogerm. Sprachwissenschaft.) 
 
 A Sanskrit-English Dictionary. Based upon the St. Petersburg 
 Lexicons. Lex. 8. VIII, 672 S. 1891. Geb. in engl. Leinwand Ji. 21 
 
 Den ausschliesslk-hen Vertrieb fur England und die Kolonien haben : Luzac & Co. 
 in London, fur die Vereinigten Staaten : Ginn & Co. in Boston ubcrnommen. 
 siehe auch: Pracandapandava und Vamanas Stilregeln. 
 Catalog der Universitats- und Landesbibliothek in Strassburg s. Katalog. 
 Clementina, herausgegeben von Paul de Lagarde. 8. 200 S. 1865. 
 
 (Ji. 8 ) Vergriffen. 
 (Aus dem Verlag von B. G. Teubner in Leipzig in den meinigen iibergegangen.)
 
 IV. Orientalische Philologie. 17 
 
 Clementis Roman! Recognitiones Syriace. Edidit Paulus Antonius 
 de Lagarde. Lex.-8. pp. VIII, 167. 1861. (Ji 20 ) Jt 14 
 
 (Aus dem Verlag von B. G. Teubner in Leipzig in den meinigen iibergegangen.) 
 
 Ehni, J., Der vedische Mythus des Yama, verglichen mit den analogen 
 Typen der persischen, griechischen und germanischen Mythologie. 8. 
 VI, 216 S. 1890. JL 5 _ 
 
 Ethe, C. H., Neupersische Litteratur siehe: Grundriss der iranischen Philologie. 
 
 Euting, Julius, Sechs phonikische Inschriften aus Idalion. 4 
 
 17 S. mit 3 Taf. 1875. Jt 4 _ 
 
 Abgesehen von der Bereicherung, welche der phonizische Sprachschatz durch die 
 Kntzifi'erung dieser Inschriften erfahrt, wird durch dieselben die Geschichte Cyperns im 
 2. und 3. Jahrhundert y. Ghr. wesentlich erhellt. Die darunter befindliche bilinguis (in 
 cyprischer und phonizischer Sprache) durfte in ihrer ersten genauen Darstellung der 
 cyprischen Schriftzeichen auch fur klassische Philologen Interesse darbieten. 
 Erlauterung einer zweiten Opferverordnung aus Carthago. 
 Herrn Prof. Dr. H .J. Fleischer zur Feier seines fiinfzigjahrigen Doctor- 
 jubilaums als Gruss dargebracht. 8. 10 S. Mit 1 Tafel. 1874. UK 1 60 
 
 Durch dieses Bruchstiick erfahrt unsere Kenntnis des phonizischen Cultus und Sprach- 
 schatzes manche schatzbare Erweiterung. Literar. Centralblatt. 
 
 -- Sammlung der Carthagischen Inschriften, herausgegeben mit 
 Unterstiitzung der k. Akademie der Wissenschaften zu Berlin. Band I. 
 Tafeln 1202 und Anhang, Tafel 16. In 4. 1883. Jt 60 
 
 siehe auch: Katalog und Qolasta; ferner XI. Alsatica, XV. Vermischtes. 
 Evangelien, Die vier, arabisch aus der Wiener Handschrift herausgeg. von 
 
 Paul de Lagarde. 8. XXXII, 143 S. 1864. (UK 5 ) ^ 3 50 
 
 (Aus dem Verlag von B. G. Teubner in Leipzig in den meinigen iibergegangen.) 
 Pranke, 0., Grammatik des klassischen Sanskrit der Grammatiker, 
 der Litteratur und der Inschriften, sowie der Mischdialekte (epischer und 
 nordbuddhistischer). (Grundriss der indo-arischen Philologie und Altertums- 
 kunde I. Band, 5. Heft.) (In Vorbereitung.) 
 
 Paligrammatiker, Paligrammatik (Grundriss der indo-arischen 
 Philologie und Altertumskunde I. Band, 7. Heft). (In Vorbereitung.) 
 
 Fiirst, Julius (Rabbiner), Glossarium graeco-hebraeum oder der 
 griechische Worterschatz der jiidischen Midraschwerke. Ein Beitrag zur 
 Kultur- und Altertumskunde. 8. 216 S. 1891. Ji 7 
 
 Garbe, Richard, Samkhya und Yoga (Grundriss der indo-arischen Philo- 
 logie III. Band, 4. Heft). Lex. 8. 54 S. 1896. MB 
 
 Wir spenden der in sehr praciser Form alles Wissenswerthe bietenden Schrift in 
 ihren beiden Theilen die verdiente Anerkennung. Literar. Centralblatt 1896. Nr. 51. 
 
 siehe auch : Vaitana Sutra. 
 
 Geiger, Wilhelm, Grammatik und Litteratur des Singhalesischen 
 (Grundriss der indo-arischen Philologie I. Band, 10. Heft). (In Vorbereitung.) 
 
 Afganisch. Baluci. Geographic von Iran. Siehe: Grundriss der 
 iranischen Philologie. 
 
 Geldner, Carl F., Studien zum Avesta. 1. Heft. 8. IX, 181 S. 1882. 
 
 (Nicht mehr erschienen.) * 5 
 
 -- Vedische Litteratur (Sruti): Die drei Ve den (Grundriss der indo- 
 
 arischen Philologie II. Band, 1. Heft a). (In Vorbereitung.) 
 
 _ _ Awestalitteratur siehe: Grundriss der iranischen Philologie. 
 Goldschmidt, Prof. Siegfr., Prakrtica. 8. 32 S. 1879. Ji 1 
 
 siehe auch : Eavanavaha. 
 
 Grierson, G. A., The Aryan Vernaculars of Modern India and their 
 Literature (Grundriss der indo-arischen Philologie und Altertumskunde 
 I. Band, 9. Heft). (In Vorbereitung.)
 
 n er enzenen etrage entste, aurc zu vermeen, ass ie einzelnen 
 
 hnitte gleich nach ihrer Ablieferung einzeln gedruckt und ausgegeben werden. Durch 
 
 gemeinsamen Titel und ein ausfuhrliches Namen- und Sachregister am Schluss 
 
 Bandes werden die einzelnen Hefte zu einem gemeinschaftlichen Ganzen zusammen- 
 
 18 VERLAGSKATALOG von KARL J. TRUBNER in Strassburg. 
 
 Grundriss der indo-arischen Philologie und Altertumskunde, unter Mit- 
 wirkung von A. Baines-London, It. G. Bhandarkar-Puna, M. Bloomfield- 
 Baltimore, J. Bwr^ess-Edinburgh, 0. Franke-Konigsherg, It. (?rie-Tiibingen, 
 W. GW^er-Erlangen, K. Geldner-Berlin, G. A. Gnerson-Calcutta,, A. Hille- 
 brandt- Breslau , H. Jacobi-Ronn, J. Jo/Zy-Wurzburg, H. Kern-Leiden, 
 E. .Kw/m-Munchen, C. It. Lanman-Cambridge (Mass.), E. Z/eMwan-Strass- 
 burg, B. Liebich-Breslau, A. Macdonell-Oxtord, B. Meringer-Vfien, R. Pischel- 
 Halle, E. J. Rapson-London, J. S. <S>pe/er-Groningen, M. A. Stein-Lahore, 
 G. Tftibaut-Mlahahad. A. Venis-Benares, Sir R. West-London, M. Winter- 
 wj'te-Oxford, Th. Zctchariae-Halle. Herausgegeben von Georg Buhler. 
 
 In diesem Werk soil zum ersten Mai der Versuch gemacht werden, einen Gesamt- 
 iiberblick iiber die einzelnen Gebiete der indo-arischen Philologie und Altertumskunde in 
 knapper und systematischer Uarstellung zu geben. Die Mehrzahl der Gegenstande wird 
 damit iiberhaupt zum ersten Mai eine zusammenhangende abgerundete Behandlung er- 
 fahren; deshalb darf von dem Werk reicher Gewinn fur die Wissenschaft selbst erhofft 
 werden, trotzdem es in erster Linie fur Lernende bestimmt ist. 
 
 Gegen dreissig Gelehrte aus Deutschland, Oesterreich, England, Holland, Indien und 
 Amerika haben sich mit Hofrat G. Buhler in Wien vereinigt, um diese Aufgabe zu 
 losen, wobei ein Teil der Mitarbeiter ihre Beitrage deutsch, die ilbrigen sie englisch 
 abfassen werden. (Siehe nachfolgenden Plan.) 
 
 Besteht schon in der raumlichen Entfernung vieler Mitarbeiter eine grossere Schwierig- 
 keit als bei anderen ahnlichen Unternehmungen, so schien es auch geboten, die Unzu- 
 traglichkeit der meisten Sammelwerke, welche durch den unberechenbaren Ablieferungs- 
 termin der einzelnen Beitrage entsteht, dadurch zu vermeiden, dass die einzelne 
 Abschnitte gleich nach ihrer Ablieferun 
 einen 
 jedes 
 gefasst. 
 
 Das Werk wird aus drei Banden Lex. 8 im ungefiihren Umfang von je 1100 Seiten 
 bestehen. Der Subskriptionspreis des ganzen Werkes betragt durchschnittlich 65 Pf. pro 
 Druckbogen von 16 Seiten; der Preis der einzelnen Hefte durchschnittlich 80 Pf. pro 
 Druckbogen. Auch fur die Tafeln und Karten wird den Subskribenten eine durchschnitt- 
 liche Ermassigung von 20% auf den Einzelpreis zugesichert. Ueber die Einteilung des 
 Werkes giebt der nachstehende Plan Auskunft. 
 
 Plan des Werkes: 
 
 NB. Die mit * bezeichneten Hefte sind bereits erschienen. 
 Band I. Allgemeines und Sprache. 
 
 1. Geschichte der indo-arischen Philologie und Altertumskunde von 
 Ernst Kuhn. 
 
 2. Vorgeschichte der indo-arischen Sprachen von R. Meringer. 
 
 3. ) Die indischen Systeme der Grammatik, Phonetik und Etymologic 
 
 von B. Liebich. 
 
 * b] Die indischen Worterbiicher (Kosa) von Th. ZctcMriae. Subskr.- 
 Preis Ji 2. , Einzelpreis Ji. 2.50. 
 
 4. Grammatik der vedischen Dialekte von C. R. Lanman (englisch). 
 
 5. Grammatik des klassischen Sanskrit der Grammatiker, der Litte- 
 ratur und der Inschriften, sowie der Mischdialekte (epischer und 
 nordbuddhistischer) von 0. Franke. 
 
 *6. Vedische und Sanskrit-Syntax von J". S. Speyer. Subskr.-Preis 
 Jl>. 4. , Einzelpreis Ji. 5. . 
 
 7. Paligrammatiker, Paligrammatik von 0. Franke. 
 
 8. Prakritgrammatiker, Prakritgrammatik von R. Pischel. 
 
 9. Grammatik und Litteratur des tertiaren Prakrits von Indien von 
 G. A. Grierson (englisch). 
 
 10. Grammatik und Litteratur des Singhalesischen von Wilh. Geiger. 
 *11. Indische Palaeographie (mit 17 Tafeln) von G. Buhler. Subskr.- 
 Preis Jl. 15., Einzelpreis Ji 18.50. 
 
 Band II. Litteratur und Geschichte. 
 1. Vedische Litteratur (Sruti). 
 
 a) Die' drei Veden von K. Geldner. 
 
 b) Atharvaveda von M. Bloomfield (englisch).
 
 IV. Orientalische Philologie. 19 
 
 Grundriss der indo-arischen Philologie etc. (Fortsetzung). 
 
 2. o) Epische Litteratur von H. Jacobi. 
 
 b) Klassische Litteratur (einschliesslich der Poetik und der Metrik) 
 von H. Jacobi. 
 
 3. Quellen der indischen Geschichte. 
 
 ) Litterarische Werke und Inschriften von G. Buhler (englisch). 
 *b) Indian Coins (with five plates) by E. J. Rapson (englisch). 
 Subskr.-Preis Jz 5. , Einzelpreis UK 6. . 
 
 4. Geographic von G. Buhler und M. A. Stein. 
 
 5. Ethnographie von A. Baines (englisch). 
 
 nG.Buhler,J. Jolly unA Sir R. West (englisch). 
 
 *8. Recht und Sitte (einschliesslich der einheimischen Litteratur) von 
 
 J. Jolly. Subskr.-Preis Jt 6.50, Einzelpreis Jt &.. 
 9. Politische Geschichte bis zur muhammedanischen Eroberung von 
 
 G. Buhler (englisch). 
 Band III. Religion, weltliche Wissenschaften und Kunst. 
 
 1. *o) Vedic Mythology by A. A. Macdonell (engl). Subskr.-Preis 
 
 Ji 7.50, Einzelpreis Ji 9.. 
 b) Epische Mythologie von M. Winternitz. 
 *2. Ritual-Litteratur, Vedische Opfer und Zauber von A. Hilhbrandt. 
 
 Subskr.-Preis Ji8., Einzelpreis ^59.50. 
 3. Vedanta und Mimamsa von G. Thibaut. 
 
 *4. Samkhya und Yoga von R. Garbe. Subskr.-Preis Ji. 2.50, Einzel- 
 preis UK 3.. 
 
 5. Nyaya und Vaisesika von A. Venis (englisch). 
 
 6. Vaisnavas, Saivas, Sauras, Ganapatas, Skandas, Saktas von R. G. 
 Bhandarkar (englisch). 
 
 7. Jaina von E. Leuinann. 
 
 *8. Manual of Indian Buddhism by H. Kern (englisch). Subskr.-Preis 
 
 JL 5 50, Einzelpreis Ji 7.. 
 9. Astronomic, Astrologie und Mathematik von G. Thibaut. 
 
 10. Medizin von J. Jolly. 
 
 11. Bildende Kunst (mil Illustrationen) von J. Burgess (englisch). 
 
 12. Musik. 
 
 Grundriss der iranischen Philologie, unter Mitwirkung von F. K. Andreas, 
 Chr. Bartliolomae, C. H. EtU, K. F. Geldner, P. Horn, H. Hiibschmann, 
 A V. W Jackson, F. Justi, Th. NSldeke, C. Salemann, A. Scan, 
 F. H. Weissbach, E. W. West und V. Zukovskij herausgegeben von 
 Wilh. Geiger und Ernst Kuhn. 
 
 Der Grundriss der iranischen Philologie wird in Lieferungen von ^' c h^< 
 
 ^^^^^ 
 
 Plan des Werkes: 
 
 1 Ban Einleitung. Geschichte der iranischen Philologie Prof. 
 Dr. E. Kuhn. 
 
 of-D, fir. 
 
 \2. Awestasprache und Altpersisch Prof. Dr. Chr. 
 (3. Mittelpersisch Akademiker Dr. C. Salemann. 
 
 4 Neupersische Schriftsprache Pnvatdozent Dr. P. Horn. 
 
 5. Die iibrigen modernen Sprachen und Dialekte. 
 
 A. Afyanisch / Prof Dr w Geiger. 
 
 B. Baluci \ 
 
 C. Kurdisch Prof. Dr. A. Soctn.
 
 20 VERLAGSKATALOG von KARL J. TRUBNER in Strassburg. 
 
 Grundriss der iranischen Philologie (Fortsetzung). 
 
 D. Ossetisch Prof. Dr. H. Hiibschmann. 
 
 P.' EGilaki etc. | "** * 
 G. Dialekte in Persien Prof. V. ZukovsMj. 
 II. Band. 
 
 Abschnitt II. Litteratur. 
 
 1. Awestalitteratur Prof. Dr. C. F. Geldner. 
 
 2. Die altpersischen Inschriften Dr. F. H. Welssbach. 
 
 3. Die Pahlavilitteratur Dr. E. W. West. 
 
 4. Das iran. Nationalepos Prof. Dr. Th. Noldeke. 
 
 5. Neupersische Litteratur Prof. Dr. C. H. Etht. 
 Abschnitt III. Geschichte und Kultur. 
 
 1. Geographie von Iran Prof. Dr. W. Geiger. 
 
 2. Geschichte Irans von den altesten Zeiten bis zum Ausgang 
 der Sasaniden Prof. Dr. F. Justi. 
 
 3. Geschichte Irans in islamitischer Zeit Privatdozent Dr. P. Horn. 
 
 4. Die iranische Religion Prof. Dr. A. V. W. Jackson. 
 
 5. Mlinzen, Gemmen Privatdozent Dr. P. Horn. 
 
 6. Schriftkunde Dr. F. K. Andreas. 
 
 Bis jetzt erschienen: I. Band. 1. Lieferung Jl 8 , 2. Lieferung Ji 4 50 
 II. 1., 2. und 3. Lieferung a Jt. 8 
 
 Hillebrandt, Alfred, [Indische] Ritual-Litteratur. Vedische Opfer und 
 Zauber (Grundriss der indo-arischen Philologie III. Band, 2. Heft). 
 Lex.-8. 189 S. 1897. Jt 9 50 
 
 Holtzmann, Adolf, Agni nach den Vorstellungen des Mahabharata. 8. 
 
 36 S. 1878. Jt,. 1 
 
 Arjuna. Ein Beitrag zur Reconstruction des Mahabharata. 8. 
 
 69 S. 1879. Ji 1 60 
 
 , >ie 2)en?toiirbigfeitcn <Sd)afj 2atna^' be 6rften bon 
 (15151570). 3lu bem Driginaltej:t sum erften 2Me iiberje^t unt> 
 ntit (Sttautetungen berfe^en. It. 8. 156 @. 1891. Jk 3 
 
 Grundriss der neupersischen Etymologic (Sammlung indo- 
 
 germanischer Worterbiicher, IV. Band). 8. XXV. 384 S. 1893. Jt 15 
 Geschichte der neupersischen Schriftsprache. Geschichte Irans in isla- 
 mitischer Zeit. Miinzen, Gemmen von Iran. Siehe : Grundriss der iranischen 
 Philologie. 
 
 Hiibschmann, H., Persische Studien. 8. 288 S. 1895. Jl 10 
 
 Jnhalt: 1. Beitrage zu Horn's Grundriss der neupersischen Etymologic. 
 2. Neupersische Lautlehre. 
 
 Etymologie und Lautlehre der ossetischen Sprache. (Samm- 
 lung indogerm. Worterbiicher, I. Band.) 8. VIII, 151 S. 1887. Jt 4 
 
 Die ossetischc Sprache siehe: Grundriss der iranischen Philologie. 
 
 Das indogerman. Vocalsystem siehe: III. Indogerm. Sprachwissenschaft. 
 Huth, Dr. Georg (Privatdozent an der Universitat Berlin), Die tibetische 
 
 Version der Naihsargikaprayac. cittikadharmas. Buddhistische 
 Suhnregeln aus dem Pratimokshasutram. Mit kritischen Anmerkungen 
 herausgegeben, iibersetzt und mit der Pali- und einer chinesischen Fassung, 
 sowie mit dem Suttavibhanga verglichen. (Diss.) 8. 51 S. 1891. Jt 2 
 
 Geschichte des Buddhismus in der Mongolei. Aus dem Tibe- 
 
 tischen des Jigs-med nam-mk'a, herausgegeben, iibersetzt und erlautert. 
 
 I. Teil: Vorrede, Text, kritische Anmerkungen. gr. 8. X, 296 S. 
 
 1892. Jt 20 
 
 II. Teil: Uebersetzung. Nachtrage zum ersten Teil. gr. 8. XXXII, 
 
 456 S. 1896. Ji 30
 
 IV. Orientalische Philologie. 21 
 
 Jackson, A. V. W., Die iranische Religion siehe: Grundriss der iranischen. 
 
 Philologie. 
 
 Jacobi, H., Epische Litteratur Indiens. Klassische Litteratur 
 Indiens einschliesslich der Poetik und der Metrik (Grundriss der indo- 
 arischen Philologie und Altertumskunde II. Band, 2. Heft a und b). 
 
 (In Vorbereitung.) 
 
 Jensen, P., Die Kosmologie der Babylonier. Studien und Materialien. 
 Mit einem mythologischen Anhang und drei Karten. gr. 8. XVI, 546 S. 
 1890. Ji 40 
 
 . . . Jensen hat in ausgezeichneter Weise nicht nur das gesamte Material verwertet, 
 er hat sich auch, ehe er an seine Arbeit ging, eingehend unter der Leitung eines tttch- 
 tigen Astronomen mit den betreffenden astronomischen Fragen bekannt gemacht. Der 
 Lohn seiner muhevollen Arbeit waren nicht unerhebliche Resultate, welche meist durch 
 Epping's und Strassmaier's Astronomisches aus Babylon* ihre Bestatigung gefundea 
 haben. Von allgemeinem Interesse ist besonders der Nachweis, dass der grossere TeiF 
 der Tierkreisbilder, ja wahrscheinlich alle, aus Babylonien stammt. . . Hochst dankens- 
 wert sind die Abschnitte der Weltsch6pfung und Weltbildung* und iiber die Sintflut 
 . . . Aber mit dem Hinweis auf den religionsgeschichtlichen Ertrag der Arbeit, der hier 
 nur angedeutet werden konnte. ist ihre Bedeutung nicht erschopft, nicht geringer ist ihr 
 Wert fur die philologische Forschung. Jensen beherrscht in ausgezeichneter Weise das 
 Material, und bei seiner tuchtigen philologischen Schulung, seinem Scharfsinn und seiner 
 hervorragenden Kombinationsgabe war es nicht anders zu erwarten, als dass er unsere 
 Erkenntnis um ein gutes Stuck fordern wiirde ; man mag im Einzelnen da und dort 
 anderer Meinung sein, im Grossen und Ganzen ist diese Arbeit unbestreitbar eine ttich- 
 tige Leistung, die kein Historiker oder semitischer Philologe ohne Nutzen aus der Hand 
 legen wird. Die dem Buche beigegebenen Karten: Lauf der Venus, der babyloniscb.fr 
 Tierkreis und die Welt nach babylonischer Vorstellung fordern wesentlich das Ver- 
 standnis. Die Ausstattung des Werkes ist mustergiiltig. 
 
 Literar. Centralblatt 1890. Nr. 15. W. N. 
 
 Die Verlagshandlung erlaubt sich ferner auf die ausfflhrlichen Besprechungen in der 
 Zeitschrift fur Assyriologie V, 1 (Zimmern), der Theologischen Literatur- 
 zeitung 1890 Nr. 7 (Budde), der Revue critique 1890 Nr. 25 (Halevy) und der Bei- 
 lage zur Allgemeinen Zeitung 1890 Nr. 34 zu verweisen. 
 
 Hittiter und Armenier. 17 Bogen mit 10 lithographischen Tafeln 
 
 und einer Karte. <& 25 
 
 Inhalt: I. Das Volk und das Land der Hatier. II. Die Inschriften der Hatier. 
 A) Verzeichnis der Inschriften. B) Uebersetzungen der Halfte des inschriftlichen Mate- 
 rials mit Commentar. III. Das Schriftsystem der Hatier. A) Die Sehriftzeiehen und 
 ihre Verwendung. B) Das Vorbild der hatischen Schrift. C) Der palaeo-armenische 
 Ursprung der hatischen Schrift. IV. Die Sprache der Hatier und das Armenische. 
 A) Grammatisches: a) zur Formenlehre, b) zur Syntax. B) Lexikalisches: a) Appellative, 
 b) Nomina propria. C) Der Lautbestand der hatischen Sprache im Verhaltnis zu dem 
 des Indogermanischen und des Armenischen. V. Zur hatisch-armcnischen Religion. 
 A) Hatische Gotterzeichen. B) Hatische Gotternamen. C) Hatische Cotter. D) Einfluss 
 des syrischen Cultus auf den der Hatier. E) Die Religion der Hatier und die der Ar- 
 menier. VI. Zur hatisch-armenischen Geschichte. 
 
 Jolly, Julius, Recht und Sitte [in Indien] einschliesslich der einheimischen 
 Litteratur (Grundriss der indo-arischen Philologie und Altertumskunde 
 II. Band, 8. Heft). Lex.-8. 160 S. 1896. ^ 8 - 
 
 Der Verfasser bietet uns das zu einer Uebersicht uber die gesammte indische Rechts- 
 und Sittengeschichte verarbeitete reife Ertriigniss seiner eigenen wie der Mitforscher 
 Untersuchungen in einer bis dahin unerreichten Vollstandigkeit. Nach Massgabc von 
 Jolly's bisherigen Beitragen zur Theorie und Geschichte des indischen Rechts durfte man 
 Bedeutendes erwarten. Dennoch ist man uberrascht zu sehen, mit wie grosser Selb- 
 standigkeit und Sicherheit im Urtheil die ganze weitschichtige Materie . . . zur Darslellung 
 gebracht wird. Literar. Centralblatt 1896. Nr. 36. 
 
 Meclizin in Indien (Grundriss der indo-arischen Philologie und 
 
 Altertumskunde III. Band, 10. Heft). (In Vorbereitung.) 
 
 - siehe auch : Biihler, Jolly und West. 
 Justi,F., Geschichte Irans von den altesten Zeiten bis zum Ausgang der Sasa- 
 
 niden. Siehe: Grundriss der iranischen Philologie. 
 Katalog der kaiserlichen Universitats- und Landesbibliothek in Strassburg 
 
 Arabische Literatur. [Verfasst von Dr. Julius Eutmg.] 4 VIII, 
 
 111 S. 1877. 
 
 Festschrift zur 400jahr. Jubelfeier der Eberhard-Karls-Universitat zu Tubingen.
 
 22 VERLAGSKATALOG von KARL J. TRUBNER in Strassburg. 
 
 Katalog der kaiserlichen Universitats- und Landesbibliothek in Strassburg. 
 Orientalische Handschr iften Theil I: Hebraische, arabische, 
 persische und tiirkische Handschriften, bearbeitet von Dr. S. Landauer. 
 4. IV, 75 S. 1881. Jl. b 
 
 Kautzsch, E. und A. Socin, Die Aechtheit der moabitischen Alter- 
 th timer, gepriift. Mit 2 Tafeln. 8. VIII, 191 S. 1876. Jl. 4 
 
 Nach dem fast einstimmigen Urteil aller Kritiker und Sachverstandigen (Noldeke^ 
 Deutsche Rundschau, Miirz 1876 ; S p r e n g e r , Academy, March 11, 1876 ; Ch. C 1 e r m o n t - 
 Ganneau, Revue critique 11 mars 1875; Jenaer Literaturzeitung Nr. 15, 1876 etc. etc.) 
 ist es den beiden Verfassern gelungen, den Beweis zu liefern, dass die in Berlin befind- 
 lichen moabitischen Altertiimer modern e False hungen sind. Jedenfalls bildet das 
 Buch den wichtigsten aller bisher erschienenen Beitrage zur Losung dieser Streitfrage 
 und besitzt bleibenden Wert durch die darin betindlichen eingehenden Untersuchungen 
 iiber alttestamentliche Geschichte, Religion und Archaologie. 
 
 Kayser, Karl, siehe: Buch von der Erkenntniss der Wahrheit. 
 Kern, H., Manual of Indian Buddhism (Grundriss der indo-arischen 
 Philologie und Altertumskunde III. Band. 8. Heft). Lex. 8. 138 S. 1896. 
 
 JL 7 - 
 
 .... Es ist Kern gelungen, seine Aufgabe mit grossem Geschick zu losen. Den 
 gerade in der buddhistischen Forschung der Neuzeit sich anmeldenden schiidlichen Folgen 
 der Isolierung durch eine die verschiedenen Arbeiten einander naher bringende Thatigkeit 
 
 entgegenzuwirken, ist ein grosses Verdienst des vorliegenden Manual, das darum im 
 Rahmen des Grundrisses.> auch so angesehen seine voile Existenzberechtigung hat. 
 
 Literar. Centralblatt 1897. Nr. 2. 
 
 Kuhn, Ernst, Geschichte der indo-arischen Philologie und Alter- 
 tumskunde (Grundriss der indo-arischen Philologie und Altertumskunde 
 I. Band, 1. Heft). (In Vorbereitung.) 
 
 Geschichte der iranischen Philologie siehe: Grundriss der irauischen 
 Philologie. 
 
 Lagarde, Paul de, Anmerkungen zur griechischen Uebersetzung 
 der Proverbien. 8. VIII, 96 S. 1863. (Jt 3 ) Vergriffen. 
 
 (Aus dem Verlag von B. G. Teubner in Leipzig in den meinigen iibergegangen.) 
 
 siehe auch : Clementina. dementis Roman! Recognitiones Syriace. 
 Evangelien, die vier. Libri Veteris Testament} Apocryphi Syriace. Ono- 
 mastica sacra. Ferner unter VI. Klass. Philologie, Neulatein: Scaligeri 
 Poemata. Titus Bostremis. 
 
 Landauer, Dr. S., siehe: Katalog der K. Univ.- n. Landesbibliothek in Strassburg: 
 
 Orientalische Handschriften. 
 Lanman, C. R., The Grammar of the Vedic Dialects (Grundriss der 
 
 indo-arischen Philologie und Altertumskunde I. Band, 4. Heft). 
 
 (In Vorbereitung.) 
 
 Leumann, E., Jain a (Grundriss der indo-arischen Philologie und Alter- 
 tumskunde III. Band, 7. Heft). (In Vorbereitung.) 
 
 tibersicht iiber die Avashyaka-Literatur mit Excerpten aus der 
 
 Silanka-Handschrift und einer photographischen Reproduktion derselben 
 sowie mit einer Pratika-Liste zum Viseshavasyakabhashya. Gross-Folio. 
 Etwa 12 Bogen Text und 35 Tafeln in Lichtdruck. (Unter der Presse.) 
 
 Leumann, E. u. J., Etymologisches Sanskritworterbuch (Sammlung 
 indogermanischer Worterbiicher, V. Band). (Unter der Presse.) 
 
 Libri Veteris Testament! Apocryphi Syriace, e recognitione Pauli Antonii 
 de Lagarde. 8. pp. XXXIX, 272. 1861. (JL 20 ). Jt 14 - 
 
 Liebich, B., Die indischen Systeme der Grammatik, Phonetik und 
 Etymologic (Grundriss der indo-arischen Philologie und Altertumskunde 
 I. Band, 3. Heft a). (In Vorbereitung.) 
 
 Macdonell, A. A., Vedic Mythology (Grundriss der indo-arischen Philo- 
 logie und Altertumskunde III. Band, 1. Heft a). Lex. 8. 177 S. 1897. 
 
 Ji. 9
 
 IV. Orientalische Philologie 
 
 Meringer, R., Vorgeschichte der indo-arischen Sprachen (Grund- 
 riss der indo-arischen Philologie und Altertumskunde I. Band, 2. Heft). 
 
 (In Vorbereitung.) 
 
 Noldeke, Theodor, Das iranische Nationalepos. Besonderer Abdruck 
 aus dem Grundriss der iranischen Philologie. 8. 82 S. 1896. Ji. 4 50 
 
 Das neue vorliegende Heft des iranischen Grundrisses (vgl. lauf. Jahrg. Nr. 12. S)>. 
 ringt zuniichst den Abschluss der meisterhaften Studio von Noldeke iiber das 
 Schahname. Wer es beim Lesn des ersten, im ersten Hefte des zweiten Bandes ent- 
 
 425 d. Bl.) br 
 
 haltenen Theiles noch nicht gemerkt hat, der wird jetzt zugeben miissen. dass in Zukunft 
 eine wissenschaftliche Beschiiftigung mit der epischen Dichtung im Allgemeinen oder 
 mit der alteren Epik eines Einzelvolkes nicht moglich ist, wenn man nicht diese Dar- 
 stellung des iranischen Nationalepos grilndlich durchgearbeitet und sich zu eigen gemacht 
 hat. Sie stellt sich wiirdig neben die ebenfalls tief eindringenden und neue Ziele er- 
 schlicssenden Untersuchungen Comparetti's iiber das finnische Nationalepos. Schon was 
 Noldeke selbst an Parallelen aus dem altgriechischen, altgermanischen und anderen 
 volksthiimlichen Epen anfiihrt, wird jedem classischen und germanischen Philologen sehr 
 erwiinscht und anregend sein. Literar. Centralblatt 1896, Nr. 43. 
 
 Onomastica sacra edidit P. de Lagarde. Zvvei Theile in einem Band. 1870. 
 8. VIII, 304, 160 S. (Ji. 15 ). (JI. 10 ) Vergriffen. 
 
 (Aus dem Verlag von B. G. Teubner in Leipzig in den meinigen ubergegangen.) 
 Pfungst, Dr. Arthur, Das Sutta Nipata siehe: Sutta Nipata. 
 Pischel, R., Prakritgrammatiker, Prakritgrammatik (Grundriss der 
 indo-arischen Philologie und Altertumskunde I. Band, 8. Heft). 
 
 (In Vorbereitung.) 
 
 Pracandapandava. Ein Drama des Rajacekhara. Zum ersten Male 
 herausgegeben von Carl Cappeller. 8. 50 S. 1885. Ji 3 50 
 
 Qolasta oder Gesange und Lehren von der Taufe und dem Aus- 
 gang der Seele. Mandaischer Text mit sammtlichen Varianten nach 
 Pariser und Londoner Manuscripten ; mit Unterstiitzung der deutschen 
 morgenlandischen Gesellschaft autographirt und herausgegeben von 
 Dr. Jul. Euting. Stuttgart 1867. gr. Fol. 40 Bogen. (Friiherer Laden- 
 preis M. 100 , ermiissigt auf Ji. 75 .) Vergriffen. 
 
 Rapson, E. J., Indian Coins. (Grundriss der indo-arischen Philologie und 
 Altertumskunde II. Band. Heft 3 b.) With 5 plates. Lex. 8. 52 S. 1898. 
 
 JL 6 
 
 Ravanavaha oder Se tub and ha. Prakrit und Deutsch herausgegeben von 
 Siegfried Goldschmidt. Mit einem Wortindex von Paul Gold- 
 schmidt und dem Herausgeber. Erste Lieferung. Text und Wortindex 
 enthaltend. 4. XXIV, 194 S. 1880. Zweite Lieferung: Ubersetzung. 
 4. 136 S. 1884. (<& 43 ) 
 
 Sabbag, Miha/il, Die Brieftaube. Schneller als der Blitz, fliichtiger als 
 die Wolke. Aus dem Arabischen. Nebst einem Anhange : Beitrage zur 
 Geschichte der Taubenpost. 8. 55 S. 1879. (UL 1 50) Vergriffen. 
 
 Grammatik der arabischen Umgangssprache in Syrien und 
 
 Aegypten. Nach der Miinchner Handschrift herausgegeben von H. 1 hor- 
 becke. 8. X, 80 S. 1886. * 4 
 
 Sa'di's, Muslicheddin, Aphorismen undSinngedichte. Zum ersten Male 
 herausgegeben und iibersetzt von Dr. W. Bacher. Mit Beitragen zur 
 Biographic Sa'di's. Mit Subvention des Autors durch die Kaiserl. Aka- 
 demie der Wissenschaften in Wien. 8. LXXIV, 200 S. 1879. Ut 6 - 
 Der Verfasser bietet im vorstehenden Werke don Freunden porsischor Literatur zum 
 ersten Male in deutscher fmetrischer) Bearbeitung die Aphorismen Sadis. 1J Ueber- 
 setzung parallel lauft der Urtext, der, sehen wir von dor Calcn tacr und Caw nporer Aus- 
 gabe dor (1,'Hiunn (werko .li.-s Dichtcrs ab, jetzt zum ersten Male aus cmor ourop Oft 
 hervorgeht, In dor voran^oschickton Biographio erhalten wir in anziehender Se&ttderang 
 ein farbenreiches Bild von dem wechselvollen Leben dieses Hauptvertreters der d.dak- 
 tischen Poesie der Perser. Literar. Centralbl. 18,9, Nr. 4 
 
 Salemann, C., Mittelpersisch. - Pamirdialekte. - Mazandaraui, Gilaki etc. 
 Siehe : Gmndriss der iranischen Philologie.
 
 24 VERLAGSKATALOG von KARL J. TRUBNER in Strassburg. 
 
 Scherman, Dr. Lucian, Philosophische Hymnen aus der Rig- und 
 
 Atliarva-Veda-Sanhita verglichen mit den Philosophemen der alteren 
 
 Upanishad's. 8. VII, 96 S. 1887. M. 2 50 
 
 -Seybold, Chr., Die arabische Sprache in den romanischen Landern. Siehe : 
 
 Grundriss der romanischen Philologie unter VIII. Roman. Philologie. 
 Silanka siehe Leumann, E. 
 Socin, A., Die kurdische Sprache siehe: Grundriss der iranischen Philologie. 
 
 siehe auch: Kautzsch n. Socin. 
 Speyer, J. S., Vedische und Sanskrit-Syntax (Grundriss der indo- 
 
 arischen Philologie und Altertumskunde I. Band, 6. Heft). Lex. 8. 96 S. 
 
 1896. ^ 5 
 
 Stackelberg, Reinh., Beitrage zur Syntax des Ossetischen. 8. V, 
 
 99 S. 1886. Ji 3 
 
 Stein, M. A., Indische Geographic siehe: Biihler und Stein. 
 Sutta Nipata, Das. Eine Sammlung von Gesprachen, welche zu den kano- 
 
 nischen Biichern der Buddhisten gehort. Aus der englischen Uebersetzung 
 
 von Prof. V. Fausboll in Kopenhagen. (Sacred books of the East, vol. 
 
 X.) Ins Deutsche iibertragen von Dr. Arthur Pfungst. 1. Lief. 8. 
 
 X, 80 S. 1889. Jt 1 50 
 
 Thibaut, G., Vedanta und Mimamsa. Astronomie, Astrologie 
 
 und Mathematik in Indien (Grundriss der indo-arischen Philologie 
 
 und Altertumskunde III. Band, 3. und 9. Heft). (In Vorbereitung.) 
 
 Thorbecke siehe: Sabbag, Arab. Grammatik. 
 Vaitana Sutra, Das Ritual des Atharvaveda. Aus dem Sanskrit iiber- 
 
 setzt und mit Anmerkungen versehen von Dr. Rich. Gar be. 8. V, 
 
 116 S. 1878. M 4 
 
 Vamanas Stilregeln, bearbeitet von Carl Cappeller. Lex. 8. XII, 38 S. 
 
 1880. Jk 1 50 
 
 Venis, A., Nyaya and Vaiseshika (Grundriss der indo-arischen Philologie 
 
 und Altertumskunde III. Band, 5. Heft). (In Vorbereitung.) 
 
 "Weissbach, F. H., Die altpersischcn Inscliriften siehe : Grundriss der iranischen 
 
 Philologie. 
 
 West, E. W., Pahlavi-Literature siehe: Grundriss der iranischen Philologie. 
 West, Sir R., siehe: Biihler, Jolly und West. 
 Winternitz, M., Epische Mythologie Indiens (Grundriss der indo-arischen 
 
 Philologie und Altertumskunde III. Band, 1. Heft b). (In Vorbereitung.) 
 Zachariae, Theodor, Die indischen Worterbucher (Kosa) (Grundriss 
 
 der indo-arischen Philologie und Altertumskunde I. Band, 3. Heft b). 
 
 Lex. 8. IV. 40 S. 1897. M 2 50 
 
 Znkovskij, V., Die persischen Dialekte siehe: Grundriss der iranischen Philo- 
 logie.
 
 V. 
 
 Dtimichen, Dr. Johannes (Professor an der Universitat Strassburg), Bau- 
 geschichte des Denderatempels und Beschreibung der ein- 
 zelnen Teile des Bauwerks nach den an seinen Mauern befmdlichen 
 Inschriften. kl. fol. 50 S. und 57 Tafeln Inschriften nebst 2 Planen. 
 
 1877. U* 60 
 Der Herausgeber des,vorstehenden Werkes, einer der fruchtbarsten und thatigsten 
 
 Aegyptologen in der Veroffentlichung unbekannter und wichtiger Inschriften des agypti- 
 schen Altertums, hat sich durch diese neue Publikation wiederum ein besonderes Ver- 
 dienst um die von ihm mit Liebe und Eifer gepflegte Wissenschaft erworben. Den Kern 
 der Arbeit, um welchen sich eine erschopfende Anzahl teils bekannter, teils unbekannter 
 Texte als erklarendc oder analoge Beispiele gruppieren, bilden die von dem Herausgeber 
 auf seiner letzten agyptischen Reise mit grossen Miihen und Kosten freigelegten unteren 
 Rander der Aussenwande des hochberuhmten Tempels von Dendera (Tentyra der Alten), 
 welche eine vollstandig erhaltene Bauurkunde unter Angabe der Maasse aller Sale, 
 Zimmer, Treppen etc. des Tempels enthalten. 
 
 Prof. Brugsch in den Gottinger gel. Anzeigen vom 28. Marz 1877. 
 
 Die Oasen der libyschen Wiiste. Ihre alten Namen und ihre 
 
 Lage, ihre vorziiglichsten Erzeugnisse und die in ihren Tempeln ver- 
 ehrten Gottheiten, nach den Berichten der agyptischen Denkmaler. Mit 
 19 Tafeln hieroglyphischer Inschriften und bildlicher Darstellungen in 
 Autographic des Verfassers. 4. VI. 34 S. 1878. Jt 15 
 
 Der Verfasser giebt die Losung eines schwierigen geographischen Ratsels, die ihm 
 vollstandig gelungen ist. Alle spateren Arbeiten uber die Gesehichte und alte Geographic 
 der Oasen werden an D.'s wichtige Arbeit und die in ihr niedergelegten Resultate anzu- 
 kniipfen haben. G. Ebers im Literar. Centralblatt 1877, Nr. 20. 
 
 Ueber die Tempel und Graber im alten Agypten und ihre 
 
 Bildwerke und Inschriften. Vorlesung, gehalten am 19. November 1872 
 in der kaiserl. Universitat zu Strassburg. 8. 32 S. 1872. Jk 60 
 M. Dumichen resume fort nettement pour 1'instruction du public une foule d'idees 
 courantes dans la science. Revue critique. 
 
 Zur Erinnerung an Richard Lepsius. 8. 23 S. 1884. Jl 1 
 
 Faselius, Aug., Altagyptische Kalenderstudien. gr. 8. IV, 82 S. 
 
 1878. * 2 40 
 Lauth, Prof. Dr. Jos., Agyptische Chronologie, basiert auf die voll- 
 
 standige Reihe der Epochen seit Bytes-Menes bis Hadrian-Antonin durch 
 3 voile Sothisperioden = 4380 Jahre. Autographiert. 8. VI, 240 S., 
 o Tafeln. 1877. 
 
 Moses-Hosarsyphos Sali' Hus Levites-A'Haron f rater Ziphorah- 
 
 Dabariah conjux Miriam-Bellet soror Elisheba-Elizebat fratria. Ex monu- 
 mento inferioris Aegypti per ipsum Mose abhinc annos MMMCD dedicate 
 nunc primum in lucem produxit Franc. Jos. Lauth. Cum duabus tabulis 
 et uno photogrammate. 4. lithogr. 248 S. 1879. Ji 25 
 
 Spiegelberg, Wilhelm (Privatdozent der Agyptologie an der Universitat 
 Strassburs), Rechnungen aus der Zeit Setis I. (ca. 1350 v. Chr.) mit 
 anderen Rechnungen des neuen Reiches herausgegeben und erklart. 
 
 2 Bande folio. 1896. 
 
 70 ' 
 
 Inhalt des Textbandes (VIII, 100 S.) I. Gesehichte der Papyri Rollin II Aenssero 
 BeschaiTenheit der Handschriften. III. Datierung und Lokalisierun? der Handsch 
 IV. Zur Palaeographie der Papyri Rollin. V. Character und Inhalt der Rechnungen 
 VI. Uebersetzunl der Rechnungen. VII. Kommentar. VIII. Anhanp, en haltpnd cine chro 
 nologisch geordnete Liste von Namen Sgyptischer Kriegs- nnd Handcl s rhifr und emcn 
 Warentarif der erwahnten Epoche der agyptischen Gesehichte. Lino b n.lic Uber da, 
 Wertverhaltnis von Gold, Silber und Kupfer, in welcher msbcsondero die 
 
 Wertverhaltnis von Gold, Silber und Kupfer, in wecer mscsonero ie an 
 Entwertung des Silbcrs im Pharaoncnreich ziffernmassig nachgew 1 esen ist, bildet den 
 Schluss der Arbeit.
 
 26 VERLAGSKATALOG von KARL J. TRUBNER in Strassburg. 
 
 Spiegelberg, Wilhelm, Rechnungen aus der Zeit Setis I. (Fortsetzung). 
 
 Der Tafelband (mit 43 Tafeln) enthalt die bearbeiteten Papyri in Autographic und' 
 Lichtdruck nebst einer hieroglyphischen Umschrift der samtlichen Texte. 
 
 Dass die von Hrn. Pleyte schon 1868 sehr unvollkommen herausgegebenen und 
 erklarten Rechntfngen der Papyri Rollin der Bibliotheque nationale zu Paris in ver- 
 besserter Form und in einer dem heutigen Standpunkt der agyptologischen Wissensehaft 
 entsprechenden Weise behandelt wtirden, erschien schon lange als ein Bedurfnis. Auch 
 Ref. hat fur die Novembersitzung 1895 der Bibl. Archaeol. Society eine Arbeit tiber einen 
 Teil der Papyri Rollin, die Backerrechnungen eingereicht, die bereits gedruckt in einem 
 der nachsten Hefte der Proceedings dieser Gesellschaft erscheinen wird. Herr Dr. Spiegel- 
 berg hat es sich aber zur Aufgabe gemacht. samtliche Rechnungen dieser Papyri in 
 corrector Weise herauszugeben und zu kommentieren. Dabei ist dem selben die . . . Ver- 
 
 lagshandlung durch wurdige Ausstattung zu Hilfe gekommen Ein Verdienst hat 
 
 sich Spiegelberg erworben, dass er den von Pleyte falsch Ra-nefer-cheper-ka gelesenen 
 Konigsnamen als den Tutmes I Ra-aa-cheper-ka, an einer andern Stelle als den Ame- 
 nophis II erkannte, und dass er eine Rcihe agyptischer Worter teils neu, teils priiziser 
 bestimmtc ... Literarisches Centralblatt 1896, Nr. 11. 
 
 . . . . The author's aims are high, and his method excellent. No fragment seems to 
 have proved too small and worn for him to attempt its decipherment ; and where he 
 attempts he succeeds. By his present work, for which in every other respect but that 
 of metrology he was so admirably qualified, he has again laid all Egyptologists under 
 great obligations ... F. LI. Griffith, The Academy, 1896, Okt. 10. 
 
 Les Papyrus Rollin do la Bibliotheque Nationale ont deja ete publics et commentes 
 en partie par, Pleyte, en 1868. L'ouvrage a rendu bon service en son temps, mais les 
 progres de PEgyptologie ont ete si rapides, depuis lors, que beaucoup des donnees qu'il 
 renferme ne peuvent plus etre considerces comme exactes. M. Spiegelberg a repris les 
 manuscrits deja examines par Pleyte, il y a joint d'autres fragments qui etaient inedits 
 pour la plupart, et il a donne du tout une transcription et une interpretation excellentes 
 sur la plupart des points. G. Maspero. Revue critique 1897, N 5. 
 
 Arbeiter und Arbeiterbewegung im Pharaonenreich unter den 
 Ramessiden (ca. 14001100 v. Chr.). Eine kulturgeschichtliche Skizze. 
 Lex. 8. IV, 25 S. Mit 1 Tafel und 2 Abbildungen. 1895. ^ 1 
 
 Die Agyptische Sammlung des Museum-Meermanno-Westreenia- 
 
 num im Haag. Hrsg. u. erLautert. 4. 29 S. Mit 5 Tafeln. 1896. 
 
 Ji - 
 
 M. DaMont-Schauberg, Strassburg.
 
 
 3V 012001 
 
 RECD YRL AUC 02
 
 A 000018534 8
 
 Uni\ 
 S