practical utility of my Dictionary, and the more so as, along with many eulogistic reviews and notices, it met with some adverse criticism, especially at the hands of German Sanskṛitists.
A Sanskrit-English Dictionary
etymologically and philologically arranged
with special reference to cognate
Indo-European languages
Sir Monier Monier-Williams, M.A., K.C.I.E.
Boden Professor of Sanskrit
Hon. D.C.L. Oxon, Hon. LL.D. Calcutta, Hon. Ph.D. Göttingen
Hon. Fellow of University College and Sometime Fellow of Balliol College, Oxford
New edition, greatly enlarged and improved with the collaboration of
Professor E. Leumann, Ph. D. of the University of Strassburg
Professor C. Cappeller, Ph. D.of the University of Jena
and other scholars
Motilal Banarsidass Publishers • Private Limited • Delhi
First Edition. Published by Oxford University Press, 1899
Reprint: Delhi, 1963, 1964, 1970, 1974, 1976, 1979, 1981, 1984, 1986, 1990, 1993, 1995, 1997
© Motilal Banarsidass Publishers Private Limited. All Rights Reserved
ISBN: 81-208-0069-9 (Deluxe Ed.) ISBN: 81-208-0065-6 (Ordinary Ed.)
Also available at: MOTILAL BANARSIDASS
41 U.A. Bungalow Road, Jawahar Nagar, Delhi 110 007
8 Mahalaxmi Chamber, Warden Road, Mumbai 400 026
120 Royapettah High Road, Mylapore, Madras 600 004
Sanas Plaza, Subhash Nagar, Pune 411 002
16 St. Mark's Road, Bangalore 560 001
8 Camac Street, Calcutta 700 017
Ashok Rajpath, Patna 800004
Chowk, Varanasi 221 001Printed in India
By Jainendra Prakash Jain at Shri Jainendra Press,
A-45 Naraina, Phase I, New Delhi 110 028
and published by Narendra Prakash Jain for
Motilal Banarsidass Publishers Private Limited,
Bungalow Road, Delhi 110 007
Contents
SECTION III. Extent of Sanskrit Literature comprehended in the Present Edition.
SECTION V. Acknowledgment of Assistance Received
The dictionary order of the Nagarī letters
The first edition of this Dictionary had the advantage of being published by the Delegates of the Oxford University Press, with the support of the Secretary of State for India in Council. The present greatly enlarged and improved work enjoys the same privileges. The first edition appeared in the summer of 1872. The extent of its indebtedness to the great seven-volumed Sanskṛit-German Thesaurus compiled by the two eminent German Sanskṛitists, Otto Böhtlingk and Rudolf Roth, with the assistance of many distinguished scholars, such as Professor A.Weber of Berlin — then only completed as far as the beginning of the letter व v — was fully acknowledged by me in the Preface.
Having regard, however, to the entire originality of the plan of my own work, I did not venture to describe it as based on the great Sanskṛit-German Wörterbuch. For that plan I claimed to be alone responsible. Every particle of its detail was thought out in my own mind, and the whole work was brought to completion by me, with the co-operation of five successive assistants — whose names were duly recorded — in about twelve years from the date of my election to the Boden Professorship in the University of Oxford.
The words and the meanings of the words of a Dictionary can scarcely be proved by its compilers to belong exclusively to themselves. It is not the mere aggregation of words and meanings, but the method of dealing with them and arranging them, which gives a Dictionary the best right to be called an original production.
In saying this I am not claiming any superiority for my own method over that of the two great German Sanskṛit scholars — which, of course, has advantages of its own. Nor am I detracting one whit from the tribute of admiration which I and other lexicographers are always desirous of rendering to the colossal monument of industry and scholarship represented by their work. I am merely repeating my claim to the production of a Sanskṛit-English Dictionary on a wholly unique plan — a plan the utility of which has been now proved by experience.
It was not thought desirable to print off more than a thousand copies of the first edition of my book. These — notwithstanding the necessarily high publishing price — were sold off in a few years. It then became a question as to how the continuous demand for the Dictionary was to be met, and the Delegates decided to provide for it by a supplementary facsimile edition, produced by a photolithographic process. Copies printed by that process have been procurable ever since. Of course I was well satisfied with the factual evidence thus afforded of the practical utility of my Dictionary, and the more so as, along with many eulogistic reviews and notices, it met with some adverse criticism, especially at the hands of German Sanskṛitists.
Not that such criticisms discouraged me. On the contrary, as soon as I became aware of the likelihood of my volume becoming out of print, I set about preparations for a new edition on the very same general plan, although with an earnest determination to improve the original work by the light of such critical animadversions as seemed to me to be pertinent. And I must at once acknowledge that in these efforts I received valuable suggestions from Professor Ernst Leumann of the University of Strassburg, who was my first collaborator at an early stage of the new undertaking (see p. xxxi). It ought, however, to be put on record that, even before Professor Leumann's co-operation, I had made the discovery that the great increase in the number of printed Sanskṛit texts and works bearing on Sanskṛit scholarship, since the issue of my first edition, would entirely preclude the idea of my producing a mere “réchauffé” of my former volume, with additions, however numerous, introduced from my own interleaved copy and the contributions of fellow-Sanskṛitists. It would necessitate the re-writing of the whole from beginning to end — a formidable task, tantamount to the production of an entirely new Dictionary. This task I began to put in hand alone at least twenty years ago, and it is only due to the authorities at the India Office, under whose auspices this work was inaugurated, and with whose assistance it has been printed, that I should explain the causes which have led to the unexpected delay in its publication.
In real truth I am bound to confess that I entered upon my third lexicographical career with a little too magnificent audacity, and a little too airy hopefulness, at a time when my energies were severely tried, not only by my ordinary duties of lecturing in Sanskṛit, but by other collateral activities.
Amongst the latter it may be mentioned that I had devoted myself to researches into Indian religions and philosophies, for a series of public lectures before the University, which I felt bound to give in my capacity of Boden Professor. And I certainly could not have ventured to carry on these researches — much less to have printed them in various books as trustworthy1 — if I had not gained a “first-hand” knowledge of my subject by placing my own mind in direct touch with the mind of the learned natives of India in their own country.
It was for this and other cognate reasons2 that — with the consent and approbation of two successive Vice-Chancellors, and at my own expense — I undertook voyages to India on three several occasions (in 1875-6, 1876-7, 1883-4), and extended my travels from Bombay to Calcutta and the confines of Tibet — from Cashmere to Madras and the extreme South, including the chief homes of Buddhism in the island of Ceylon.
On each occasion I was cordially assisted by the Governor-General and local Governments of the day3. On each occasion, too, I found scattered throughout vast areas old fellow-students and pupils of my own administering immense provinces, and eager to help me in my investigations; and on each occasion I met to my surprise with learned and thoughtful natives — not only in the cities and towns, but even in remote villages — able and willing to converse with me in Sanskṛit, as well as in their own vernaculars, and to explain difficult points in their languages, literatures, religions, and philosophies.
It may well be believed that these Indian journeys were of great value in extending the horizon of my own knowledge, and increasing my power of improving the Dictionary, but it must be confessed that they interrupted its continuous prosecution.
And, in very deed, the intermittent character of my latest lexicographical career would have made its completion during my life-time almost hopeless, had I not been ably aided by successive assistants and fellow-workers, whose co-operation is gratefully acknowledged by me subsequently (p. xxxi); that of Professor C. Cappeller having extended over far the larger portion of the work.
And this is not all that I have to urge in extenuation of my apparent dilatoriness. A still more unavoidable cause of delay has been the unlooked-for amount of labour involved. This is fully explained in the Introduction (see p. xvi), but I may briefly mention here that it has consisted in adding about 60,000 Sanskṛit words to about 120,000 — the probable amount of the first edition; in fitting the new matter into the old according to the same etymological plan; in the verification of meanings, old and new; in their justification by the insertion of references to the literature and to authorities; in the accentuation of nearly every Sanskṛit word to which accents are usually applied; in the revision and re-revision of printed proofs; until at length, after the lapse of more than a quarter of a century since the publication of the original volume, a virtually new Dictionary is sent forth.
It would, of course, be unreasonable to look for perfection in the result of our combined efforts. The law of human liability to error is especially applicable to the development of a new method of any kind. Nor are the imperfections of this volume ever likely to become better known to the most keen-sighted critics than they are to the compilers themselves.
It is said of the author of a well-known Dictionary that the number of mistakes which his critics discovered in it, were to him a source of satisfaction rather than annoyance. The larger a work, he affirmed, the more likely it was to include errors; and a hypercritical condemnation of these was often symptomatic of a narrow-mindedness which could not take in the merit of any great performance as a whole.
Without having recourse to this convenient way of discomfiting critics of the Chidrânveskin type, and without abating one iota of justifiable confidence in the general trustworthiness of the present Dictionary, its compilers can yet be keenly alive to its thoroughly human character.
Speaking for myself I may say that blended with my thankfulness for the longevity that has enabled me to see a protracted undertaking brought to a completion, is a deep consciousness that I am not young enough to consider myself infallible. Indeed it is at once the joy and sorrow of every true scholar that the older he grows the more he has to confess himself a learner rather than a teacher, and the more morbidly conscious he becomes of his own liability to a learner's mistakes.
From all true scholars I do not fear, but court, criticism. Such critics will understand how a sense of responsibility may increase with advancing age, putting an author out of conceit with his own performances, and filling him with progressively intensified cravings after an impossible perfection. They will make due allowance for the difficulties besetting the production of so many densely printed pages, often comprising column after column of unbroken serried type, and abounding with countless dots and diacritical marks. Nor will they be surprised at occasional inequalities of execution in a work representing efforts spread over numerous years. Nor will they need to be reminded that occasional distractions, trials of health and weariness of spirit are unavoidably incident, not only to the responsible head of a prolonged undertaking, but to his several assistants. Indeed it is no disparagement to those who have contributed to the detail of this work to admit that a compilation, which is the result of the collaboration of so many different personalities, must in some degree reflect the idiosyncrasies and infirmities peculiar to each.
Yet notwithstanding my desire that due weight should be given to such considerations, I may be pardoned if I express my confident expectation that the volume now offered to students of both Sanskṛit and comparative philology, will supply them with the most complete and useful one-volumed Sanskṛit-English Dictionary ever yet produced — a Dictionary, too, which in its gradual progress has, I trust, kept pace with the advancing knowledge and scholarship of the day.
At all events I feel sure that I may affirm for my collaborators, as well as for myself, that we have earnestly striven to secure for this new volume, even more than for the old, the possession of four principal characteristics, namely: — I. Scholarly accuracy; 2. Practical utility; 3. Lucidity of arrangement, designed to set forth, as clearly as possible, the etymological structure of the language, and its bearing on that of the cognate languages of Europe; 4. Completeness and comprehensiveness, at least to the fullest extent attainable in the latest state of Sanskṛit research, and to the utmost limit compatible with compactness and compression into a single volume.
And here it is my duty to notify, in justice as much to my assistants as to myself, that I must be held primarily responsible, not only for the plan, but for the general character of the whole Dictionary. This will be understood when I state that I have from the first exercised a strict superintendence over the details of both editions — not only in carefully supervising the manuscript, but in adding new words, in modifying or amplifying meanings, in subjoining explanations from my own literary notes — made during my sojourning at the chief seats of learning in India — in examining and re-examining every proof-sheet.
I ought to state, however, that during occasional attacks of illness I have been 
compelled to trust more to my collaborators than at other times4; and I must also make an exception in regard to the Additions, the abundance of which is justifiable by the circumstance that many of them are taken from texts and books published quite recently. For although a manuscript list of all the words and meanings in the supplementary pages was submitted for my approval, and although many words in the list have been eliminated by me, while others have been added from my own notes, yet the necessity for passing the worst winter months in a Southern climate has made it impossible for me to have at hand every new book needed for the verification of every addition which I have allowed to be retained.
With regard to a strictly personal criticism in which I have for many years been content to acquiesce without comment, I may perhaps advantageously — now that I have nearly arrived at the end of my career — make a brief explanation. Some of my critics and a few candid friends have expressed surprise that I should have devoted so much of my long tenure of the Boden Professorship to the dry, dreary and thankless drudgery of writing Dictionaries and Grammars, and to practical researches carried on among the Pandits of India in their own country, rather than to the duty of proving the profundity of my learning and my fitness to occupy a high Professorial position by editing or translating obscure Sanskṛit texts which have never been edited or translated before5.
In explanation I must draw attention to the fact that I am only the second occupant of the Boden Chair, and that its Founder, Colonel Boden, stated most explicitly in his will (dated August 15, 1811) that the special object of his munificent bequest was to promote the translation of the Scriptures into Sanskṛit, so as “to enable his countrymen to proceed in the conversion of the natives of India to the Christian Religion6.”
It was on this account that, when my distinguished predecessor and teacher, Professor H. H. Wilson, was a candidate for the Chair in 1832, his lexicographical labours were put forward as his principal claim to election.
Surely then it need not be thought surprising, if following in the footsteps of my venerated master, I have made it the chief aim of my professorial life to provide facilities for the translation of our sacred Scriptures into Sanskṛit7, and for the promotion of a better knowledge of the religions and customs of India, as the best key to a knowledge of the religious needs of our great Eastern Dependency. My very first public lecture delivered after my election in 1860 was on “The Study of Sanskṛit in Relation to Missionary Work in India” (published in 1861).
For the rest, I have already alluded to the advantage which this Dictionary has derived from the support of the Governments of India, and I ought here to acknowledge with gratitude that, without the subsidy granted by successive Secretaries of State in Council, the present volume could not have been sold to the public at the price at which it is now offered. In regard to the Oxford University Press it will be sufficient to say that this volume adds to the countless evidences of its efficiency and of the wealth of its resources. But I may be permitted to congratulate its Delegates and Controller on their good fortune in possessing an unrivalled Oriental Press-reader in Mr. Pembrey, For more than forty years he has read the final proofs of all my books; and I can from my own experience, and without exaggeration, affirm, that I doubt whether any one can surpass him in the perfection to which he has brought the art of detecting errors due to the inadvertence of both authors and printers.
It is only necessary for me to add that having been alone responsible for the singularity of the plan of the original Dictionary, I thought it desirable to prefix to the first edition several sections of introductory explanations. In the same way my supremacy in the production of the present new work necessitates my undertaking the responsibility of writing a new series of explanations, in which I have deemed it desirable to pursue the main lines of my previous method, and not to discard any portion of the old matter which could be advantageously retained.
Indian Institute, Oxford, 1899
MONIER MONIER-WILLIAMS.
POSTSCRIPT
This Dictionary, to which my father devoted so many years of labour, was completed by him a few days before his death, which took place at Cannes, in the south of France, on April 11, 1899. It had been his hope to see this work published shortly after his return to England. Although this desire was not granted, it was a satisfaction to him to know that the last revise had received his final corrections, and that the book would be issued from the University Press within a few weeks of his death.
May 4, 1899.
M. F. MONIER-WlLLIAMS.
Statement of the circumstances which led to the peculiar System of Sanskṛit Lexicography introduced for the first time in the Monier- Williams Sanskṛit-English Dictionary of 1872.
To enable me to give a clear account of the gradual development of the plan of the present work, I must go back to its earliest origin, and must reiterate what I stated in the Preface to the first edition, that my predecessor in the Boden Chair, Professor H. H. Wilson, once intended to compile a Sanskṛit Dictionary in which all the words in the language were to be scientifically arranged under about 2,000 roots, and that he actually made some progress in carrying out that project. Such a scientific arrangement of the language would, no doubt, have been appreciated to the full by the highest class of scholars. Eventually, however, he found himself debarred from its execution, and commended it to me as a fitting object for the occupation of my spare time during the tenure of my office as Professor of Sanskṛit at the old East India College, Haileybury. Furthermore, he generously made over to me both the beginnings of his new Lexicon and a large MS. volume, containing a copious selection of examples and quotations (made by Pandits at Calcutta under his direction8) with which he had intended to enrich his own volume. It was on this account that, as soon as I had completed the English-Sanskṛit part of a Dictionary of my own (published in 1851), I readily addressed myself to the work thus committed to me, and actually carried it on for some time between the intervals of other undertakings, until the abolition of the old Haileybury College on January 1, 1858.
One consideration which led my predecessor to pass on to me his project of a root-arranged Lexicon was that, on being elected to the Boden Chair, he felt that the elaboration of such a work would be incompatible with the practical objects for which the Boden Professorship was founded9.
Accordingly he preferred, and I think wisely preferred, to turn his attention to the expansion of the second edition of his first Dictionary10 — a task the prosecution of which he eventually intrusted to a well-known Sanskṛit scholar, the late Professor Goldstücker. Unhappily, that eminent Orientalist was singularly unpractical in some of his ideas, and instead of expanding Wilson's Dictionary, began to convert it into a vast cyclopædia of Sanskṛit learning, including essays and controversial discussions of all kinds. He finished the printing of 480 pages of his own work, which only brought him to the word Ariṁ-dama (p. 87 of the present volume), when an untimely death cut short his lexicographical labours.
As to my own course, the same consideration which actuated my predecessor operated in my case, when I was elected to fill the Boden Chair in his room in 1860.
I also felt constrained to abandon the theoretically perfect ideal of a wholly root-arranged Dictionary in favour of a more practical performance, compressible within reasonable limits — and more especially as I had long become aware that the great Sanskṛit-German Wörterbuch of Böhtlingk and Roth was expanding into dimensions which would make it inaccessible to ordinary English students of Sanskṛit.
Nevertheless I could not quite renounce an idea which my classical training at Oxford had forcibly impressed upon my mind — viz. that the primary object of a Sanskṛit Dictionary should be to exhibit, by a lucid etymological arrangement, the structure of a language which, as most people know, is not only the elder sister of Greek, but the best guide to the structure of Greek, as well as of every other member of the Āryan or Indo-European family — a language, in short, which is the very key-stone of the science of comparative philology. This was in truth the chief factor in determining the plan which, as I now proceed to show, I ultimately carried into execution.
And it will conduce to the making of what I have to say in this connexion clearer, if I draw attention at the very threshold to the fact that the Hindūs are perhaps the only nation, except the Greeks, who have investigated, independently and in a truly scientific manner, the general laws which govern the evolution of language.
The synthetical process which comes into operation in the working of those laws may be well called saṁskaraṇa, “putting together,” by which I mean that every single word in the highest type of language (called Saṁskṛita11) is first evolved out of a primary Dhātu — a Sanskṛit term usually translated by “Root,” but applicable to any primordial constituent substance, whether of words, or rocks, or living organisms — and then, being so evolved, goes through a process of “putting together” by the combination of other elementary constituents.
Furthermore, the process of “putting together” implies, of course, the possibility of a converse process of vyākaraṇa, by which I mean “undoing” or “decomposition;” that is to say, the resolution of every root-evolved word into its component elements. So that in endeavouring to exhibit these processes of synthesis and analysis, we appear to be engaged, like a chemist, in combining elementary substances into solid forms, and again in resolving these forms into their constituent ingredients.
It seemed to me, therefore, that in deciding upon the system of lexicography best calculated to elucidate the laws of root-evolution, with all the resulting processes of verbal synthesis and analysis, which constitute so marked an idiosyncrasy of the Sanskṛit language, it was important to keep prominently in view the peculiar character of a Sanskṛit root — a peculiarity traceable through the whole family of so-called Āryan languages connected with Sanskṛit, and separating them by a sharp line of demarcation from the other great speech-family usually called Semitic12.
And here, if I am asked a question as to what languages are to be included under the name Āryan — a question which ought certainly to be answered in limine, inasmuch as this Dictionary, when first published in 1872, was the first work of the kind, put forth by any English scholar, which attempted to introduce comparisons between the principal members of the Āryan family — I reply that the Āryan languages (of which Sanskṛit is the eldest sister13, and English one of the youngest) proceeded from a common but nameless and unknown parent, whose very home somewhere in Central Asia cannot be fixed with absolute certainty, though the locality may conjecturally be placed somewhere in the region of Bactria (Balkh) and Sogdiana, or not far from Bokhara and the first course of the river Oxus14. From this centre radiated, as it were, eight principal lines of speech — each taking its own course and expanding in its own way — namely the two Asiatic lines: (A) the Indian — comprising Sanskṛit, the various ancient Prākṛits, including the Prākṛit of the Inscriptions, the Pāli15 of the Buddhist sacred Canon, the Ardha-Māgadhī of the Jains, and the modern Prākṛits or vernacular languages of the Hindūs, such as Hindī, Marāṭhī, Gujarātī, Bengālī, Oriya &c. (B) the Iranian — comprising the Avesta language commonly called Zand or Zend16, old Persian or Akhæmenian, Pahlavī, modern Persian, and, in connexion with these, Armenian and Pushtu; and then the six European lines: (A) Keltic, (B) Hellenic, (C) Italic, (D) Teutonic, (E) Slavonic, (F) Lithuanian, each branching into various sub-lines as exhibited in the present languages of Europe. It is this Asiatic and European ramification of the Āryan languages which has led to their being called Indo-European.
Now if I am asked a second question, as to what most striking feature distinguishes all these languages from the Semitic, my answer is, that the main distinction lies in the character of their roots or radical sounds; for although both Āryan and Semitic forms of speech are called “inflective17,” it should be well understood that the inflectiveness of the root in the two cases implies two very different processes.
For example, an Arabic root is generally a kind of hard tri-consonantal framework consisting of three consonants which resemble three sliding but unchangeable upright limbs, moveable backwards and forwards to admit on either side certain equally unchangeable ancillary letters used in forming a long chain of derivative words. These intervenient and subservient letters are of the utmost importance for the diverse colouring of the radical idea, and the perfect precision of their operation is noteworthy, but their presence within and without the rigid frame of the root is, so to speak, almost overpowered by the ever prominent and changeless consonantal skeleton. In illustration of this we may take the Arabic tri-consonantal root KTB, “to write,” using capitals for the three radical consonants to indicate their unchangeableness; the third pers. sing. past tense is KaTaBa, “he wrote,” and from the same three consonants, by means of certain servile letters, are evolved with fixed and rigid regularity a long line of derivative forms, of which the following are specimens: — KaTB, and KiTāBat, the act of writing; KāTiB, a writer; maKTūB, written; taKTīB, a teaching to write; muKāTaBat, and taKāTuB, the act of writing to one another; mutaKāTiB, one engaged in mutual correspondence; iKTāB, the act of dictating; maKTaB, the place of writing, a writing-school; KiTāB, a book; KiTBat, the act of transcribing.
In contradistinction to this, a Sanskṛit root is generally a single monosyllable18, consisting of one or more consonants combined with a vowel, or sometimes of a single vowel only. This monosyllabic radical has not the same cast-iron rigidity of character as the Arabic tri-consonantal root before described. True, it has usually one fixed and unchangeable initial letter, but in its general character it may rather be compared to a malleable substance, capable of being beaten out or moulded into countless ever-variable. forms, and often in such a way as to entail the loss of one or other of the original radical letters; new forms being, as it were, beaten out of the primitive monosyllabic ore, and these forms again expanded by affixes and suffixes, and these again by other affixes and suffixes19, while every so expanded form may be again augmented by prepositions and again by compositions with other words and again by compounds of compounds till an almost interminable chain of derivatives is evolved. And this peculiar expansibility arises partly from the circumstance that the vowel is recognized as an independent constituent of every Sanskṛit radical, constituting a part of its very essence or even sometimes standing alone as itself the only root.
Take, for example, such a root as Bhū, “to be” or “to exist.” From this is, so to speak, beaten out an immense chain of derivatives of which the following are a few examples: — Bhava or Bhavana, being; Bhāva, existence; Bhāvana, causing to be; Bhāvin, existing; Bhuvana, the world; Bhū or Bhūmi, the earth; Bhū-dhara, earth-supporter, a mountain; Bhū-dhara-ja, mountain-born, a tree; Bhū-pa, an earth-protector, king; Bhūpa-putra, a king's son, prince, &c. &c.; Ud-bhū, to rise up; Praty'-ā-bhū, to be near at hand; Prôdbhūta, come forth, &c.20
Sanskṛit, then, the faithful guardian of old Indo-European forms, exhibits these remarkable properties better than any other member of the Āryan line of speech, and the crucial question to be decided was, how to arrange the plan of my Dictionary in such a way as to make them most easily apprehensible.
On the one hand I had to bear in mind that, supposing the whole Sanskṛit language to be referable to about 2,000 roots or parent-stems21, the plan of taking root by root and writing, as it were, the biographies of two thousand parents with sub-biographies of their numerous descendants in the order of their growth and evolution, would be to give reality to a beautiful philological dream — a dream, however, which could not receive practical shape without raising the Lexicon to a level of scientific perfection unsuited to the needs of ordinary students.
On the other hand I had to reflect that to compile a Sanskṛit Dictionary according to the usual plan of treating each word as a separate and independent entity, requiring separate and independent explanation, would certainly fail to give a satisfactory conception of the structure of such a language as Sanskṛit, and of its characteristic processes of synthesis and analysis, and of its importance in throwing light on the structure of the whole Indo-European family of which it is the oldest surviving member.
I therefore came to the conclusion that the best solution of the difficulty lay in some middle course-some compromise by virtue of which the two lexicographical methods might be, as it were, interwoven.
It remains for me to explain the exact nature of this compromise, and I feel confident that the plan of the present work will be easily understood by any one who, before using the Dictionary, prepares the way by devoting a little time to a preliminary study of the explanations which I now proceed to give.
Explanation of the Plan and Arrangement of the Work, and of the Improvements introduced into the Present Edition.
Be it notified, at the very threshold, that there are four mutually correlated lines of Sanskṛit words in this Dictionary: — (1) a main line in Nāgarī type, with equivalents in Indo-Italic type22; (2) a subordinate line (under the Nāgarī) in thick Indo-Romanic type; (3) a branch line, also in thick Indo-Romanic type, branching off from either the first or the second lines with the object of grouping compound words under one head; (4) a branch line in Indo-Italic type, branching off from leading compounds with the object of grouping together the compounds of those compounds. Of course all four lines follow the usual Sanskṛit Dictionary order of the alphabet (see p. xxxvi).
The first or main line, or, as it may be called, the “Nāgarī line,” constitutes the principal series of Sanskṛit words to which the eye must first turn on consulting the Dictionary. It comprises all the roots of the language, both genuine and artificial (the genuine being in large Nāgarī type), as well as many leading words in small Nāgarī, and many isolated words (also in small Nāgarī), some of which have their etymologies given in parentheses, while others have their derivation indicated by hyphens.
The second or subordinate line in thick Indo-Romanic type is used for two purposes: — (a) for exhibiting clearly to the eye in regular sequence under every root the continuous series of derivative words which grow out of each root; (5) for exhibiting those series of cognate words which, to promote facility of reference, are placed under certain leading words (in small Nāgarī) rather than under the roots themselves.
The third or branch line in thick Indo-Romanic type is used for grouping together under a leading word all the words compounded with that leading word.
The fourth or branch Indo-Italic line is used for grouping under a leading compound all the words compounded with that compound.
The first requires no illustration; the second is illustrated by the series of words under कृ 1. kṛi (p. 300) beginning with 1. Kṛit, p. 301, col. 3, and under कर 1. kará (p. 253) beginning with 1. Karaka (p. 254, col. 1); the third by the series of compounds under कर 1. kará (p. 253, col. 1), and Káraṇa (p. 254, col 1); the fourth by the series of compounds under -vīra (p. 253, col. 3).
And this fourfold arrangement is not likely to be found embarrassing; because any one using the Dictionary will soon perceive that the four lines or series of Sanskṛit words, although following their own alphabetical order, are made to fit into each other without confusion by frequent backward and forward cross-references. In fact, it will be seen at a glance that the ruling aim of the whole arrangement is to exhibit, in the clearest manner, first the evolution of words from roots, and then the interconnexion of groups of words so evolved, as members of one family descended from a common source. Hence all the genuine roots of the language are brought prominently before the eye by large Nāgarī type; while the evolution of words from these roots, as from parent-stocks, is indicated by their being printed in thick Romanic type, and placed in regular succession either under the roots, or under some leading word connected with the same family by the tie of a common origin. It will be seen, too, that in the case of such leading words (which are always in Nāgarī type), their etymology — given in a parenthesis — applies to the whole family of cognate words placed under them, until a new series of words is introduced by a new root or new leading-word in Nāgarī type. In this way all repetition of etymologies is avoided, and the Nāgarī type is made to serve a very useful purpose.
It will also be seen that words which are different in meaning, but appear identical in form, are distinguished from each other by the figures 1, 2, 3, &c., placed before the Indo-Romanic or Indo-Italic transliterated forms : — see, for example, अशीत 1. a-śīta, अशीत 2. aśīta (p. 113)23; 1. Ā'pya, आप्य 2. āpya, आप्य 3. āpya (pp. 142, 144); बृह् 1. bṛih, बृह् 2. bṛih, बृह् 3. bṛih (p. 735).
In regard to the roots of the language, it will be observed that they are treated of in the present work — both in respect of the meanings and of the exhibition of tenses, participles, and verbal forms evolved from them — more thoroughly and exhaustively than has hitherto been attempted in a Dictionary24.
Furthermore, all the verbs formed from the roots with prepositions (as, for example, अनुकृ anu-√kṛi, p. 31, समभिव्याहृ sam-abhi-vy-ā-√hṛi, p. 1156) are arranged according to the method followed in Greek and Latin Lexicons; that is to say, such verbs are to be looked for in their own alphabetical order, and not under the roots kṛi and hṛi. The practical convenience resulting from this method, and the great advantage of exhibiting the connexion of every verb and its meanings with its derivatives, constitute in my opinion an invaluable gain, especially to the student who studies Sanskṛit as he would Greek and Latin, and makes it a guide to the study of the other members of the Indo-European family. At all events it forms one of the unique features of the present work, stamping it with an individuality of its own, and differentiating it from all other Sanskṛit Dictionaries. The labour entailed in the process — necessarily a wholly pioneering process — of thus rearranging the verbs in a language so rich in prepositions, can only be understood by those who have undergone it.
As to the separation of meanings it must be noted that mere amplifications of preceding meanings are separated by a comma, whereas those which do not clearly run into each other are divided by semicolons. A comma, therefore, must always be taken as marking separate shades of meaning, except it occurs in parenthetical observations.
Let it be observed, however, that all the meanings of a word belonging to a group are not always given in full, if they may be manifestly gathered from the other members of the group. This applies especially to participles and participial formations.
Observe too that all remarks upon meanings and all descriptive and explanatory statements are given between ( ), all remarks within remarks and comparisons with other languages between [ ].
I was told by a friendly critic, soon after the appearance of the first edition, that meanings and synonyms had been needlessly multiplied, but when the book had been fairly tested by repeated and extended application to various branches of the literature, it was found that apparently superfluous synonyms often gave the precise meanings required to suit particular passages. In the present edition — to save space — some synonyms which seemed mere surplusage have been rejected; and I fear I may have occasionally gone too far in sanctioning some of these rejections. For experience proves that the practical utility of a Dictionary is less impaired by a redundancy than by a paucity of meanings.
Again, a glance at the following pages will show that the arrangement of compound words under a leading word, as introduced in the edition of 1872, and continued with modifications in the present edition, is entirely novel.
It may perhaps be objected that there are too many of these compounds; but once more it may be urged that a Sanskṛit Dictionary must not be tried by ordinary laws in this respect, for Sanskṛit has developed more than Greek and German and any other Āryan language the faculty of forming compounds. The love of composition is indeed one of its most characteristic features. To exclude compounds from a Sanskṛit Lexicon would be, so to speak, to “unSanskṛitize” it. Not only are there certain compounds quite peculiar to Sanskṛit, but, in the grammar, composition almost takes the place of Syntax, and the various kinds of compound words are classified and defined with greater subtlety and minuteness than in any other known language of the world. When a student is in doubt whether to translate compounds like Indra-śatrn as Bahuvrīhis or Tatpurushas, the Dictionary is surely bound to aid in clearing up his perplexity. Even as it is, many useful compounds have, I fear, been sacrificed to the exigencies of space. The meanings of these, however, can be easily inferred from the meaning of their component members. Take, for example, such a word as saṁyuktâkshara, “a compound or conjunct letter.”
Another distinctive peculiarity of this Dictionary consists in the articles on mythology, literature, religion, and philosophy, scattered everywhere throughout its pages. My own collection of notes from various sources, especially those made during my three Indian journeys and published in the books named in the Preface to this volume (see p. vi, with note), have enabled me to furnish students with much useful information on many subjects not hitherto treated of in Sanskṛit Dictionaries. It will, I feel sure, be admitted that the knowledge gained by me from personal contact with Indian Pandits and educated men in their own Universities, and with all sorts and conditions of Hindūs in their own towns and villages, has been a distinct advantage to this Dictionary. It has enabled me to give much useful information not found in other works, and to avoid many mistakes made by Sanskṛitists who have only a book-knowledge of India.
A further peculiar feature is the introduction of a large number of names of persons and places. This may be objected to as a needless extension of the scope and limits of a Dictionary. In extenuation I contend that greater liberty ought to be allowed to a Sanskṛit Dictionary in this respect than to Greek and Latin Lexicons, because Oriental alphabets have no capital letters enabling such names to be distinguished from ordinary nouns.
Then again, in regard to the names of works, which are also multiplied to an unusual extent, Sanskṛit literature is so vast that, although — as I hold — very little worthy of attention remains to be edited, yet it may often be of great importance to have attention drawn to unknown treatises, or to commentaries on well-known works ascertained to exist in manuscript in the libraries of Europe or India.
As to plants and trees, the adjective qualifying the name of a plant, as well as the name of the plant itself, ought occasionally to be marked, according to the rules of botanical science, with an initial capital letter. But it is often difficult for a non-botanist to decide as to the correct usage. It was therefore thought better to use capital letters for both substantive and adjective, especially as in the new edition, to save space, the word “plant” is omitted. Hence the second capital letter, though often inappropriate, serves as a symbol for denoting that the epithet is that of a plant.
I need scarcely draw attention to the comparisons from cognate languages which manifestly constitute a special feature of this volume. Many doubtful comparisons have been eliminated from the present edition. A few questionable ones have, I fear, been retained or rashly inserted, but they will be easily detected (e.g. under Ayāsya, p. 85).
In regard to what may be thought a needless multiplication of indecent words and meanings, offensive to European notions of delicacy, I am sorry to say that they had to be inserted, because in very truth Sanskṛit, like all Oriental languages, abounds with words of that character, and to such an extent, that to have omitted them, would have been to cut out a large percentage of the language. A story is told of a prudish lady who complimented Dr. Johnson on having omitted all bad words from his English Dictionary; whereupon he replied: “Madam, it is true that I have done so, but I find that you have been looking for them.” In point of fact students of Sanskṛit literature cannot sometimes avoid looking for such words. Nor have I, except in rare instances, veiled their meaning under a Latin translation which only draws attention to what might otherwise escape notice.
In extenuation it may fairly be urged that in India the relationship between the sexes is regarded as a sacred mystery, and is never held to be suggestive of improper or indecent ideas.
After the foregoing explanation of the general plan of the work it remains to describe some of the more noteworthy changes and improvements introduced into the present edition.
And let me at once say that, as it was intended to give explanations of even more Sanskṛit words than are treated of in the great Wörterbuch of Böhtlingk and Roth, and in the later Wörterbuch of the former, and, as it was decided that to prevent its expansion beyond the limits of one compact volume, the number of pages in the new edition should not be augmented by much more than a hundred and fifty, it became a difficult problem to devise a method of making room for the ever-increasing number of words which, as the work grew under our hands, continually pressed more and more for admission into its purview.
Let any critic, then, who may feel inclined to pass a severe judgment on the contrivances for abbreviation in the present edition of the Dictionary, think for a moment of the difficulties in which its compilers found themselves involved. It was only gradually that the actual fact revealed itself- — the very startling fact that we had to provide for the treatment of about one half more Sanskṛit words, simple and compound, than in the first edition. That is to say, calculating as I had done that the number of Sanskṛit words — simple and compound — in the first edition amounted to about 120,000, it became evident to us, as the work proceeded, that the number to be provided for in the new edition could not be reckoned at less than 180,000. It was as if a builder employed in repairing one of his own buildings had been told that he had to provide for the crowding of 1,800 human beings into a room, originally constructed by him to hold only twelve hundred.
Or perhaps the difficulty may be better illustrated thus: — A traveller, after having made a voyage round the world, starts some time afterwards for a second similar journey. The rules of the ship in which he embarks only permit of his taking a limited amount of baggage into his cabin, and naturally his first idea is to take the same box which accompanied him on the first occasion. Into this he begins by packing his possessions, with perhaps a little more compression than before. He soon finds, however, that the lapse of time has added to his acquisitions, and that no close packing will enable him to make room for them. What then is he to do? He is permitted to make his one box a little longer and deeper; but even then he has not room enough. His only resource is to make his one receptacle hold more by filling up every crevice, and fitting one article into the other by various ingenious devices.
This is an illustration of the difficulties encountered in the process of compressing the immense mass of new matter which had to be brought within the compass of the new edition. It has been possible to lengthen the pages of the new volume by about an inch, so that each column now contains about eight or nine lines more than in the first edition, and the volume has been increased in thickness by more than one hundred pages (and with the Addenda by 147 pages). These enlargements have given considerable additional space, but not nearly as much as was needed. All sorts of contrivances for contracting, abridging, and abbreviating had, therefore, to be adopted, so as to secure the greatest economy of space without impairing the completeness of the work — considerations which will, I hope, be a valid excuse for the occasional violations of uniformity which forced themselves upon us, as the need for greater comprehensiveness, within a limited circumference, became more and more imperative.
Perhaps the necessity for such measures will be better understood if I here enumerate some of the sources whence the additional matter in the present volume has been derived.
Imprimis, all the latter portion of the great seven-volumed Wörterbuch of the two great German lexicographers beginning with the letter व v. Next, all the additions in Geheimrath von Böhtlingk's later compilation, and especially his Nachträge. Then all my own manuscript Addenda in the interleaved copy of my first edition25; and lastly all the words from many important pure Sanskṛit and Buddhistic Sanskṛit works printed and published in recent years, most of which will be named in the sequel.
Doubtless, therefore, in describing the improvements which mark this new Dictionary, the first place should be given to the vast mass of new matter introduced into it. This I venture to assert, after a somewhat rough calculation, amounts to very little short of 60,000 additional Sanskṛit words with their meanings.
And a still further increase has resulted from the introduction of references to authorities, and to those portions of the literature in which the words and meanings recorded in the Dictionary occur. The reason given by me for abstaining from more than a few such references in the first edition, was that abundant quotations were to be found in the great seven-volumed Thesaurus — so often named before — which all who used my Dictionary could easily find means of consulting. In real fact, however, not a few words and meanings in the earlier portion of the first edition of my book were entered on the authority of Professor H. H. Wilson, while many more in the middle and towards the end were inserted from sources investigated independently by myself, and were not supported by any of the quotations given in the Thesaurus. It followed as a matter of course that, very soon after the publication of my first edition in 1872, the almost entire absence of independent references of my own was animadverted upon regretfully by even friendly critics.
Naturally, therefore, I determined to remedy an evident defect by introducing a large number of references and quotations into the new edition. Nor is it surprising that this determination grew and strengthened in the course of execution, so much so, indeed, that after the printing of page 60 I decided, with Professor Leumann's co-operation, to give no words and no series of meanings without quoting some authority for their use, or referring to the particular book or portion of literature in which they occur.
And further, it became a question whether we were not bound to indicate by a reference in every case not merely the particular books, but the chapter and line in which each word was to be found, and sometimes even to quote entire passages. This, in fact, as will be seen, has been occasionally done, but it soon became evident, that the immense copiousness of Sanskṛit literature — a copiousness far exceeding that of Greek and Latin — would preclude the carrying out of so desirable an object in full, or even to a somewhat less extent than in the great St. Petersburg Thesaurus — unless indeed my new Dictionary was to be enlarged to a point beyond the limits of a single compact volume. Nay, it soon became clear that the exigencies of space would make the mere enumeration of all the works in which a word occurs impossible. In the end it was found that the use of the symbol &c., would answer all the purpose of a full enumeration.
Hence it must be understood that RV.26 &c. &c. denotes that a word occurs in the whole literature — both Vedic and Post-Vedic — beginning with the Rig-veda, while Mn. &c. signifies that the use of a word is restricted to the later literature beginning with Manu.
And again, when a word had not yet been met with in any published literary work, but only in native lexicons, it was decided to denote this by the letter L.
As to the words and meanings given on my authority and marked MW., many of them have been taken by me from commentaries or from the notes which I made after conversations with learned Pandits in their own country. For it seems to me that Sanskṛit Dictionaries ought sometimes to give important modern words and meanings as used by modern educated Sanskṛit scholars in India — such, for example, as the meaning of prâṇa-pratishṭhā27 (see Additions under Prâṇa, p. 1330).
Then a third improvement in the present edition, as every true scholar will admit, is the accentuation of words occurring in accentuated texts, although it will be found, I fear, that occasional accidental omissions occur, and in cross-references the accent has often been designedly dropped. Many accents, too, which are only known from Pāṇini and the Phiṭ-sūtras have been intentionally omitted.
It is admitted that accentuation is marked only in the oldest Vedic texts, and that in later times it must have undergone great changes — so far at least as the spoken accent was concerned. And this led me to decide that in preparing a practical Dictionary which employed so many complicated diacritical marks, it would be better not to increase the complication by adding the marks of accentuation. All accentuation was, therefore, designedly omitted in the first edition. But the careful study of Pāṇini's grammar, which my higher lectures, during the period of my active occupancy of the Boden Chair (1860-1888), obliged me to carry on, forced upon me the conviction that, inasmuch as at the time when the great Indian Grammarian — the chief authority for both Vedic and classical grammar — elaborated his wonderful system, every word in Sanskṛit, as much in the ordinary language as in the Vedic, had its accent28, a knowledge of accents must be often indispensable to a right knowledge of the meaning of words in Sanskṛit.
And in real truth the whole of Pāṇini's grammar is interpenetrated throughout by the ruling idea of the importance of accentuation to a correct knowledge of words and their meanings.
For example, we learn from Pāṇ. vi, 1, 201, that the word ksháya means “abode,” but kskayá with the accent on the last syllable means “destruction.” And again, from Pāṇ. vi, 1, 205, that datta, “given,” which as a p. participle has the accent on the second syllable (dattá) is accentuated on the first syllable (i.e. is pronounced dátta) when it is used as a proper name. On the other hand, by Pāṇ. vi, 1, 206, dhṛíshṭa has the accent on the first syllable, whether as a participle, or as a name (not dhṛishṭá at p. 519).
Further, by Pāṇ. vi, 1, 223 and vi, 2, 1 all compounds have different meanings according to the position of the accent. Hence Indra-śatru means either “an enemy of Indra” or “having Indra as an enemy,” according as the accent is on the last or first member of the compound (Indra-śatrú or Índra-śatru; see Additions, p. 1321). These examples may suffice to show the importance of accentuation in affecting meanings.
That this holds good in all languages is shown by the careful way in which accentuation is marked in modern English Dictionaries. How, indeed, could it be otherwise when the transference of an accent from one syllable to another often makes such important alteration in the sense as may be noted in the words “gállant” and “gallánt,” “récord” and “recórd” “présent” and “presént,” “aúgust” and “augúst,” “désert” and “desért.” The bearing, too, of Sanskṛit accentuation on comparative philology will be evident to any one who has noted the coincidences between the accentuation of Greek and Sanskṛit words.
Manifestly then it would have been inexcusable had we omitted all accentuation in the present enlarged and improved work29. It must be admitted, however, that incidence of accent has not been treated with exact uniformity in every page of this volume.
In Pāṇini's system, as is well known, the position of the accent is generally denoted by some indicatory letter, attached to the technical names given by him to his affixes and suffixes, including the terminations of verbs and of verbal derivatives (called pratyaya). Thus, by Pāṇ. vi, i, 163 the letter c added to a suffix (as in ghurac, Pāṇ. iii, 2, 161), indicates that the derivative bhaṅgura formed by that suffix is accented on the last syllable (e.g. bhaṅgurá).
In Vedic texts printed in Nāgarī character the accents are denoted by certain short lines placed above and below the letters, but in the present Dictionary we have not thought it necessary to mark the accent of words printed in Nāgarī, but only of their equivalents in Romanic and Italic type, the common Udātta or acute accent being marked by ʹ, and the rarer Svarita by ̀.
And in this connexion it should be mentioned that the employment of the long prosodial mark ( ¯ ) to denote long vowels (e.g. ā) has manifestly one advantage. It enables the position of an accent to be indicated with greater clearness in cases where it falls on such vowels (e. g. ā).
Next to the three principal improvements thus explained ought certainly to be reckoned the increased mechanical aids provided for the eye, to facilitate the search for words in pages overcrowded with complicated and closely printed type. And most conspicuous among these aids is the employment of thick “Clarendon” type (see p. xiv, note I) in place of the Italics of the previous edition, both for the derivatives under roots and under leading words and for the compounds under such words; thus allowing the Italic type to be reserved for compounds of compounds.
Then another improvement of the same kind has been effected by the distribution of the compounds belonging to leading words under two, three, or even more separate heads, according to the euphonic changes in the finals of these words. Thus in the first edition all the compounds belonging to the leading word Bahis were arranged under the one word Bahis (=Vahis); but in the present edition these compounds are far more readily found by their segregation under the five heads of Bahis, Bahiḥ, Bahir, Bahiś, and Bahish (see pp. 726, 727).
Furthermore, among useful changes must be reckoned the substitution of the short thick line (not necessarily expressive of a hyphen30) for the leading word in all groups of compounds whose first member is formed with that leading word. Take, for example, such an article as that which has the leading word Agni, at pp. 5, 6. It is easy to see that the constant repetition of Agni in the compounds formed with that word was unnecessary. Hence -kaṇa, -karman &c. are now substituted for Agni-kaṇa, Agni-karman &c. By referring to such an article as Mahā, at pp. 794-802, an idea may be formed of the space economized by this simple expedient.
And here I must admit that a few changes may possibly be held to be doubtful improvements, the real fact being that they have been forced upon us by the necessity for finding room for those 60,000 additional Sanskṛit words with their meanings, the accession of which to the pages of the Dictionary — as already mentioned — became a paramount duty.
For instance, towards the end of the work, the exigencies of space have compelled us to use Italics with hyphens, not only in the case of sub-compounds (as, for example, -maṇi-maya under canđra-kānta at p. 386, col. 3, is for candrakānta-maṇi-maya), but also in the case of compounds falling under words combined with prepositions (as, for example, under such words as 2. Vi-budha, Vi-bhāga, at p. 977).
The same exigencies of space compelled us to group together all words compounded with 3. vi (see p. 949) and with 7. sa (see under sa-kaṅkaṭa, p. 1123 &c.).
The same considerations, too, have obliged us to make a new departure in extending the use of the little circle ° to English words. Its ordinary use, of course, is to denote that either the first or last part of a Sanskṛit word has to be supplied. For instance, such a word as keśa-v°, coming after I. Vapanīya at p. 919 stands for keśa-vapanīyay while °da, °data, °dasva after codati, at p. 400, are for coda, codata, codasva; and similarly °dyotana under Pra-dyota at p. 680 is for Pra-dyotana.
The application of this expedient to English words has enabled us to effect a great saving. It must be understood that this method of abbreviation is only applied to the leading meaning which runs through a long article, or to English words in close juxtaposition. For example, the leading signification of ratha under the article 1. rátha (p. 865) being “chariot,” this is shortened to “ch°” in the remainder of the article; and “clarified butter” in one line is shortened to “cl° b°” in the next. By referring to such an article as sahasra, at p. 1195, it will be seen what a gain in space has thus been effected.
In cases like -°nśa under kalā (p. 261) the ° denotes that -°nśa is not a complete word without the prefixing of a, which is not given because it has become blended with the final ā of the leading word kalā.
Much space, too, has been gained by the application of the symbols 
(adopted at Professor Leumann's suggestion) to denote the blending of short and long vowels. Thus denotes the blending of two short vowels (as of a + a into 
); denotes the blending of a short with a long vowel (as of a + ā into 
); denotes the blending of a long with a short (as of ā + a into 
); denotes the blending of two long vowels (as of ā + ā into 
), and so with the other vowels, e.g. ê for a + i, ô for a + u, 
for a + ū &c. (see for example kṛitâtgni for kṛita + agni, kṛitôdaka for kṛita + udaha, at p. 303).
A further economy has been effected by employing the symbol √ for root.
In this new edition, too, the letters “mfn.” placed after the crude stems of words, have been generally substituted for the forms of the nominative cases of all adjectives, participles, and substantives (at least after the first 100 pages), such nominative forms being easily inferred from the gender. But it must be borne in mind that nearly all feminine stems in ā and ī are also nominative forms. In cases where adjectives make their feminines in ī this has been generally indicated, as in the previous edition. Occasionally, too, the neuter nominative form (am) is given as an aid to the eye in marking the change from one gender to another.
Other contrivances for abbreviation scarcely need explanation; for instance, “N.” standing for “name” is applicable to epithets as well as names, and when it applies to more than one person or object in a series, is omitted in all except the first; e.g. “N. of an author, RV.; of a king, MBh.” &c.
Also, the figures 1, 2, 3 &c. have been in some cases dropped (see note 1, p. xv), and the mention of cl. 8 is often omitted after the common root kṛi.
Finally, I have thought it wise to shorten some of the articles on mythology, and to omit some of the more doubtful comparisons with the cognate languages of Europe.
Extent of Sanskṛit Literature comprehended in the Present Edition.
I stated in the Preface to the first edition of this work — written in 1872 — that I had sometimes been asked by men learned in all the classical lore of Europe, whether Sanskṛit had any literature. Happily, since then, a great advance in the prosecution of Indian studies and in the diffusion of a knowledge of India has been effected. The efforts and researches of able Orientalists in almost every country have contributed to this result, and I venture to claim for the Oxford Indian Institute and its staff of Professors and Tutors a large share in bringing this about.
Nevertheless much ignorance still prevails, even among educated English-speakers, in respect of the exact position occupied by Sanskṛit literature in India — its relationship to that of the spoken vernaculars of the country and the immensity of its range in comparison with that of the literature of Europe. I may be permitted therefore to recapitulate what I have already said in regard to the term “Sanskṛit,” before explaining what I conceive ought to be included under the term “Sanskṛit literature.”
By Sanskṛit, then, is meant the learned language of India — the language of its cultured inhabitants — the language of its religion, its literature, and science — not by any means a dead language, but one still spoken and written by educated men in all parts of the country, from Cashmere to Cape Comorin, from Bombay to Calcutta and Madras31. Sanskṛit, in short, represents, I conceive, the learned form of the language brought by the Indian branch of the great Āryan race into India. For, in point of fact, the course of the development of language in India resembles the course of Āryan languages in other countries, the circumstances of whose history have been similar.
The language of the immigrant Āryan race has prevailed over that of the aborigines, but in doing so has separated into two lines, the one taken by the educated and learned classes, the other by the unlearned — the latter again separating into various provincial sub-lines32. Doubtless in India, from the greater exclusiveness of the educated few, and the desire of a proud priesthood to keep the key of knowledge in their own possession, the language of the learned classes became so highly elaborated that it received the name Saṁskṛita, or “perfectly constructed speech” (see p. xii), both to denote its superiority to the common dialects (called in contradistinction Prākṛita) and its more exclusive dedication to religious and literary purposes. Not that the Indian vernaculars are exclusively spoken languages, without any literature of their own; for some of them (as, for example, Hindī, Hindūstānī, and Tamil, the last belonging to the Drāviḍian and not Āryan family) have produced valuable literary works, although their subject-matter is often borrowed from the Sanskṛit.
Next, as to the various branches of Sanskṛit literature which ought to be embraced by a Dictionary aiming, like the present, at as much completeness as possible — these are fully treated of in my book “Indian Wisdom” (a recent edition of which has been published by Messrs. Luzac & Co.). It will be sufficient therefore to state here that Sanskṛit literature comprises two distinct periods, Vedic and Post-Vedic, the former comprising works written in an ancient form of Sanskṛit which is to the later form what the language of Chaucer is to later English.
Vedic literature begins with the Rig-veda (probably dating from about 1200 or 1300 B.C.), and extending through the other three Vedas (viz. the Yajur, Sāma, and Atharva-veda), with their Brāhmaṇas, Upanishads, and Sūtras, is most valuable to philologists as presenting the nearest approach to the original Āryan language. Post-Vedic literature begins with the Code of Manu (probably dating in its earliest form from about 500 B.C.), with its train of subsequent law-books, and extending through the six systems of philosophy, the vast grammatical literature, the immense Epics33; the lyric, erotic, and didactic poems, the Nīti-śāstras with their moral tales and apothegms, the dramas, the various treatises on mathematics, rhetoric, prosody, music, medicine, &c., brings us at last to the eighteen Puranas with their succeeding Upa-purāṇas, and the more recent Tantras, many of which are worthy of study as repositories of the modern mythologies and popular creeds of India. No one person, indeed, with limited powers of mind and body, can hope to master more than one or two departments of so vast a range, in which scarcely a subject can be named, with the single exception of Historiography, not furnishing a greater number of texts and commentaries or commentaries on commentaries, than any other language of the ancient world. To convince one's self of this one need only glance at the pages of the present Dictionary, and note the numerous works named there, which, if the catalogue were complete, would probably amount to a total number not far short of the 10,000 which the Pandits of India are said to be able to enumerate.
Nor is it their mere number that astonishes us. We are appalled by the length of some of India's literary productions as compared with those of European countries. For instance, Virgil's Æneid is said to consist of 9,000 lines, Homer's Iliad of 12,000 lines, and the Odyssey of 15,000, whereas the Sanskṛit Epic poem called Mahā-bhārata contains at least 200,000 lines, without reckoning the supplement called Hari-vanśa34. In some subjects too, especially in poetical descriptions of nature and domestic affection, Indian works do not suffer by a comparison with the best specimens of Greece and Rome, while in the wisdom, depth, and shrewdness of their moral apothegms they are unrivalled.
More than this, the Hindūs had made considerable advances in astronomy, algebra, arithmetic, botany, and medicine, not to mention their superiority in grammar, long before some of these sciences were cultivated by the most ancient nations of Europe. Hence it has happened that I have been painfully reminded during the progress of this Dictionary that a Sanskṛit lexicographer ought to aim at a kind of quasi omniscience. Nor will any previous University education, such at least as was usual in my youth, enable him to explain correctly the scientific expressions which — although occasionally borrowed from the Greeks — require special explanation.
In answer then to the question: What extent of Sanskṛit literature is comprehended in this Dictionary? I reply that it aims at including every department, or at least such portions of each department as have been edited up to the present date.
And here 1 must plainly record my conviction that, notwithstanding the enormous extent of Sanskṛit literature, nearly all the most important portions of it — Vedic or Post-Vedic — worthy of being edited or translated have been already printed and made accessible in the principal public libraries of the world35.
No doubt the vast area of India's philosophical literature has not yet been exhaustively explored; but its most important treatises have been published either in India or in Europe. In England we may appeal with satisfaction to the works of our celebrated scholar Colebrooke, of the late Dr. Ballantyne, and more recently of such writers as E. B. Cowell, A. E. Gough, and Colonel Jacob, all of whom have contributed to the elucidation of this most difficult, but most interesting branch of study, while among Continental scholars the names of Deussen, Garbe, and Thibaut are most distinguished.
There is also much still to be done in what may be called Epigraphic or Inscription literature, in which Dr. Fleet, Dr. E. Hultzsch, and Professor F. Kielhorn are labouring so effectively. And I am happy to say that we have occasionally availed ourselves of their labours in the following pages.
The Tantras, too, present a field of research almost wholly untrodden by European scholars, and these books at one time attracted much curiosity as likely to present a hopeful mine for exploitation. I therefore, during my Indian journeys, searched everywhere for good MSS. of the most popular Tantras, with a view to making the best procurable example of them better known in Europe by a good printed edition and translation. Everywhere I was told that the Rudra-yāmala Tantra was held in most esteem36. But after a careful examination of its contents I decided that it was neither worth editing nor translating (see my “Brāhmanism and Hindūism.” pp. 205-208).
As to translations, the long array of “Sacred Books of the East” might well be supposed to have exhausted the whole reservoir of Sanskṛit works worthy of being translated; even admitting that the entire range of Sanskṛit literature is held to be more or less sacred. Yet the series is still incomplete37.
Assuming then my opinion on this point to be correct, I think I may fairly claim for the present Dictionary as great an amount of comprehensiveness as existing circumstances make either possible or desirable. Of course the earlier part of the work must perforce be less complete than the later. Nor can it be said to deal with every branch of literature with equal thoroughness, but its defects are, I hope, fairly remedied by the ample Additions at the end of the volume.
Reasons for applying the Roman Alphabet to the expression of Sanskṛits with an account of the Method of Transliteration employed in the Present Dictionary.
As I cherish the hope that this Dictionary may win its way to acceptance with the learned natives of India, I must ask European scholars to pardon my diffuseness if I state with some amplitude of detail my reasons for having applied the Roman or Latin alphabet to the expression of Sanskṛit more freely than any other Sanskṛit lexicographer.
For indeed I know full well that all who belong to the straitest sect of Hindū scholars will at once flatly deny that their divine Sanskṛit can with any propriety be exhibited to the eye clothed in any other alphabetical dress than their own “divine Nāgarī.” Na hi pūtaṁ syād go-kshīraṁ śva-dṛitau dhṛitam, “let not cow's milk be polluted by being put into a dog's skin.” How can it possibly be, they will exclaim, that the wonderful structure of our divine language and the subtle distinctions of its sacred sounds can be properly represented by such a thoroughly human and wholly un-Oriental graphic system as a modern European alphabet?
Let me, then, in the first place point out that our so-called European alphabet, as adopted by the Greeks, Romans, and modern nations of Europe, is really Asiatic, and not European in its origin. And secondly, let me try to show that it has certain features which connect it with the so-called divine Nāgarī alphabet of the Brāhmans. Nay more, that it is well suited to the expression of their venerated Sanskṛit; while its numerous accessory appliances, its types of various kinds and sizes, its capital and small letters, hyphens, brackets, stops &c., make it better suited than any other graphic system to meet the linguistic requirements of the coming century — a century which will witness such vast physical, moral, and intellectual changes, that a new order of things, and almost a new world and a new race of beings, will come into existence. In that new world some of the most inveterate prejudices and peculiarities now separating nation from nation will be obliterated, and all nationalities — brought into fraternal relationship — will recognize their kinship and solidarity.
Even during the present century the great gulf dividing the West from the East has been partially bridged over. Steam and electricity have almost destroyed the meaning of differences of latitude and longitude; and nations which were once believed to be actually and figuratively the antipodes of each other have been brought to feel that mere considerations of distance are no obstacles to the reciprocal interchange of personal intercourse, and no bar to the adoption of all that is best in each other's customs and habits of thought.
And a still more remarkable event has happened. Europe has learnt to perceive that in imparting some of the benefits of her modern civilization to Eastern races, she is only making a just return for the lessons imparted to her by Asiatic wisdom in past ages.
For did she not receive her Bible and her religion from an Eastern people? Did not her system of counting by twelves and sixties come to her from Babylonia, and her invaluable numerical symbols and decimal notation from India through the Arabs ? Did not even her languages have their origin in a common Eastern parent? It cannot, therefore, be thought surprising if her method of expressing these languages by graphic symbols also came to her from an Eastern source.
We cannot, indeed, localize with absolute certainty the precise spot whence issued the springs of that grand flow of speech which spread in successive waves — commencing with the Sanskṛit in Asia and the Keltic in Europe — over a large proportion of those two continents. Nor can we fix, beyond all liability to question, the local source of the first known purely phonographic alphabet. But we stand on sure ground when we assert that such an alphabet is to be found inscribed on Phœnician monuments of a date quite as early as the cognate Moabite inscription on the stone of King Mesha, known to belong to the middle of the ninth century B.C.38
It was of course a priori to be expected that Phœnicia — one of the chief centres of trade, and the principal channel of communication between the Eastern and Western worlds in ancient times — should have been compelled to make use of graphic symbols of some kind to enable her to carry on her commercial dealings with other nations; and it may fairly be conjectured that a mere system of ideograms would have been quite unsuited to her needs. But this does not prove that the phonographic signs on Phœnician inscriptions were invented all at once, without any link of connexion with previously current ideographic prototypes. And it is certainly noteworthy that the discovery at Tel-el-Amarna in Egypt of letters from an ancient king of Jerusalem written on tablets in the early Babylonian cuneiform script39 proves that a Babylonian form of ideographic writing existed in Palestine and the neighbourhood of Phœnicia as early as the fifteenth century B.C.
Those, however, who have conjectured that the Phœnician phonograms were developed out of the Babylonian cuneiform symbols, cannot be said to support their hypothesis by any satisfactory proof, literary or epigraphic.
Nor does the theory which makes the South Semitic or Himyaritic scripts40 the precursors and prototypes of the Phœnician seem to rest on sufficiently clear evidence.
On the other hand it is certain that if we investigate the development of the Egyptian hieroglyphic ideograms, we shall find that they passed into a so-called “hieratic” writing in which a certain number of phonograms were gradually introduced. And it is highly probable that Phœnicia in her commercial intercourse with a country so close to her shores as Egypt, or perhaps through a colony actually established there, became acquainted in very early times with this Egyptian hieratic script.
Furthermore, a careful comparison of the elaborate tables printed in the latest edition of the Encyclopaedia Britannica, and in the Oxford “Helps to the Study of the Bible” — giving the Egyptian and Phœnician symbols side by side — tends no doubt to show a certain resemblance of form between five or six of the Phœnician and corresponding Egyptian letters.
Nevertheless, the comparison by no means makes it clear that all the Phœnician letters were derived from Egyptian models41, nor does it invalidate the fact that existing epigraphic evidence is in favour of regarding Phœnicia as practically the inventor of that most important factor in the world's progress — a purely phonographic alphabet.
Here, however, I seem to hear some learned native of India remark: — It may be true that the Phœnician inscriptions are prior in date to those hitherto discovered in India; but do you really mean to imply that India's admirably perfect Deva-nāgarī alphabet, which we hold to be a divine gift42, was borrowed from the imperfect alphabet of a nation of mere money-making traders, like the Phœnicians? Is it not the case that the earliest elements of civilization and enlightenment have always originated in the East, and spread from the East to the West — not from the West to the East? And if, as is generally admitted, the symbols for numbers, which were as essential to the world's progress as letters, originated in India and passed through Semitic countries into Europe, why should not alphabets have had the same origin and the same course? Did not the Hindūs invent for themselves their own grammar, their own science of language, their own systems of philosophy, logic, algebra, and music ? Have they not an immense literature on these and other subjects, much of which must have been written down at least 600 years B.C.? And are there not references in this literature to the existence of writing in India in very ancient times? for instance, in the Vāsishṭha Dharma-sūtra of the later Vedic period, in the Laws of Manu43, in Pāṇini, who lived about 400 B.C.44, in the Pāli Canon of the Buddhists which refers to writing schools and writing materials45. And again, do not the actual inscriptions of King Aśoka of the third century B.C. exhibit a remarkably perfect system of alphabetical signs, and many varying forms in different districts of India, postulating several centuries of antecedent development46? And if no Indian epigraphs of an earlier date than the reign of Aśoka have yet been discovered, is not that due to the circumstance that the art of incising letters on stone and metal only came into use when great Hindū kings arose, whose empire was sufficiently extensive to make it necessary to issue edicts and grants to their subjects? Bearing all this in mind, may it not be contended that if there has been any plagiarism in the matter of alphabets, the borrowing may have been from the Hindūs rather than by them?
Such questions as these have often been addressed to me by learned Pandits, and it must be confessed that they are by no means to be brushed aside as unworthy of consideration. Quite the reverse. They contain many statements to which no exception can be taken. But my present object is not to furnish incontestable proof of the derivation of Indian alphabets from a Phœnician source. It is rather to point out to Indian scholars that even admitting (with some eminent authorities) that there is good ground for claiming an indigenous origin for Hindū alphabets, many of the letters composing them offer points of contact and affinity with those of Phœnicia, and therefore with those of Greece and Rome and modern Europe.
And at the outset it must be frankly acknowledged that the first phonographic alphabet brought to light on ancient Phœnician monuments constituted by no means a perfect alphabetic system. It had, no doubt, advanced beyond the ideographic stage, and even to some extent beyond the syllabic, but its phonograms were only twenty-two in number, and mainly represented consonants. It had not attained to the level of an alphabet in which vowel symbols are promoted to an equality of representation with consonantal, and treated as compeers, not as mere secondary appendages. And even to this day, the Semitic alphabets connected with the Phœnician — viz. the Hebrew, Aramæan, and Arabian — are nearly as imperfect, and very little better than, so to speak, consonantal skeletons, wanting the life-blood which vowels only can impart.
Indeed, the imperfection of the Phœnician script is well shown by the fact that the Greeks who, as every one admits, were indebted to the Phœnicians for their rudimentary consonantal method of writing, had no sooner received it (probably quite as early as 800 B. C.) than they began to remedy its defects, and gradually developed out of it a true alphabetic method of their own, which was ultimately made to flow from left to right in opposition to the Semitic method.
Similarly, too, the Romans when they had accepted the Phœnician graphic signs from the Greeks, found it necessary to improve upon them, and ultimately developed out of them an even more practical alphabetic system.
But surely these two facts may be appealed to as making it not improbable that if the Greeks and Romans, two highly intellectual races, sprung from the same Āryan stock as the Brāhmans, condescended to accept certain rudimentary phonograms from the Phœnicians, and to expand them into alphabets suited to the expression of their own languages, the Brāhmans also might have deigned, if not to accept a foreign alphabet, at least to improve their own graphic system by modifications introduced through contact with Semitic races.
Nor should it be forgotten that in later times the Hindūs did actually borrow a Semitic alphabet from Arabia for the expression of their vernacular Hindī47.
No doubt it must be admitted that, had any overmastering conviction of the necessity for the general use of written signs taken hold of the Hindū mind in early times, India would not have consented to be beholden to other countries for even improvements in her own forms of writing.
But the most patriotic of India's patriots must acknowledge that the Hindūs have always preferred oral to written communications. Indeed, although a vast literature exists in Sanskṛit, no word exists exactly corresponding to our English word “literature48;” and even if such a word were available, true Indian Pandits would prefer to designate the immense series of their sacred books by such words as Veda, or Vidyā (from vid, “to know”), Śruti (from śru, “to hear”), Śāstra (from śās, “to teach”), Smṛiti (from smṛi, “to remember”); the reason being that, like Papias, Bishop of Hierapolis (whose date, according to Dean Farrar, is 140 A.D.), they consider “that the things from books are not so advantageous as things from the living and abiding voice.” Nor must we forget that the climate of India was unfavourable to the preservation of such writing material as existed in ancient times.
And besides this may it not be conjectured that the invention and general diffusion of alphabetic writing was to Indian learned men, gifted with prodigious powers of memory, and equipped with laboriously acquired stores of knowledge, very much what the invention and general use of machinery was to European handicraftsmen? It seemed to deprive them of the advantage and privilege of exercising their craft. It had to be acquiesced in, and was no doubt prevalent for centuries before the Christian era, but it was not really much encouraged. And even to this day in India the man whose learning is treasured up in his own memory is more honoured than the man of far larger acquirements, whose knowledge is either wholly or partially derived from books, and dependent on their aid for its communication to others49.
It seems, therefore, not unreasonable to assume that, when the idea of the necessity for inventing alphabetic signs began to impress itself on the minds of Semitic races, it had not taken such deep root among the inhabitants of India as to lead to the invention or general adoption of any one fixed system of writing of their own. It seems, indeed, more probable that learned men in that country viewed the art of writing too apathetically to make a stand against the introduction of alphabetical ideas from foreign sources.
At all events there can be no antecedent improbability in the theory propounded by German Sanskṛitists that an early passage of phonographic symbols took place from a Phœnician centre eastward towards Mesopotamia and India, at about the same period as their passage westward towards Europe, namely, about 800 B.C.
It is not asserted that the exact channel by which they were transmitted has been satisfactorily demonstrated. Some think — and, as it seems to me, with much plausibility — that they may have been introduced through contact with the Greeks50. Perhaps a more likely conjecture is that Hindū traders, passing up the Persian Gulf, had commercial dealings with Aramaean traders in Mesopotamia, and, becoming acquainted with their graphic methods, imported the knowledge and use of some of their phonetic signs into India.
This view was first propounded in the writings of the learned Professor A. Weber of Berlin, and has recently been ably argued in a work on “Indische Palæographie,” by the late Professor Bühler of Vienna (published in 1896). If Indian Pandits will consult that most interesting standard work, they will there find a table exhibiting the most ancient of known Phœnician letters side by side with the kindred symbols used in the Moabite inscriptions of King Mesha — which, as before intimated, is known to be as old as about 850 B.C. — while in parallel columns, and in a series of other excellent tables, are given the corresponding phonographic symbols from the numerous inscriptions of King Aśoka scattered everywhere throughout Central and Northern India51.
These inscription-alphabets are of two principal kinds: —
The first kind is now called Kharoshṭhī (or “Ass's lip” form of writing, lipi being understood)52. This belongs to the North-west corner of the Panjāb and Eastern Afghānistān. It was used by King Aśoka for a few of his rock and stone inscriptions, and is a kind of writing the prototype of which was probably introduced into Persia about 500 B.C., and brought by Persian rulers into Northern India in the fourth century B.C. At all events, it is well known that the Persian monarchs of the Akhæmenian period employed Aramæan scribes, and that the Kharoshṭhī writing, even if originally Indian (according to Sir A. Cunningham and others), has assumed under their hands a manifestly Aramaic character, flowing like all Semitic writing from right to left. Possibly, however, as it seems to me, Grecian influences (which penetrated into India before the time of Alexander) may have partially operated in assimilating this early North-western Indian script to a Phœnician type. It may be excluded from our present inquiry, because it never became generally current in India, and never developed into a form suitable for printing.
The second kind of ancient Indian script is called Brāhma (or Brāhmī lipi). This is without doubt the oldest of the two principal forms53. Its claim to greater antiquity is proved by its name Brāhma — given to it by the Brāhmans, because, as they assert, it was invented by their god Brahmā54 — an assertion which may be taken as indicating that, whatever its origin, it was moulded into its present form by the Brāhmans.
And undeniably it is this Brāhma writing (Brāhmī lipi) which has the best right to be called the true Indian Brahmanical script. It must have been the first kind of writing used when Sanskṛit literature began to be written down (perhaps six centuries B.C.), and it is the script of the Aśoka inscriptions of Central and Northern India — and even of North-western India, where it is found concurrently with the Kharoshṭhī. It was employed to express the Prākṛit dialect55 of the Buddhist kings, and flowed, like its later development called Nāgarī, from left to right. Its first appearance on actually existing inscriptions — so far as at present discovered — cannot be placed earlier than the date of these kings in the third century B.C.
But it is important to note that the existence of the Brāhmī lipi in India must be put back to a period sufficiently early to allow for its having once flowed from right to left like the Kharoshṭhī, probably as early as the sixth century B.C. This is made clear by the direction of the letters on an ancient coin discovered by Sir A. Cunningham at Eraṇ56 — a place in the central provinces remarkable for its monumental remains. One can scarcely accept seriously the suggestion that the position of the short ि i in the present Nāgarī is a survival of the original direction of the writing57.
If then any unprejudiced Hindū scholar will examine attentively the tables in Professor Bühler's book, he will, I think, be constrained to admit that the Indian Brahma letters have certain features which connect them with the ancient Phœnician script, and therefore with the Greek and Roman.
It should not, however, be forgotten that an interval of nearly seven centuries separates the Phœnician from the Brāhma inscription-letters, and that to make the affinity between the two alphabets clearer the side-lights afforded by collateral and intermediate Semitic scripts ought to be taken into account58. Nor should it be forgotten that when the Hindūs, like the Greeks, changed the direction of their writing, some of the symbols were turned round or their forms inverted, or closed up or opened out in various ways.
The further development of the Brāhma symbols into the modern Deva-Nāgarī and its co-ordinate scripts59 is easily traceable. It must, however, be borne in mind that the later Pandits tried to improve the ancient graphic signs by setting them up as upright as possible and by drawing a horizontal stroke to serve as a line from which the letters might hang down, and so secure a system of straight writing — often conspicuously absent in Hindūstānī and Persian caligraphy60.
I here append a table consisting of seven columns, in which I have so arranged the letters as to illustrate the view that the Phœnician alphabet spread about 800 B.C. first westward towards Greece and Italy, and secondly eastward towards India.
The column marked 1 gives ten Phœnician letters. That marked 2, to the left of I, gives the ten corresponding Greek letters; that marked 3 the corresponding Roman; and that marked 4 the corresponding English letters. Then the column marked 2, to the right of I, gives the ten corresponding Brāhma letters; that marked 3 shows the gradual developments of the Brāhma symbols as exhibited on various inscriptions; and that marked 4 gives the corresponding letters in modern Nāgarī61.
Let any one study this Table and he must, I think, admit that it indicates an original connexion or family likeness between the Phœnician and earliest Indian or Brāhma letters, whilst it also illustrates the fact that the plastic hand of the Brāhmans has greatly modified and expanded the original germs, without, however, obliterating the evident indications of their connexion with the Phœnician.
* This is for the Greek theta, which is represented in this Dictionary, according to present usage, by th, although ϑ would be a more scientific symbol.
§ According to Professor Bühler, the Brāhma 
. became Nāgarī ध dh, from which द d was evolved.
And indeed the modest equipment of twenty-two letters which satisfied the Phœnicians, Greeks, and Romans, to whom the invention of writing was a mere human contrivance for the attainment of purely human ends, could not possibly have satisfied the devout Hindū, who regarded his language as of divine origin, and therefore not to be expressed by anything short of a perfect system of equally divine symbols. Even the popular Prākṛit of King Aśoka's edicts seems to have required nearly forty symbols62, and the amount needed for the full Brāhmī lipi, as used for the Sanskṛit of that period, could not have been less than fifty (if the symbols for ai, au, ṛi, ṛī, lṛi, lṛī, and ḷa be included).
Then, if we turn to the Brāhma alphabet in its final development, called Nāgarī, we see at a glance that it is based on the scientific phonetic principle of “one sound one symbol” — that is, every consonantal sound is represented by one invariable symbol, and every shade of vowel-sound — short, long, or prolated — has one unvarying sign (not as in English where the sound of e in be may be represented in sixteen different ways). Hence, for the expression of the perfectly constructed Sanskṛit language there are sixteen vowel-signs (including am and aḥ and excluding the prolated vowel forms), and thirty-five simple consonants, as exhibited on p. xxxvi of this volume.
Of course a system of writing so highly elaborated was only perfected by degrees63, and no doubt it is admirably adapted to the purposes it is intended to serve. Yet it is remarkable that even in its latest development, as employed in the present Dictionary, it has characteristics indicative of its probable original connexion with Semitic methods of writing, which from their exclusively consonantal character are admittedly imperfect.
For the Pandits, unlike the Greeks and Romans, cannot in my opinion be said to have adopted to the full the true alphabetic theory which assigns a separate independent position to all vowel-signs. And my reason for so thinking is that they make the commonest of all their vowels — namely short a64 — inherent in every isolated consonant, and give a subordinate position above or below consonants to some of their vowel-signs. And this partially syllabic character of their consonantal symbols has compelled them to construct an immense series of intricate conjunct consonants, some of them very complicated, the necessity for which may be exemplified by supposing that the letters of the English word “strength” were Nāgarī letters, and written सतरेनगथ. This would have to be pronounced satarenagatha, unless a conjunction of consonantal signs were employed, to express str and ngth, and unless the mark called Virāma, “stop,” were added to the last consonant. So that with only thirty-three simple consonants and an almost indefinite number of complex conjunct consonants the number of distinct types necessary to equip a perfect Sanskṛit fount for printing purposes amounts to more than 500.
Surely, then, no one will maintain that, in these days of every kind of appliance for increased facilities of inter-communication, any language is justified in shutting itself up behind such a complex array of graphic signs, however admirable when once acquired. At all events such a system ought not to have the monopoly for the expression of a language belonging to the same family as our own and in a country forming an integral part of the British Empire. The Sanskṛit language, indeed, is a master-key to a knowledge of all the Hindū vernaculars, and should moreover be studied as a kind of linguistic bond of sympathy and fellow-feeling between the inhabitants of the United Kingdom and their Indian fellow-subjects. But to this end every facility ought to be afforded for its acquirement.
And if, as we have tried to show, the Brāhmī lipi, the Nāgarī, and the Greek and Romanic alphabets are all four related to each other — at least, in so far as they are either derived from or connected with the same rudimentary stock — it surely cannot be opposed to the fitness of things, that both the Nāgarī and Romanic alphabets should be equally applied to the expression of Sanskṛit, and both of them made to co-operate in facilitating its acquisition.
Nor let it be forgotten that in the present day the use of the English language is spreading everywhere throughout India, and that it already co-exists with Sanskṛit as a kind of lingua franca or medium of communication among educated persons, just as Latin once co-existed with Greek. So much so indeed, that, contemporaneously with the diffusion of the English language, the Roman graphic system, adopted by all the English-speaking inhabitants of the British Empire, has already forced itself on the acceptance of the Pandits, whether they like it or not, as one vehicle for the expression of their languages; just as centuries ago the Arabic and Persian written characters were forced upon them by their Muhammadan conquerors for the expression of Hindi.
It is on this account that I feel justified in designating the European method of transliteration employed in this Dictionary by the term “Indo-Romanic alphabet.”
And be it understood that such an acceptance of the Romanic alphabet involves no unscientific adaptation of it to the expression of Sanskṛit like our chaotic adaptation of it to the expression of English; or like the inaccurate use of it by native writers themselves in transliterating their own Indian words65. Quite the reverse. The Roman alphabet adapts itself so readily to expansion by the employment of diacritical points and marks, that it may be regarded as a thoroughly scientific instrument for the accurate expression of every Indian sound, and probably of nearly every sound, in every language of the world. And it may, I think, be confidently predicted that before the twentieth century has closed, man's vision, overtasked by a constantly increasing output of literary matter, will peremptorily demand that the reading of the world's best books be facilitated by the adoption of that graphic system which is most universally applicable and most easily apprehensible. Whether, however, the Roman symbols will be ultimately chosen in preference to other competing systems as the best basis for the construction of a world's future universal alphabet no one can, of course, foretell with the same confidence.
One thing, I contend, is certain. Any ordinary scholar who consults the present work will be ready to admit that it derives much of its typographical clearness from certain apparently trifling, but really important, contrivances, possible in Romanic type, impossible in Nāgarī. One of these, of course, is the power of leaving spaces between the words of the Sanskṛit examples. Surely such a sentence as sādhu-milrāny akuśalād vārayanti is clearer than sādhumitrāṇyakuśalādvārayanti. Again, who will deny the gain in clearness resulting from the ability to make a distinction between such words as “smith” and “Smith,” “brown” and “Brown,” “bath” and “Bath?” not to speak of the power of using italics and other forms of European type. And, without doubt, the use of the hyphen for separating long compounds in a language where compounds prevail more than simple words66, will be appreciated by all. I can only say that, without that most useful little mark, the present volume must have lost much in clearness, and still more in compactness; for, besides the obvious advantage of being able to indicate the difference between such compounds as su-tapa and suta-pa which would have been impossible in Nāgarī type, it is manifest that even the simplest compounds, like sad-asad-viveka, sv-alpa-keśin, would have required, without its use, an extra line to explain their analysis67.
Fairness, however, demands that a few of the obvious defects of the Indo-Romanic system of transliteration adopted in this volume should be acknowledged. In certain cases it confessedly offends against scientific exactness; nor does it always consistently observe the rule that every simple vowel-sound should be represented by a single symbol. For instance, the Sanskṛit vowels ऋ and ॠ are not represented in this Dictionary by the symbols ṛ and ṝ, according to the practice of some German scholars — a practice adopted by the Geneva Transliteration Committee — but by ṛi and ṛī. And my reason is that, inasmuch as in English Grammar r is not regarded as a semi-vowel, ṛ and ṝ are unsuitable representatives of vowel-sounds. Moreover, they are open to this objection, that when the dot under the r is accidentally dropped or broken off, as often happens in printing, especially in India, the result is worse than if the r were followed by i. For example, Krshṇa is surely worse than Krishṇa.
So again in the case of aspirated consonants, the aspiration ought not to be represented by a second letter attached to them. Indeed, in the case of ch employed by Sir W. Jones for the palatal च, and chh for छ, the inconvenience has been so great that in the present edition I have adopted (in common with many other Sanskṛitists) the simple c for च, the pronunciation being the same as c in the Italian dolce or as ch in “church,” the latter of which would, if a Sanskṛit word, be written “curc.” Similarly ch has been adopted for छ68.
As to the transliteration of the palatal sibilant श, I have preferred ṡ to the ś employed in the first edition, and I much prefer it to the German and French method of using ç. Experience proves that the cedilla is often either broken off in printing or carelessly dropped, and as a consequence important words such as Aśoka are now often wrongly printed and pronounced Acoka.
So also I should have preferred the symbol ṣ for the cerebral sibilant, but have felt it desirable to retain sh in the present edition. There is the same objection to ṣ as to the ṛ mentioned above. This will be clear if we write the important word ṛishi in the way German scholars write it, namely ṛṣi, and then omit the dots thus, rsi.
In regard to the nasals I have in the present edition adopted ṅ for ङ and ñ for ञ. In these changes I am glad to find myself in accord with the Geneva Transliteration Committee.
As to the method of using italic k, kh for च, छ and italic g, gh for ज, झ — adopted in the “Sacred Books of the East” — the philological advantage thought to be gained by thus exhibiting the phonetic truth of the interchange of gutturals and palatals, appears to me to be completely outweighed by the disadvantage of representing by similar symbols sounds differing so greatly in actual pronunciation. For instance, to represent such common words as “chinna” by “khinna” and “jaina” by “gaina” seems to me as objectionable as to write “Khina” for “China” and “Gapan” for “Japan.” The plan of using Italics is no safeguard, seeing that in printing popular books and papers the practice of mixing up Roman and Italic letters in the same word is never adhered to, so that it is now common to find the important Indian sect of Jains printed and pronounced “Gains69.”
Having felt obliged by the form in which this Dictionary is printed to dwell at full length on a matter of the utmost importance both in its bearing on the more general cultivation of Sanskṛit and on the diffusion of knowledge in our Eastern Empire, I must now repeat my sense of the great assistance the cause of the transliteration of Indian languages into Romanized letters formerly received at the hands of the late Sir Charles Trevelyan. He was the first (in his able minute, dated Calcutta, January, 183470) to clear away the confusion of ideas with which the subject was perplexed. He also was the first to awaken an interest in the question throughout England about forty-two years ago. His arguments induced me to take part in the movement, and our letters on the subject were published by the “Times,” and supported by its advocacy. Since then, many Oriental books printed on a plan substantially agreeing with Sir W. Jones' Indo-Romanic system, have been published71. Moreover, on more than one occasion I directed the attention of the Royal Asiatic Society72, and of the Church Missionary Society73, and Bible Society, to this important subject, and at the Congress of Orientalists held at Berlin in September, 1881, I read a paper, and submitted a proposal for concerted international action with a view to the fixing of a common scheme of transliteration. The discussion that followed led to the appointment of the first Commission for settling a common international system of transcription, and it may, I think, be fairly assumed that the agitation thus set in motion, and carried on for so many years, was one of the principal factors in bringing about the proposed international scheme issued by the Transliteration Committee of the Geneva Oriental Congress in September, 1894.
Acknowledgment of Assistance Received
In the Preface to the first edition I made special mention of the name of an eminent scholar who was a member of the Oxford University Press Delegacy when the publication of that edition was undertaken — Dr. Robert Scott, sometime Master of Balliol, afterwards Dean of Rochester, and co-author with Dr. Liddell of the well-known Greek Lexicon. He had been one of my kindest friends, and wisest counsellors, ever since the day I went to him for advice during my first undergraduate days at Balliol, on my receiving an appointment in the Indian Civil Service, and I need scarcely repeat my sense of what this Dictionary, in its inception, owed to his support and encouragement.
Nor need I repeat the expression of my sense of obligation to my predecessor in the Boden Chair, Professor H. H. Wilson, who first led me to the study of Sanskṛit about sixty years ago (in 1839), and furnished me with my first materials for an entirely new system of Sanskṛit lexicography (see p. xi). All the words and meanings marked W. in the following pages in the present work rest on his authority.
Nevertheless, sincerity obliges me to confess that, during my long literary career, my mind has had to pass through a kind of painful discipline involving a gradual weakening of faith in the trustworthiness of my fellow men, not excepting that of my first venerated teacher. I began my studies, indeed, with much confidence in the thought that one man existed on whom I could lean as an almost infallible guide; but as I grew a little wiser, and my sensitiveness to error sharpened, I discovered to my surprise that I was compelled to reject much of his teaching as doubtful. Nay, I am constrained to confess that as I advanced further on the path of knowledge, my trustfulness in others, besides my old master, experienced by degrees a series of disagreeable and unexpected shocks; till now, that I have arrived at nearly the end of my journey, I find myself left with my faith in the accuracy of human beings generally — and certainly not excepting myself — somewhat distressingly disturbed. Such painful feelings result, I fear, in my own case from a gradual and inevitable growth of the critical faculty during a long lifetime, and are quite consistent with a sense of gratitude for the effective aid received from my collaborators, without which, indeed, I could not have brought this work to a conclusion.
In my original Preface I expressed my thanks to each and all of the scholars who aided me in the compilation of the first edition, and whose names in the chronological order of their services were as follow: —
The late Rev. J. Wenger, of the Baptist Mission, Calcutta; Dr. Franz Kielhorn, afterwards Superintendent of Sanskṛit Studies in Deccan College, Poona, and now Professor of Sanskṛit in the University of Göttingen; Dr. Hermann Brunnhofer; Mr. A. E. Gough, M.A., of Lincoln College, Oxford, sometime Professor in the Government Colleges of Benares, Allāhābād, and Calcutta; and lastly, Mr. E. L. Hogarth, M.A., of Brasenose College, sometime Head Master of the Government Provincial School at Calicut.
It is now my duty to express my grateful obligations to the able and painstaking Assistants who have co-operated with me in producing the present greatly enlarged and improved work.
No one but those who have taken part in similar labours can at all realize the amount of tedious toil — I might almost say dreary drudgery — involved in the daily routine of small lexicographical details, such as verifying references and meanings, making indices and lists of words, sorting and sifting an ever-increasing store of materials, revising old work, arranging and re-arranging new, writing and re-writing and interlineating “copy,” correcting and re-correcting proofs — printed, be it remembered, in five kinds of intricate type, bristling with countless accents and diacritical points, and putting the eyesight, patience, and temper of author, collaborators, compositors, and press-readers to severe trial. I mention these matters not to magnify my own labours, but to show that I could not have prosecuted them without the able co-operation of others.
The names of my new Assistants in chronological order are as follow: —
First, Dr. Ernst Leumann (a native of Switzerland), who worked with me in Oxford from October 3, 1882, until April 15, 1884, when he accepted a teachership in the Kantonschule of Frauenfeld in Switzerland. I have already acknowledged my obligations to him.
He was succeeded by the late Dr. Schönberg (a pupil of the late Professor Bühler), who came to me in a condition of great physical weakness, and whose assistance only extended from May 20, 1884, to July 19, 1885, when he left me to die. He was a good scholar, and a good worker, but impatient of supervision, and, despite my vigilance, I found it impossible to guard against a few errors of omission and commission due to the rapid impairment of his powers.
Then followed an interval during which my sources of aid were too fitful to be recorded.
In September, 1886, Dr. Leumann, who had meanwhile been appointed Professor of Sanskṛit in the University of Strassburg, renewed his co-operation, but only in an intermittent manner, and while still resident in Germany. Unhappily the pressure of other duties obliged him in September, 1890, to withdraw from all work outside that of his Professorship. He laboured with me in a scholarly way as far as p. 474; but his collaboration did not extend beyond 355 pages, because he took no part in pp. 137-256, which represent the period of Dr. Schönberg's collaboration.
It was not till December, 1890, that Dr. Carl Cappeller, Professor of Sanskṛit in the University of Jena, began his painstaking co-operation, which, starting from the word Dāda (p. 474), he has prosecuted perseveringly to the completion of the Dictionary. And it should be put on record that, although his collaboration had to be carried on contemporaneously with the discharge of his duties at Jena — involving the necessity for a constant interchange of communications by post — yet it resulted in the production of 834 finished pages between March, 1891, and July, 1898. It should also be recorded that, from the beginning of the letter प p, he had a careful assistant in Dr. Blau of Berlin, who also occasionally read the proof-sheets and contributed a certain number of words for the Addenda.
Furthermore, I must express my gratitude to Herr Geheimrath Franz Kielhorn, C. I. E., Ph.D., Professor of Sanskṛit in the University of Göttingen, who was my assistant soon after the inception of the first edition, for his free and generous supervision of the grammatical portions of the present edition from about the year 1886; and his readiness to place-at my disposal the experience which he gained during his labours for many years as Superintendent of Sanskṛit Studies at the Government College, Poona.
I have finally to record my grateful appreciation of the value of the principal work: used or consulted by my collaborators and myself in compiling this Dictionary. Some of these, and a few important grammatical works — such as the Mahā-bhāshya (in the excellent edition of Professor Kielhorn), the Siddhânta-kaumudī &c. — besides many other texts, such as that of Manu, the Bṛihat-saṁhitā &c., did not exist in good critical editions when the great Thesaurus of the two German Lexicographers was being compiled.
Professor Ernst Leumann informs me that during the period of his collaboration he was much aided by Grassmann's Rig-veda, Whitney's Index Verborum to the published text of the Atharva-veda; Stenzler's Indices to the Gṛihya-sūtras of Āśvalāyana, of Pāraskara, Śāṅkhāyana, Gobhila, and the Dharma-śāstra of Gautama; the vocabularies to Aufrecht's edition of the Aitareya Brāhmaṇa; Bühler's Āpastamba Dharma-sūtra; Garbe's Vaitāna-sūtra; Hillebrandt's Śāṅkhāyana Śrauta-sūtra &c. He states that in his portion of the work his aim was rather to verify and revise the words and meanings given in the Petersburg Dictionaries than to add new and unverifiable matter. In regard to quotations he refers the reader to the Journal of the German Oriental Society, vol. xlii, pp. 161-198.
Professor C. Cappeller states that in addition to the books enumerated above he wishes to name in the first place Böhtlingk's Upanishads, his Pāṇini (2nd ed.) and Kāvyâdarśa as well as the valuable critical remarks of that honoured Nestor of Sanskṛitists on numerous texts, published in various journals; further the Jaiminīya Upanishad Brāhmaṇa edited by H. Oertel, and various Sūtra works with their indices by F. Knauer, M. Winternitz, J. Kirste, and W. Caland. For some additions contributed from the Drāhyāyaṇa Śrauta-sūtra he is indebted to Dr. J. N. Reuter of Helsingfors. He also made use of the Vaijayantī of Yādava-prakāśa (edited by G. Oppert, London, 1893); the Uṇādigaṇa-sūtra of Hemacandra (edited by J. Kirste, Vienna, 1895); the Dictionaries of Apte (Poona, 1890), of A. A. Macdonell (London, 1893), of C. Cappeller (Strassburg, 1891); Whitney's Roots, Verb-forms, and Primary Derivatives of the Sanskṛit Language (Leipzig, 1885); Lanman's Noun-inflection in the Veda (New Haven, 1880); Jacob Wackernagel's Altindische Grammitik (Göttingen, 1896); Delbrück's Altindische Syntax (Halle, 1888); Regnaud's Rhétorique Sanskṛite (Paris, 1884); Lévi's Théatre Indien (Paris, 1890); Macdonell's Vedic Mythology (Strassburg, 1897), &c.
For Vedic interpretation Roth and Grassmann have been the chief authorities, but it will be seen that neither Sāyaṇa nor such modern interpreters as Pischel and Geldner in Vedische Studien (Stuttgart, 1889-1897), and Bloomfield for the Atharva-veda (in S. B. E., vol. xlii) have been neglected.
The Buddhistic portion of the Dictionary has chiefly been enriched by the following: — Aśvaghosha's Buddha-carita (edited and translated by Professor E. B. Coweil of Cambridge); Divyâvadāna (edited by Cowell and Neil, Cambridge, 1886); Jātaka-mālā (edited by H. Kern, Boston, 1891); the two Sukhāvatī-vyūhas (S. B. E., vol. xlix) and the Dharma-saṁgraha (Anecdota Oxoniensia, 1885). It is evident, that until new and complete Pāli and Prākṛit Dictionaries are published, the idiomatic Sanskṛit used by Buddhists and Jains and the authors of certain inscriptions cannot be dealt with satisfactorily.
Of course many portions of the Indische Studien (edited by Professor A. Weber of Berlin) have been consulted, and valuable aid has been received from some of the translations contained in the “Sacred Books of the East,” as well as from many other works, the names of which will be found in the List of Works and Authors at p. xxxiii.
As to the books used by myself, many of them, of course, are identical with those named above. Others are named in the first edition, and need not be referred to again here. I ought, however, to repeat that some of the words marked MW. in the present edition rest on the authority of the Śabda-kalpa-druma of Rādhā-kānta-deva (published in eight volumes at Calcutta in the Bengali character). I am also, of course, responsible for some words and meanings taken from my own books, such as “Brāhmanism and Hindūism,” “Buddhism,” “Indian Wisdom” (see note 1 to p. vi of Preface), my Sanskṛit Grammar and Nalôpâkhyānam (with vocabulary, published by the Delegates of the Oxford University Press), text of the Śakuntalā (with index and notes, published by the same), as well as from the notes appended to my English translation of the Śakuntalā (published by Messrs. Harmsworth among Sir John Lubbock's hundred best books of the world), &c.
MONIER MONIER-WILLIAMS.
INDIAN INSTITUTE, OXFORD
[The order is that of the English Alphabet, The letters outside the parentheses represent the abbreviated forms used in the references."
[In the progress of a work extending over many years it has been found almost impossible to preserve absolute uniformity in the use of abbreviations and symbols, but it is hoped that most of the inconsistencies are noticed in the following table
The dictionary order of the Nagarī letters
with their indo-romanic equivalents and their pronunciation exemplified bv English words
Consonants
* Sometimes printed in the form 
, see p. 174, col. 3.
The conjunct consonants are too numerous to be exhibited above, but the most common will be found at the end of “A Practical Sanskṛit Grammar by Monier-Williams,” published by the Delegates of the Clarendon Press, fourth edition.
For the correct pronunciation of the aspirated consonants, kh, eh, ţht th, ph, &c., see p. xxix, note 4 of the foregoing Introduction.
Observe that n represents the true Anusvāra in the body of a word before the sibilants and h, as in anśa, ansa, anhati: ṁ as the symbol of any nasal will often be found at the end of a word, as in dānaṁ ca; but may also represent Anusvāra, when final m is followed by initial semivowels, sibilants and h, and in words formed with preposition sam, like taṁ-veśa, saṁ-śaya, tsaṁ-hata: the word Sanskṛit is now too Anglicized to be written Saṁskṛit. Visarga, as a substitute for final s, is a distinctly audible aspirate, so that the ḥ at the end of such a word as đevaḥ must be clearly heard.
The dictionary order of the Indo-Romanic letters without their nagarl equivalents.
1 Some of these books are referred to in the present Dictionary; for example, that on “Hindūism” (published by the S.P.C.K., 13th edition); that on “Brāhmanism” &c. (also called “Religious Thought and Life in India;” published by Mr. J. Murray, Albemarle Street, 4th ed., referred to as RTL.); that on “Indian Wisdom” (published by Messrs. Luzac of Great Russell Street, 4th ed., referred to as IW.); that on “Buddhism” (also published by Mr. Murray, 2nd ed., referred to as MWB.).
2 One of these was the founding of an Indian Institute for the promotion of Indian studies in my own University of Oxford. Another was to induce the Government of India to found six Government scholarships for enabling deserving Indians to finish their education at our University.
3 The three Viceroys were Lord Northbrook, the late Lord Lytton, and Lord Ripon. I owe a deep debt of gratitude to Lord Ripon for receiving me as his guest at Government House, Calcutta, in 1883-4; and to Sir Richard Temple for receiving me at Government House, Belvedere, during the Prince of Wales' visit in 1875-6; and to Sir James Fergusson for receiving me at Government House, Bombay, in 1884.
4 I cannot allow myself to think that the Dictionary has suffered much from this cause, except perhaps during the collaboration of the late Dr. Schönberg, the rapid impairment of whose powers did not at first strike me so as to make me aware of the necessity for increased vigilance on my part (see p. xxxi).
5 I believe it is held that for an Alpine climber to establish a reputation for mountaineering he must ascend some peak, however comparatively insignificant, that has never been ascended before. But the application of such a principle as a sole proof of scholarship in the present day, can no more hold good in Sanskṛit than in Greek and Latin. At all events let any one who claims a reputation for superior scholarship on that sole ground associate with Indian Pandits in their own country and he will find out that far severer proofs of his knowledge and acquirements will be required of him there.
6 Lieutenant-Colonel Boden, of the-Bombay Native Infantry, returned to England in 1807 and died at Lisbon, Nov. 21, 1811. His daughter died Aug. 24,1827, whereupon his bequest passed to the University of Oxford, but the first election to the Chair, for some reason unknown to me, did not take place till 1832.
7 In his address proposing himself for election to the Boden Electors, Professor H. H.Wilson laid stress on what he had done for “the rendering of Scripture Terms into the Sanskṛit language.” It was doubtless on this account that after he was elected he urged me to compile an English-Sanskṛit Dictionary — a work never before attempted, I laboured at this for about seven years, and although the result (published in a thick volume by the Directors of the East India Company in 1851) cannot, I fear, be said to meet the needs of the present day, yet it should be borne in mind that it was pioneering work. Nor can it be said to have been useless, seeing that seven years after its publication the following testimony to its utility was voluntarily tendered by the Rev. J. Wenger, translator of the Bible into Sanskṛit and Editor of Dr. Yates' Sanskṛit Dictionary: — “I received a copy of Professor Monier Williams' English and Sanskṛit Dictionary at a time when I was about to commence a translation into Sanskṛit of portions of the Old Testament. I have used it daily for the last seven years, and the more I have consulted it the more excellent I have found it. I feel bound to say that he appears to have succeeded, not only beyond my previous ideas of what was likely, but also of what was feasible, to be accomplished at the present time. The Pundits whom I employ have likewise expressed their unqualified admiration of the labour and erudition which his volume displays. The Rev. J. Parsons of Benares, who has been engaged for some years past in preparing a new Hindee version of the New Testament, has likewise derived material assistance from Professor M. W.'s work. Indian missionaries generally owe him a large debt of gratitude.”
Let me add that I hope the present Sanskṛit-English Dictionary will furnish some young scholar with materials for the compilation of a far more satisfactory English-Sanskṛit Dictionary than that which I began to compile more than half a century ago.
8 This will be found in the library presented by me to the Indian Institute, Oxford.
9 The main object was really a missionary one, as I have shown in the Preface to this volume (p. ix), and in my Life of H. H.Wilson appended to my Reminiscences of Old Haileybury College (published by A. Constable & Co.).
10 His first Dictionary was published in 1819, and his second in 1832, while he was a candidate for the Boden Professorship.
11 Sanskṛit is now too Anglicized a word to admit of its being written as it ought to be written according to the system of transliteration adopted in the present Dictionary — Saṁskṛit.
12 The name Semitic or Shemitic is applied to Assyrian, Hebrew, Aramaic (or Aramæan), Arabic, and Himyaritic, because in the tenth chapter of Genesis, Shem is represented as father of the principal nations speaking these languages — e.g. Assur (Assyria), Aram (Syria), and of Arphaxad, grandfather of Eber, from whom came the Hebrews — or Trans-Euphratian race, the name Hebrew coming from 
, and really meaning “one who lives beyond (a river)” — and Joktan, the father of many of the tribes inhabiting South Arabia. It is usual, too, to reckon among Semitic races the people of Abyssinia, whose sacred and literary language is the Ethiopic or Ge'ez, while their spoken dialects are Tigré for the north and north-east, and Amharic for the centre and south, all presenting affinities with the ancient Himyaritic Arabic of South Arabia (Yaman). Hence, speaking generally, we may classify Semitic languages under the two heads of: — 1. “North Semitic,” comprising Assyrian, Hebrew, and Aramaic; 2. “South Semitic,” comprising Arabic, Himyaritic, and Ethiopic.
13 Though the younger sisters sometimes preserve older forms.
14 According to some German Theorists the cradle-land of the Āryans was in the steppes of Southern Russia. Others have fantastically placed it in Northern Europe. Most scholars hold to the old idea of “somewhere in Central Asia,” and probably in the region of Bactria (Balkh) and Sogdiana, although there might have been a second centre of migration. I myself firmly believe that Balkh was once a chief ancient home of Āryan civilization. Its ruins are said to extend for twenty miles.
15 See note 3, p. xxv [this note is reprinted below], on Pāli and on the Prākṛit of the inscriptions.
[Aśoka, who called himself Priya-darśin, and was the grandson of Candra-gupta, did for Buddhism what Constantine did for Christianity, by adopting it as his own creed. Buddhism then became the religion of the whole kingdom of Magadha, and therefore of a great portion of India; and Aśoka's edicts, inscribed on rocks and pillars (about the middle of the third century B.C.), furnish the first authentic records of Indian history. Yet the language of these inscriptions cannot be said to be exactly identical with so-called Māgadhī Prākṛit, nor with the Pāli of the Buddhist sacred scriptures, although those forms of Prākṛit may be loosely called either Māgadhī or Pāli. Nor was the name Pāli originally applied to the language of the Buddhist Canon, but rather to the line or series of passages constituting a text (cf. the use tantra). According to Professor Oldenberg the Vinaya portion of the texts existed in its present form as early as 400 B.C. The later Buddhist texts were written down not long after, and commentaries have since been compiled in Pāli and the languages of Ceylon, Siam, and Burma; the Pāli of Ceylon being affected by intercourse with Kalinga (Orissa).]
16 As to the Avesta, commonly called Zend (more correctly Zand), this is that ancient language of Eastern Īrān in which are written the sacred books of the Zoroastrians, commonly called Zend-Avesta — books which constitute the bible and prayer-book of the Pārsīs — those fugitives from Persia who are scattered everywhere throughout India, and are now among the most energetic and loyal of our Indian fellow-subjects.
17 As distinguished from unchangeably“monosyllabic” like the Chinese, and “agglutinative” like the Drāviḍian of Southern India, and like the Turkish and other members of an immense class of languages, in which there are no so-called “inflections,” but merely affixes or suffixes “glued” as it were to the root or body of a word, and easily separable from it, and not blending intimately with it, and so, as it were, inflecting it.
18 Of course it is well understood that there are in Sanskṛit a certain number of dissyllabic roots, but I am here merely contrasting Semitic and Āryan roots generally.
19 The vikaraṇa of a root may be called an “affix,” and the verbal termination &c. a “suffix.”
20 For other illustrations of this see I. kṛi, p. 300; 1. śru, p. 1100; 1. sthā, p. 1262 of this volume.
21 The number of distinct Dhātus or radical forms given in some collections is 1,750, but as many forms having the same sound have different meanings, and are conjugated differently, they are held to be distinct roots and the number is thereby swelled to 2,490. It should be noted, too, that a great many of these Dhātus are modifications or developments of simpler elements, and this Dictionary does not always decide as to which of two, three or more roots is the simplest, although when roots are allied their connexion is indicated. Probably the real number of elementary radicals in Sanskṛit might be reduced to a comparatively small catalogue — even, as some think, to a list of not more than about 120 primitive roots. Many Sanskṛit roots have alternative Prākṛit forms or vice versâ, and both forms are allowed to co-exist, as bhan and bhaṇ, dhan and dhaṇ, nṛit and naṭ; others whose initials are aspirated consonants have passed into other aspirated consonants or have retained only the aspirate, as in bhṛi, dhṛi, dhvṛi, hvṛi, hṛi &c. Again, such a root as svad is probably nothing but a compound of su and root ad, and such roots as stubh, stumbh, stambh are plainly mere modifications of each other.
22 I use the expression “Indo-Romanic” and “Indo-Italic” to denote the expanded Roman and Italic alphabets adapted by the use of diacritical points and marks to the expression of Sanskṛit and other Indian languages. The thick Indo-Romanic type employed in this volume is a product of the Oxford Clarendon Press, and therefore named Clarendon type.
23 In this first case the hyphen used in the transliterated form is no doubt sufficient to distinguish the two forms from each other. Hence, to economize space, the figures have occasionally towards the end of the work been omitted (see samāná, Sa-māna, p. 1160).
24 I must, however, here repeat the acknowledgment of my original indebtedness to “Westergaard's Radices;” nor must I omit to mention Whitney's valuable Index of Roots, Verb-forms and Primary Derivatives.
25 Unfortunately in noting down words for insertion I omitted to quote the sources whence they were taken, as I did not at the time contemplate improving my new edition by the addition of references.
26 Rig-Veda has now become an Anglicized word, and the dot under the R has been omitted in the Dictionary for simplicity.
27 I am sorry to have to confess that imbued as I once was with false notions as to the deadness of Sanskṛit, I have sometimes omitted to give the meanings of important modern words like prâṇa-pratiskṭhā in the body of the Dictionary.
28 The absence of accent was only permitted in calling out to a person in the distance, Pan. i, 2, 33.
29 The importance of correct accentuation and intonation in a language, the very sound of which is held by the Hindūs to be divine, and the bearing of Sanskṛit accentuation on that of Greek, had become so impressed on me, that when I was sent as a Delegate to the Berlin International Congress of Orientalists by the Government of India in 1881, I requested Pandit Śyāmajī Kṛishṇa-varmā (who was also a Government Delegate) to illustrate my paper on Vedic hymns by repeating them with the right accentuation. The Pandit's illustrations were not only much appreciated, but received with grateful acknowledgments at the time by the eminent Chairman, Prof. A. Weber, and other Sanskṛit scholars present, but were misconstrued by one of my auditors — the well-known and most energetic Hon. Secretary of the Royal Asiatic Society. That gentleman made the Pandit's illustrative additions the subject of an extraordinary criticism in a paper on “Oriental Congresses,” written by him and published in the Calcutta Review, No. CLXI (1885), and quite recently reprinted. A letter lately received by me from Professor A. Weber, and printed last year in the Asiatic Quarterly Review, expresses the astonishment which we both felt at the statements in that paper.
30 Some compound words which are formed by Taddhita affixes supposed to be added to the whole word ought not strictly to have a hyphen.
31 A paper written by Pandit Śyāmajī Kṛishṇa-varmā on “Sanskṛit as a living language in India,” was read by him at the Berlin Oriental Congress of 1881, and excited much interest. He argues very forcibly that “Sanskṛit as settled in the Ashṭâdhyāyī of Pāṇini was a spoken vernacular at the time when that great grammarian flourished.” In the same paper he maintains that Sanskṛit was the source of the Prākrits, and quotes Vararuci's Prākṛita-pra-kāśa xii, a (Prakṛitiḥ saṁskṛitam, “Sanskṛit is the source”). Of course the provincialized Prākṛits — though not, as I believe, derived directly from the learned language, but developed independently-borrowed largely from the Sanskṛit after it was thus elaborated.
32 It has been recently stated in print that Russian furnishes an exception to the usual ramification into dialects, but Mr. Morfill informs me that it has all the characteristics of Āryan languages, separating first into Great and Little Russian and then into other dialects.
33 See the chapters on the Epic poems in “ Indian Wisdom,” and my edition of the “Story of Nala,” published at the Clarendon Press, and my little work on “Indian Epic Poetry” (now scarce).
34 The late Professor Bühler has shown that the inscriptions of about 500 A. D. quote the Mahā-bhārata and describe it as containing 100,000 verses.
35 I do not mean this remark to apply to Buddhistic literature, which is very extensive, and is partly in Sanskṛit, and has much still unedited and untranslated. The Divyâvadāna, edited by Professor E. B. Cowell and Mr. Neil, is an example. It is written in Sanskṛit or rather in a kind of Sanskṛitized Pāli, or Pāli disguised in Sanskṛit garb. Other Buddhist Texts, written in Sanskṛit, are now being ably edited by the well-known Tibetan traveller, Rai Śarat Candra Dās, Bahādur, C. I. E., to whom I was greatly indebted for help in my researches at Darjeeling and its neighbourhood. Much Jaina philosophical literature, too, is still unedited, although well worthy of attention, and although only occasionally referred to in this Dictionary. It is written in Sanskṛit as well as in Ardha-Māgadhī Prākṛit, for the elucidation of which Professor Leumann has done such excellent work. In fact, the Sanskṛit form of Jaina philosophical literature (now being ably expounded by Mr. Vīrcand Ghāndhi at Chicago) still offers an almost wholly unexplored field of investigation. Furthermore, it must be admitted that in some cases better editions of pure Sanskṛit works are needed. For example, a better critical edition of the Mahā-bhārata than those of Calcutta and Bombay is a desideratum. The Southern Recension of that immense work is I believe engaging the attention of Dr. Lüders, Librarian of the Indian Institute.
36 A section of it has been printed in Calcutta.
37 The use made of some of the series is thankfully acknowledged at p. xxxii; but it is surprising that the long line of 49 thick octavo volumes includes no complete translation of India's most sacred book — the Rig-veda. Only about 180 out of 1017 hymns are translated in vols, xxxii and xlvi, when a continuous English version of all the hymns might have been given in one volume. It is regrettable, too, that vol. xiii only gives about a third of the Atharva-veda hymns, and that the Bhāgavata-purāṅa, which is a bible of modern Hindūism, has no place in the list, while some volumes give translations of far less important works, and some give re-translations of works previously translated by good scholars.
38 The Phœnician inscriptions have been deciphered by assuming that the Phœnician language must have been akin to Hebrew. Although their age cannot be ascertained with absolute certainty, yet there is good reason to believe that some of them are of greater antiquity than the cognate Moabite inscription of King Mesha which was found at Dibon, a little N.E. of Jerusalem and south of Heshbon.
39 Some of these tablets show that diplomatic correspondence passed between Babylonia and Egypt through Palestine. In fact, “ Babylonian” was in those days the language of diplomacy, as French once was in Europe. Other tablets in Babylonian cuneiform character have proved to be letters written by the king of Jerusalem to the Egyptian monarch to whose suzerainty he appears to have been subject.
40 There are two kinds of Himyaritic inscriptions, viz. Sabæan and Minæan.
41 Notwithstanding the elaborate proofs given by the Abbé Van Drival in his ingenious and interesting treatise on “l'origine de l'écriture.”
42 See note 2, p. xxvi.
43 In Book viii, 168 written legal documents are mentioned.
44 He gives the words lipi and libi in one of his rules (iii, 2, 21).
45 The bark of the Bhoj (or Birch) tree and the leaf of the palm seem to have constituted the chief material used by the Hindūs till the introduction of paper by the Muhammadans. No such durable materials as Egyptian papyrus or European parchment — the latter being prohibited on account of its impurity — seem to have been employed.
46 See note 3, p. xxv. [This note is placed below]
[In the same way the great Arabian Teacher Muhammad declared in the first Sura of the Kurān (according to Rodwell, p. a, and Sale, p. 450 with note) that “God taught the use of the pen.” Even some Christians may not be indisposed to agree with Hindūs and Muhammadans in holding that the faculty of writing, as an instrument for the expression of thought — although dormant through all the early ages of the world's history — is as much a divine gift as language. Muhammad's view, however, of the divine origin of writing consisted in declaring that the Kurān descended ready written from heaven.]
47 Hindī when so transliterated is called Hindūstānī or Urdū.
48 Litera, “a letter,” is derived from lino, “to smear,” just as Sanskṛit lipi from lip. If a corresponding word were to be used in Sanskṛit it would be lipi-śāstra. The word akshara, which is the Sanskṛit for a letter, properly means “indelible,” and this meaning seems to point to the use of letters in early times for inscriptions on stones and metal. Similarly the first meaning of lekha is “scratching with a sharp point.”
49 Pandit Śyāmajī in his second paper, read at the Leyden Congress, said: “We in India believe even at the present day that oral instruction is far superior to book-learning in maturing the mind and developing its powers.”
50 Certainly, as I think, the change of direction in the writing may have been due to Greek influence. Pāṇini, who probably lived about 400 B.C, gives as an example of feminine nouns the word Yavanānī which Kātyāyana interprets to mean “the Greek alphabet;” and we know that Greek coins and imitations of Greek coins, unearthed in North-western India, prove the existence of that alphabet there before Alexander the Great's time. Hindū receptivity of Greek influences is illustrated by the number of astronomical words derived directly from the Greeks to be found scattered throughout the pages of the present Dictionary.
51 Aśoka, who called himself Priya-darśin, and was the grandson of Candra-gupta, did for Buddhism what Constantine did for Christianity, by adopting it as his own creed. Buddhism then became the religion of the whole kingdom of Magadha, and therefore of a great portion of India; and Aśoka's edicts, inscribed on rocks and pillars (about the middle of the third century B.C.), furnish the first authentic records of Indian history. Yet the language of these inscriptions cannot be said to be exactly identical with so-called Māgadhī Prākṛit, nor with the Pāli of the Buddhist sacred scriptures, although those forms of Prākṛit may be loosely called either Māgadhī or Pāli. Nor was the name Pāli originally applied to the language of the Buddhist Canon, but rather to the line or series of passages constituting a text (cf. the use tantra). According to Professor Oldenberg the Vinaya portion of the texts existed in its present form as early as 400 B.C. The later Buddhist texts were written down not long after, and commentaries have since been compiled in Pāli and the languages of Ceylon, Siam, and Burma; the Pāli of Ceylon being affected by intercourse with Kalinga (Orissa).
52 See this Kharoshṭhī fully described in Professor Bühler's book. The first names given to it were Ariano-Pāli, Bactro-Pāli, Indo-Bactrian, North Aśoka &c. Sir A. Cunningham called it Gāndhārian. Pandit Gaurī-Śaṁkar, in his interesting work Prācīna-lipi-mālā written in Hindī, calls it Gāndhāra-lipi. Some think that Kharoshṭhī is derived from the name of the inventor.
53 A variation of it called Bhaṭṭiprolu is described by Bühler.
54 In the same way the great Arabian Teacher Muhammad declared in the first Sura of the Kurān (according to Rodwell, p. a, and Sale, p. 450 with note) that “God taught the use of the pen.” Even some Christians may not be indisposed to agree with Hindūs and Muhammadans in holding that the faculty of writing, as an instrument for the expression of thought — although dormant through all the early ages of the world's history — is as much a divine gift as language. Muhammad's view, however, of the divine origin of writing consisted in declaring that the Kurān descended ready written from heaven.
55 For the language of the inscriptions, see p. xxv, note 3. [This note is placed below]
[In the same way the great Arabian Teacher Muhammad declared in the first Sura of the Kurān (according to Rodwell, p. a, and Sale, p. 450 with note) that “God taught the use of the pen.” Even some Christians may not be indisposed to agree with Hindūs and Muhammadans in holding that the faculty of writing, as an instrument for the expression of thought — although dormant through all the early ages of the world's history — is as much a divine gift as language. Muhammad's view, however, of the divine origin of writing consisted in declaring that the Kurān descended ready written from heaven.]
56 These letters are shown in Professor Bühler's tables.
57 Our invaluable decimal notation certainly came from India, and may be said to conform to Semitic methods in the direction of the notation, inasmuch as units are placed on the right, while tens and hundreds are on the left.
58 Professor Bühler's first table in his work on Indian Palaeography would have been more convincing had he given examples of collateral and intermediate Semitic forms.
59 Such as the Bengālī, the Marāṭhī, Gujarātī &c., some of which may be usefully studied as presenting forms more closely resembling the ancient Brahma letters.
60 A similar line is often drawn in English copybooks and on writing paper as an aid to straight writing, but always below, not above the letters.
61 Dr. Lüders, of the Indian Institute, has kindly assisted me in the right formation of some of the inscription letters. The roughness of some is due to their being photographs from original impressions.
62 Some of the inscriptions had not the full complement of vowel-signs. As a matter of fact I find that in some inscriptions a list of only thirty-five letters in all is given, while in others there are thirty-six, and in others again thirty-nine. Professor Bühler says (p. 82 of his latest work published in 1898) that the ordinary Brahma alphabet has forty-four letters traceable in the oldest inscriptions (including the Bhaṭṭiprolu) which with au (derived from o) would make forty-five, and with the mark for Visarga which “first occurs in the Kushana inscriptions” forty-six. The common reckoning for the vowels, as taught in indigenous schools, makes them only twelve.
63 The oldest known inscription in Sanskṛit is on a rock at Junā-garh in Kāthiāwār. It is called the Rudra-dāman inscription, and dates from the second century a.d. It is not in Nāgarī, but in old inscription letters. The Bower MS. of about 400 a.d. shows a great advance towards the Nāgarī, while Danti-durga's inscription of about 750 A.D. exhibits a complete set of symbols very similar to the Nāgarī now in use. It is noteworthy, however, that the first manuscript in really modern Nāgarī is not older than the eleventh century a. d.
64 This ă is the a of our words “vocal organ” (pronounced vocul orgun), Sanskṛit does not possess the sound of a in our “man,” nor that of o in our “on.” As a consonant cannot be pronounced without a vowel, the Brābmans chose the commonest of their vowels for the important duty of enabling every consonant to be pronounced. Hence every consonant is named by pronouncing it with a (e. g. ka, kha, ga &c.). It is, I suppose, for a similar reason that we have used the common vowel symbol e for naming many of our English letters.
65 Take, for example, the following transliterated words in a recent pamphlet by a native: — Devi, puja, Durga, Purana, ashtami, Krshna, Savitri, Acoka, Civa &c. I have even seen crab written for the Hindūstānī , “bad.”
66 Forster gives an example of one compound word consisting of 152 syllables. This might be matched by even longer specimens from what is called Campū composition.
67 We may, at least, entertain a hope that the hyphen will not be denied to Sanskṛit for the better understanding of the more complex words, such, for example, as vaidikamanvādipraṇītasmṛititvāt, karmaphalarūpaśarīradhārijīvanirmitatvābhāvamātreṇa, taken at haphazard from Dr. Muir's Texts. We may even express a hope that German scholars and other Europeans, who speak forms of Āryan speech, all of them equally delighting in composition, may more frequently condescend to employ the hyphen for some of their own Sesquipedalia Verba, thereby imitating the practical Englishman in his Parliamentary compounds, such, for example, as Habeas-corpus-suspension-act-continuance-Ireland-bill.
68 In the paper on transliteration, which 1 read at the Berlin International Congress, I proposed a kind of mark of accentuation to represent aspirated consonants, as, for example, k', p'. To say (as at p. xxxvi) that aspirated k or p is like kh in inkhorn or ph in uphill is to a certain extent misleading. It is simply k or p pronounced as in Ireland with a forcible emission of the breath.
69 Surely we ought to think of our Indian fellow-subjects who in their eagerness to learn the correct pronunciation of English would be greatly confused if told that such good old English words as pinch, catch, chin, much, jump, jest, ought to be written pink, cak, kin, muk, gump, gest.
70 This will be found at p. 3 of the “Original Papers illustrating the History of the Application of the Roman Alphabet to the Languages of India,” edited by me in 1859.
71 Among other numberless publications a most accurate edition of the Rig-veda itself, edited by Professor Aufrecht, was printed in the Roman character, and published in two of the volumes of Professor Weber's Indische Studien.
72 See especially my paper read before the R. A.S., April 21, 1890.
73 In 1858 I wrote strong letters to the Rev. Henry Venn, deprecating the system of transliteration then adopted by the C. M. S. It has been recently remodelled on the lines of the Geneva Congress report.