Gonda, J. Some Notes On The Study of Ancient Indian Religious Texts
Gonda, J. Some Notes On The Study of Ancient Indian Religious Texts
Gonda, J. Some Notes On The Study of Ancient Indian Religious Texts
1, No. 2 (Winter, 1962), pp. 243-273 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/1062054 . Accessed: 08/05/2012 17:21
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to History of Religions.
http://www.jstor.org
J. Gonda
SOME THE
NOTES STUDY
ON OF
Those students of comparative history of religions who are acquainted with the history of research in the special field of ancient Indian Weltanschauung and who take cognizance of the moot points and questions under discussion among Vedists and historians of Indian thought will have noticed that our knowledge of, and insight into, Vedic religion largely depend on a correct understanding of a considerable number of Indian words and phrases, many of which have now been debated for nearly a century. They will have observed that not rarely opinions on the exact sense of important religious terms continue to diverge widely, and in other cases solutions offered with much self-confidence and suggestiveness appear to be, sooner or later, open to justifiable criticism. It is not my intention in this article to dwell at length on some of the factors which have contributed to this state of affairs, which, after all, is unavoidable in any comparable field of scientific research: the distance in time, space, and cultural environment between Vedic mankind and most modern specialists; the incompleteness of our sources; the reinterpretations suggested by the traditional views of the Indians; the prejudices and limitations
243
on the part of the hearer;'1by studying the importance of the contexts and situations in which a word or word group is with a certain regularity used." Whereas, moreover, the study of semantics has for many years been mainly concerned with semantic change, that is, with historical problems of the semantic development of individual words, interest began, in the twenties and thirties of this century, to be focused also on a study of coherent, coexistent word groups forming so-called semantic fields and their relations to similar "fields" composed of the same or similar names as they existed at a later date.12 Eyes were opened to the possibility of distinguishing semantic "structures" and "structurations"-the latter term denoting the dynamic aspect, "le processus d'organisation structurelle."13It has been found that the "meanings" of the elements of a vocabulary group themselves so as to constitute wholes which are to a certain extent organized, the constituents maintaining mutual relations to each other as well as to the whole. There are "microstructures": "meanings" which are complex, consisting of semantic aspects, grouped round a "kernel"; there are also macrostructures or "fields" composed of groups of words which are in some way or other-morphologically, notionally, etc.-more closely associated. The very idea of "meaning" has, moreover, been subjected to criticism. We now know that "words" do not mean "things." "Meaning" is, in brief, a reciprocal relation between name (= Wortformor Wortkorper)and sense (Sinn or Begriff), between symbol and "thought" or "reference," which enables them to call up one another,14 the "idea" or "reference" relating to the "thing itself." This insight, however, implies that, in studying the meanings of, for instance, religious terminology of
10See, e.g., M. Leumann, "Zum Mechanismus des Bedeutungswandels,"InXLV (1927), 105 ff. (=Kleine Schriften[Zurich,1959], dogermanische Forschungen, p. 286). n J. Stocklein, Untersuchungenzur lateinischen Bedeutungslehre (Dillingen, 1895). 12 See, e.g., L. Weisgerber, "Vorschlage zur Methode und Terminologie der Indogerm. Wortforschung," Forsch.,XLVI (1928), 305 ff.; and by the same author, und Geistesbildung Muttersprache (Gottingen, 1929); J. Trier, "Das sprachliche Feld," Neue Jahrbucher fur Wissenschaftund Jugendbildung,X (1934), 428 ff. We cannot enter into details, e.g., into the question as to how far semantic distinctions were, in particular cases, assumed under the influence of those who, afterward, began to reflect upon definitions, border-line cases, "synonyms," etc.; problems connected with the "adaptation" of terms when received into anthe history of Indian religion and philosophy. 13 Tatiana Cazacu, "La 'structuration dynamique' des significations,"in Mdlanges linguistiques(Bucharest:Acad6mieRoumaine, 1957), pp. 113 ff. 14See, e.g., C. K. Ogden and I. A. Richards, The Meaningof Meaning (London, 1923), 3d ed., 1930, esp. chap. i; Ullmann, op. cit., pp. 65 ff.; K. Ammer, EinI (Halle a.S., 1958), 55 ff. fihrung in die Sprachwissenschaft, 245
other community, etc. As is well known, these cases are far from imaginary in
headings of the time-honored "figures of speech" and their modern reductions to the three logical categories of narrowing, widening, and transfer of sense17are largely dominated by a priori conceptions and are little more than highly simplified schematic formulations of very complicated and often prolonged processes.'8 Backgrounds, determining factors of a historical, social, and psychological order are left out of consideration; complex phenomena of different character are classified under one and the same denominator, because it is only the results of semantic shifts-if there are any-that are in a very superficial way taken into account. Little indeed, with a view to a deeper understanding of ancient Indian thought and Weltanschauung, and of Vedic man's endeavor to penetrate into the hidden world beyond the phenomena, is gained by calling a definite contextual connotation of a word a metaphor or a "transferred meaning," or in observing that, for example, the Vedic amSu, meaning "the filament of the soma," may, by way of metonomy, be used for the soma-juice. What matters is to know why "these two meanings combined," what made the Vedic poets use this word in what would appear to us to be "two senses." What we would really like to know is by way of which association definite words were used in a "figurative" way-for example, the verb tan- "to stretch," to denote the idea of "performing the sacrifice''9-or word groups were formed which impress us as metaphorical-what was, for instance, the exact meaning of the words Rgveda 8, 48, 6 translated by Geldner: "wie das ausgeriebene Feuer sollst du (0 Soma) mich in Feuer setzen"?20We would like to know whether there exists a preference for using words belonging to definite semantic groups in so-called transferred senses; how far the use of identical words reflects ideological identifications, etc. We may go further: When Geldner,21in a note to the Soma-hymn Rgveda 9, 29, 3 vardhd samudram "fill the ocean," observes that "ocean" here means "die mit dem Meere verglichene Menge des gepreszten Somas in der Kufe," the term "metaphor" would conceal the important fact that the ancient priests considered the celestial ocean (not an ordinary sea) and the soma-vessel to be identical, however much modern men would be inclined to take the existence of a mere sensual association between
Ullmann, op. cit., p. 204. Cf. also J. R. Firth, Papers in Linguistics (London, 1957), p. 10. 19 A. A. Macdonell puts it as "figurativelyin the sense of to extend the web of the sacrifice" (A Vedic Readerfor Students[Oxford,1928], p. 198). 20 K. F. Geldner, Der Rigvedain Auswahl, I (Stuttgart, 1907), 83. 21K. F. III (Cambridge,Mass., 1951), 28. Geldner, Der Rig-vedaubersetzt, 247
18
17 See
Ancient-Indian
Religious Terminology
the soma contained in the large vessel and a real sea for granted. Thus freier Raum and Ausweg (aus der Not) are not completely adequate "equivalents" of "an original" and a "transferred" meaning of varivas (RIV. 4, 24, 2), or rather: Whereas the German expressions may be related to each other as proper sense and metaphor, the Indian word appears to express two context-bound nuances of one and the same "vague concept," which is subject to semantic association and amplification. And here the question also arises as to how far these expressions which impress us as "metaphores," transferred meanings, or figurative speech were "motivated" (i.e., felt as vivid, active, and expressive) and how far they were cum or sine fundamento in re, that is to say, either transferred or "figurative" uses based on the intuition of some real likeness of relations and belonging to the well-known and highly frequent type that has become ingrained into our common habits of expression, or indicative of a propensity to "identifications" and belonging to those products of speculative thought and imagination which play such an important role in the Weltanschauung of prescientific communities.22 What deserves special notice is the inclination of lexicographers and commentators to distribute the aspects of the total meaning of a term over a number of "senses" arranged in an order which though impressing the reader as reflecting a historical development is only a product of the ancient procedure of "logical" classification. Thus damsas, which means something like "marvelous skill or power," is believed to "mean": "1, feat, Meisterwerk; 2, iibernatiirlichesVermaya is said to have, in the Rgveda, two distinct meanings: mogen";23 whereas this "1, Verwandlung, Zauberkraft;2, Illusion, Tduschung,"24 term as far as I am able to see25has, in fact, denoted "an incomprehensible wisdom and power ascribed to mighty beings and enabling its possessors to create or to do something which is beyond the ability of ordinary men"; druh is considered to be, on the one hand, dharman, according Falsch, Falschheit, and on the other, Tduschung;26 to the dictionaries, "established order of things," "steadfast decree" as well as "practice and custom." Sometimes the occurrence of a "specialized" meaning is assumede.g., ild "invigoration, sp6cialis6 en breuvage invigorant (offert a
22 See, e.g., E. Leisi, Der Wortinhalt, seine Struktur im Deutschen und Englischen (Heidelberg, 1953). 23 Geldner, Der Rigveda in Auswahl, I, 78. 24 26
Ibid., p. 135.
See my "Sense and Etymology of Sanskrit Maya," in Four Studies in the Language of the Veda (The Hague, 1959), pp. 119 ff. 26 Geldner, Der Rigveda in Auswahl, I, 88. 248
un dieu au sacrifice.. .)"27-or an abstract term is said to express a concrete sense where a closer investigation into the use of the term and the idea for which it stands may have us question the correctness of the statement. We should not forget that all men, especially those who have not undergone a special intellectual training, are often inclined to refer to manifestations, results, materializations, etc., of power rather than abstractions and generalizations.28The sprachlichen Vorstellungen normally result from experience acquired in numberless concrete situations in which the results and consequences are, as a rule, more evident than causes and determining factors, individual cases more significant than generalizations; representations, localizations, and manifestations more perceptible than the "powerconcepts" themselves. Hence the well-known feature of many vocabularies to refer to "power-concepts" and their manifestations, to actions and effects, to ideas and their materializations by the same word.29 The Greek vifpts, for instance, is "outrage" as well as "insolence," and lexicographers remark that "it is often difficult to separate the concrete sense from the abstract"; aper' is "excellence" and "glorious deed" or "active merit; reward of excellence." In Sanskrit, sravas does not only denote "glory" but also "glorious deed(s)"; yasas not rarely refers to those objects or circumstances from which man derives honor, and a horse may be called a vdja (which roughly speaking seems to be the generative power by which new food and new life is obtained).30 Often powers and divinities are essentially identical with their manifestations and vice versa.31 Daseinsmdchte, which we would like to interpret as "abstract ideas,' mainly were the totality of all objects, persons, and phenomena, in which and by which they manifested themselves. At a certain stage of development "un Mo27L. Renou, "Hymnes a Varuna," in Etudes vediques et pdnineennes, VII (Paris, 1960), 10. 28 See, e.g., W. Havers, Handbuch der erkldrenden Syntax (Heidelberg, 1931), p. 115; Kronasser, op. cit., pp. 114 ff. 29 This is, of course, not to deny that an "abstract" term can assume a "concrete" sense. 30 These facts may, of course, also be illustrated by "ethnological parallels," but they do not stand or fall with their reliability, as is suggested by P. Thieme ("Brahman," Zeitschrift der deutschen morgenldndischen Gesellschaft, CII, 97), who, pursuing his own lines of thought, has completely misunderstood my argument. It is not clear to me how my words: "all that is connected with such power-concepts or represents them can, in principle, bear the same name (i.e., all that is connected with vaja may be called vdja, all that is of the nature of ild may bear the name ild, etc.)" (Notes on Brahman [Utrecht, 1950], p. 39), should be interpreted as: "Brahman kann alles, was nur irgend mit einer Kraftvorstellung verbunden ist, bezeichnen." 31See also P. Radin, Die religiose Erfahrung der Naturvolker (Zurich, 1951), pp. 58, 75. 249
1950). 34E. Sapir, Selected Writings (Los Angeles, 1949), pp. 160 ff. and 389 ff.; B. L. Whorf, Four Articles on Metalinguistics (Washington, D.C., 1949); J. H. Greenberg, "Concerning Inferences from Linguistic to Nonlinguistic Data," in Language in Culture, ed. H. Hoijer ("American Anthropological Association Mem.," No. 79 [Chicago, 1954]), pp. 8 ff.; S. Newman, "Semantic Problems in Grammatical Systems and Lexemes, in Language in Culture,p. 89; H. Hoijer, "The Relation of Languageto Culture,"in Anthropology Today,ed. A. L. Kroeber (Chicago, 1953), pp. 554 ff.; R. Lado, Linguistics across Cultures (Ann Arbor, 1957), pp. 77-78. ed. H. Hoijer ("AmericanAnthro36F. G. Lounsbury,in Languagein Culture, pological Association Mem.," No. 79 [Chicago, 1954]), p. 137. 250
33 See,
van der Leeuw, op. cit., p. 35. e.g., L. Weisgerber, Vom Weltbildder deutschenSprache (Diisseldorf,
jeweils ein in sich vollkommen geschlossenes Seinsbild schafft."36 Striking examples may be given with respect to terms relating to the physical environment. Whereas the speakers of the modern Western languages, which have about seven or eight principal color termswhite, yellow, red, blue, green, brown, black-are, in a way which is for themselves a matter of course, accustomed to divide the continuum of the natural color spectrum in the first instance into these "principal colors"; the ancient Greeks, whose language has another classification, had, for instance, to resort to one and the same word where we would say either "yellow," "green," or "grayish-brown." Whereas the American language, Navaho, has two terms roughly corresponding to our "black," it denotes "blue" and "green" by a single term. This has nothing to do with color-blindness on the part of the ancient Greeks and other peoples, as was believed by some classical philologists some sixty years ago. Nor does it prevent the speakers of these languages from using terms comparable to "cornflower blue," "blood red" to indicate color nuances. Although the conclusion that those speaking a language can be aware only of those distinctions which are provided by semantic differences in words and idioms would, indeed, be an exaggeration, the "world" in which they live is to a large extent unconsciously built up on the language habits of the community to which they belong. That the difficulties encountered in translating are for the greater part due to the differences in what was called by Von Humboldt the "inner speech-form" has over and over again been argued, by Schopenhauer37-who, while drawing attention to the differences between German Geist, French esprit, English wit; Greek opjni, Latin impetus, German Andrang; French malice, German Bosheit, English wickedness, observed that all translations necessarily are imperfect and defective: "fast nie kann man irgendeine charakteristische, pragnante, bedeutsame Periode aus einer Sprache in die andere so iibertragen, dasz sie genau und vollkommen dieselbe Wirkung hat"38-and by modern linguists and anthropologists who have attempted to penetrate into the different "worlds of reality" in which peoples speaking different languages live: the understanding of a text "involves not merely an understanding of the single words in their average significance, but a full comprehension of the whole life of the community as it is mirrored in the words, or as it is suggested
Weisgerber, Weltbild, p. 159. Schopenhauer, Parerga und Parallipomena, Vol. II, chap. xxv. 38 Cf. also, e.g., H. Giintert and A. Scherer, Grundfragen der Sprachwissenschaft (Heidelberg, 1956), pp. 54-55.
37 36
251
and devotional;42 It is man's participatingof God, at once "intellectual" of the atman's total subservienceto God, it is the constant rememorization inspiredand animatedby a perfect love of worshipin which the knowledge of God as the possessorof all perfections,as the mercifulsaviourand as the sole cause of the universecompletelyterminates.It culminatesin a mystic ecstasy of love so ardentthat the aspirantcannotlive for a momentseparated from God: all his happinessdepends on his contact with God; his most of his all-pervading love for God. humbleact is an expression When, therefore, in some recent publications in the field of Vedic religion attempts were made to translate important Sanskrit terms by one single modern European word, there is a strong a priori probability that the conclusions at which the authors arrive are to some extent erroneous. In his remarkable posthumous book on Varuna H. Liiders43 endeavors to show that the much discussed rta, of which the god is said to be a "guardian," is completely identical with German die Wahrheit.Although this sense is somewhat specified: "Rta bezeichnet ausschlieszlich die Wahrheit des gesprochenen Wortes oder des Gedankens,"44 no definition is given. But here we are
39H. Hoijer, in Language in Culture, p. 92. 40See, e.g., H. v. Glasenapp, Die Philosophie der Inder (Stuttgart, 1949), pp. 60, 488. 41By J. A. B. van Buitenen, Rdmdnuja on the Bhagavadgltd (thesis, Utrecht, 1953), p. 22. 42 These terms too should not lead us astray! 43 H. Liiders, Varuna (Gottingen, 1951-59). For an ample discussion of the special problem under consideration see my review which is to appear in the periodical Oriens (Istanbul-Leiden, in press). 44 Liiders, op. cit., p. 635. 252
confronted with another difficulty which would appear to be likewise minimized by many authors, namely, the extreme vagueness of many words and idioms in any language.45Being largely based on unanalyzed mental wholes, "names" as used by the ordinary speaker often stand for vague and unanalyzed "ideas," which are often surrounded by an aura of emotions and impressions. What is Wahrheit?"Quid est ergo tempus?" St. Augustine46exclaimed, "si nemo ex me quaerat, scio, si quaerenti explicare velim, nescio." Implicit vagueness is indeed, though highly variable, the most striking characteristic of word sense. It is a consequence of the process of abstraction by which our "concepts" are evolved. There is a wide gap between the virtual sense of a word in the language system and the actualized sense of speech contexts. Scholars are too often inclined tacitly to assume the existence, in the usage of the average speaker, of the clear-cut demarcation lines delimiting their own scientific concepts. In reality, the sense of a word is essentially "open," inviting supplementation. This openness and lack of firm contours is, Ullmann rightly observes,47 reflected in the "zonal" structure of the sense, the belts of varying determinateness clustering around its inner core. The mental content corresponding to abstract notions is admittedly still less distinct, the lack of sharp demarcation being not rarely a property of the referent itself. Often one can hardly imagine how an abstraction could exist at all without the help of language.48 What then is, according to Liiders, Wahrheit? Is it some "idea" vaguely opposite to "lie" or "falsehood," or is it something like "sincerity" or some other indefinite notion applied by those who speak German without unanimousness to a variety of concrete facts or situations? Or should we believe Wahrheitto express the substantival idea corresponding to what is, in explanation, added to the adjective "wahr" in some authoritative German dictionary? Or should we look for a definition in the works of a distinguished German philosopher? To these questions Liiders does not answer. Nor does he inform us of his view as to whether rta may, or must, be translated, into French
46 On the lack of precision of many words see, e.g., K. O. Erdmann, op. cit. (4th ed., Leipzig, 1925); S. Ullmann, op. cit., pp. 92 ff., 107-8 (with a Bibliography) and by the same author, Pr6cis de s6mantiquefrancaise (Paris-Berne, 1952), pp. 132 ff.; F. Paulhan, "Qu'est-ceque le sens des mots," Journal de psychologie, XXV (1928), 289 ff. 46 Augustine, Confessionsxi. 26. 47Ullmann, Principles, p. 93. 48 When anything is describedby a single word, the idea is apt to be represented as an actualization without accidents of a thing in itself, endowed with an independent existence. See also Toshihiko Izutsu, Languageand Magic (Tokyo, 1956), chaps. v and vi.
253
Ancient-Indian
Religious Terminology
by verite, into English by truth. (As is well known German Wahrheit, English truth, Latin veritas "true or real nature, reality; truthfulness, truth, integrity, etc.," are not synonyms.) To put it briefly, did Liders really mean that the inherent vagueness, the connotations, and the range of applicability of German Wahrheit-whatever its value as a practical expedient in a rough translation-coincide with the Vedic rta, which forms part of the vocabulary of a community whose views of reality, the nature, power, and function of human speech, words, and statements, and the mutual relations between the spoken word and reality were different from those of both the average German and the modern German scholars and philosophers?49 How easily we may be liable to misunderstandings with regard to the content and range of application of words belonging to archaic and foreign cultures may appear from Lilders' argument50that the term satya by which rta was in the course of time replaced, and which is, in German, likewise translated by wahr, was a synonym ("rta und satya ... (sind) zwei ganz gleiche Dinge"). It would rather appear to me that both words symbolize complementary ideas-compare, for example, Taitt. Samh. 5, 1, 5, 8 rtam satyam ity aheyam vd rtam asau satyam " 'Ttamsatyam,' he says, this (earth) is rta, yonder (sky) is satyam." A thorough investigation into the sense expressed and the syntactic combinations formed by these words-which cannot, of course, be instituted here-will no doubt reveal a considerable number of more or less similar marginal meanings as well as a difference in semantic kernel and range of application between these two terms.51The etymological sense of satya "belonging to, related to the sat, that is, the existent, being, real" is not rarely undeniable; it is often used to qualify an "object" as really being what it is said or thought to be, as being in harmony or agreement with real facts or reality. That however "reality" (sat) and its oppositum denoted by asat were to the mind of Vedic man not identical with our concept of reality-in whatever sense we would prefer to take it-may
49 No more than passing mention can be made here of the critical remarks made by other scholars. Renou (op. cit., VII, 16), while justly observing that "aucune traduction ne saurait rendre rta, terme h6rit6, qui 6tait sans doute per9u comme une entit6 inanalysable par les ri" is, in contradistinction to P. Thieme, who regards the problem as settled (op. cit., CI, 418), and M. Mayrhofer, Kurzgefasztes etymologisches Worterbuchdes Altindischen, I (Heidelberg, 1953), 122, who from the point of view of meaning leaves his readers in the dark: "rta 'Wahrheit' zu *ar'fiigen"'-of the opinion that a meaning "order" "couvre commodement l'ensemble de cette pensee 'corr6lative' qu'on salt depuis Bergaigne gtre la trame meme du 1lgveda; 'verit6' n'est qu'aspect, a notre avis, de 'ordre' et un aspect secondaire qu'il n'y a pas profit a promouvoir au rang d'acception l16mentaire." 60 Liiders, op. cit., pp. 406 ff., 642. 51Some details may be found in my above review of Liiders' book.
254
appear from the important article by Norman Brown52on which it would have been interesting to learn Liiders' opinion. After having collected the references to the structure of the universe, Professor Brown arrives at the conclusion that the universe,as .Rgvedic man saw it, was in two parts.Onebeingthat in which the gods and men live..., this he called Sat "the Existent." Below the earth... was a place of horror,inhabited only by demons, the Asat (the Non-Existent).... To make the Sat operate perfectly, every creature had his duty, his personalfunction (vrata),and when he lived by it he was an observerof the Rta, the inhabitantsof the Asat lookingfor every opportunity to injurethe R.ta-observing beingsof the earth and sky. Norman Brown therefore translates rta by "universal cosmic law," which, of course, is also an attempt at elucidating what the Vedic authors may have meant rather than an equivalent. Here we encounter another weak point of many arguments in the field of the history of religious thought: the ease with which two or more indigenous terms are declared to be synonymous, whereas competent linguists are agreed that total synonymity is an extremely rare occurrence.53The senses of two "names," though superficially regarded as identical, are indeed rarely coextensive, partly because of their inherent vagueness and partly because of their different emotive "overtones." Terms such as "liberty" and "freedom" or "aid" and "assistance" are only pseudo-synonyms, because they cannot, without suggesting any difference in either cognitive of emotive import, replace each other in any given context. If, therefore, Liders' opinion54that "rta in (Rgveda) 1, 46, 41 ein Synonym von gir, stoma, hava, brahman, pratistuti und mantra ist"-these names are rendered by "(Kult)lied"-should be understood literally, it would be hard to substantiate.55All those terms have their own connotations, their own range of meaning, referring to definite aspects of ideas for which we, perhaps, have terms of our own; or rather, they denote, in definite contexts, special aspects or applications of "ideas"-their semantic kernels-for which we often have no simple names, and of which we cannot always easily determine the dominant semantic
52 W. Norman Brown, "The ligvedic Equivalent for Hell," in Journal American Oriental Society, LXI (1941), 76; "The Creation Myth of the Rig-Veda," op. cit., LXII (1942), 85. 63 See, e.g., L. Bloomfield, Language (London, 1935), p. 145; Ullmann, Principles, pp. 108 ff. and passim; Ch. Bally, Trait6de stylistique francaise, I2 (Heidelberg-Paris), 96-97, 140 ff. 54 Liiders, op. cit., p. 438. 66The differences between some terms belonging to this "semantic field" were discussed by Renou, "Les pouvoirs de la parole dans le R.gveda," Etudes et pdnin6ennes,I (Paris, 1955), 1 ff. vediques
255
ancient Indian terms-like the often hopelessly divergent explications and translations of one and the same word59-is only a consequence of the impossibility of translating them into our languages. I for one am not convinced that those texts,60 which exhibit rta in the sense of weltschaffende und welterhaltendeMacht are from a "logical" and "historical" point of view secondary in character. Rather, it would appear to me, that rta, in the Rgveda, is a cosmic, metaphysical Daseinsmacht6--that is, "power-substance" which, within some form of experience, is supposed to be present in persons, things, nature, and phenomena and by virtue of which these are, each in their own way, powerful, influential, effective, and endowed with something which is beyond the bounds of normal human understanding-which makes its existence felt in the regular course of the natural phenomena, in the harmony and regularity of the normal (and therefore right) and natural (and therefore real) condition and character of the processes in nature and cosmos, in the world of men as well as in that of the gods; that it is a constructive and fundamental principle accepted to express the belief in a harmonic structure of the universe and a regular course of the phenomena occurring in it. This principle which gives manifold evidence of its existence may also materialize in human speech, in the word of the poet by which it is stated and described and which, if it is believed to be in harmony with the rta, assumes the character of "truth." A point on which professors Thieme62and Renou63disagree concerns the application of a principle adopted by the latter to establish, wherever possible, the sens initial of a name. As, however, the great difficulty is that the initial sense is in so many instances not known, Thieme advocates the view that we must hazard a conjecture as to what might be a likely "initial meaning" (or acception authentique, linguistiquement valable); the correctness of that conjecture must be established experimentally: if the "central idea" hypothetically adopted is recognizable in all the passages of the Rgveda-why should
69 Thus dharmawas, in the last decade, rendered by "the divinely ordained norm of good conduct" (Basham); "moraland religiousduties" (R. C. Majumdar and others); "law, nature, rule, ideal, norm, quality, entity, truth, element, category" (P. T. Raju); "moral law, merit, virtue," or "ethical living" (Radhakrishnan); "a religion which sets up laws and rules" or "Tugendiibung; das geheiligte Gesetz" (Eidlitz); "divine moral order"or "life-task and duty" (Zimmer). 60 Quoted by Luders, op. cit., pp. 568-80. 61For Daseinsmachte see H. von des indischen Glasenapp, Entwicklungsstufen Denkens (Halle a.S., 1940), pp. 9 ff. 62 See Thieme, Review of Renou's Etudes vediqueset panin6ennes,I, Journal AmericanOrientalSociety, LXXVII (New Haven, 1957), 51 ff. 63 Renou, "Les pouvoirs de la parole dans le Itgveda," op. cit., I, 1 ff.
257
258
Not infrequently, however, authors make, on the tacit assumption that a Vedic weltanschaulicheterm may be translated by one modern word, an attempt at testing a hypothesis with regard to the "meaning" of that term by investigating whether it fits in all the passages in which it occurs. In following this procedure, they have, however, sometimes overestimated the validity of its results and the cogency of their argumentation. In many cases the Procrustean method, of which we have already disapproved, allows them to regard any text, in which the substitution of a modern term for the original Vedic does not lead to a manifest absurdity, as a confirmation of their hypothesis.67 Another source of errors lies in the supposition that a, or the, meaning which belongs to a definite word in post-Vedic times must have been its "semantic nucleus" from the earliest texts. Both pitfalls proved detractive to the merits of the book on vrata-one of the key words of the Rgveda, a correct understanding of which is vital for gaining an insight into the religious attitude of its poets-by H. P. Schmidt,68 in which "die konstante Ubersetzung 'Geliibde' sowohl zu merkwiirdigen inhaltlichen Konsequenzen fiihrt ['das ganze Naturgeschehen beruht nach diesen beiden Strophen auf Geliibden,' S. 26], als auch von vornherein die Moglichkeit sprachlicher Entwicklung ausschlieszt."69 The translation Geliibde ("vow, solemn and inviolable promise") is, however, manifestly incorrect, because in the R1gveda a vrata-the term occurs over 200 times-is never, like a vow, made or taken, and practically limited to the sphere of the gods; it is, moreover, impossible to describe the fact that a god has extended sky and earth (I.V. 3, 6, 5), marked off the expanse of the earth (8, 42, 1), or simply came (2, 24, 12) as his Gelibden.70 The same term vrata-which sometimes seems to verge on the ideas of rule of conduct, fixed and regular behavior, function, observance-may serve to illustrate another methodical imperfection: a definite "meaning"-which, as already stated, often exists only in a translation-is considered to be from the historical point of view primary or original on account of etymological71 arguments. Accord67For similar criticism see W. P. Schmid, in Kratylos, V (Wiesbaden, 1960), 44. H. P. Schmidt, Vedisch "vrata" und awestisch "urvata" (Hamburg, 1958). 69 W. P. Schmid, op. cit., p. 45. 70 See also Renou, op. cit., VII (Paris, 1960), 9; "Gelubde: traduction plausible A condition qu'on y integre conventionnellement les valeurs que definit Schmidt mais que le mot "vceu" est incapable de porter sans commentaire." 71 The technical term "etymology" is used here in the traditional sense: "the tracing of a word back to its original form and meaning by the methods of comparative linguistics," because that is what it means to the authors quoted and what is meant in the text. For a more modern view of the task of the etymologist see W. von Wartburg, Einfuhrung in Problematik und Methodik der Sprachwissenschaft (Halle a.S., 1943), pp. 105-6.
68
259
D. Whitney, Journal American Oriental Society, XI (1885), 229 ff. See, e.g., A. A. Macdonell, Vedic Mythology (Strasbourg, 1897), pp. 21-22. 77 Max Miiller, Anthropological Religion (London, 1892), p. 82. 260
to construct an I.-E. dyeus "heaven, sky, day, also as a deity";78 it is but difficult to decide whether the lack of prominence of the deity (the "personal meaning" of the word) is inherited from the original Indo-Europeans-what was, in harmony with the evolutionist trends of thought of his days affirmed by Macdonell79-or has arisen from a special prehistoric development in Indo-Iranian, or was due to a preference, in the cultural milieu reflected by our Vedic texts, to other gods, for instance, to Indra.80It is, moreover, beyond doubt that the Sanskrit deva "god," like the Latin deus, derives from the same stem *dyeu-, which underlies the above *dyeus; but it would be imprudent to follow Hertelsl and Apte82in regarding the Vedic devas integrally or even as "luminaries"83-' das arische as "gods of light," Lichtmdchte Wort daiva, vedisch deva ist... abgeleitet von *diu 'Himmelslicht' ...,demgemasz sind alle arischen daiva Licht- oder Feuerwesen ..." because the texts, though sometimes associating the devas with the celestial light (see, e.g., R.V. 1, 19, 6) and connecting the latter with attribute the name to various kinds of the names of definite devas,84 superhuman and powerful beings fulfilling a variety of functions and concerned with different provinces of thought and nature. "It is absurd to suggest that when gods are opposed to demons the sky gods alone are meant, still more absurd to find them alone designated when gods, fathers, and men are discriminated."85How the "semantic shift" -which from the point of view of traditional semantics is only a "widening of meaning"-took place, how daeva, in the Avesta, came
78 See, e.g., M. Mayrhofer, Kurzgefasztes etymologisches Worterbuch des Altindischen, II (Heidelberg, 1957), 70. 79Macdonell, Vedic Mythology, p. 22, who is even inclined to defend the thesis that "the personification" was in Rgvedic times of a more advanced type than in the period of original Indo-European. 80 See, e.g., Max Muller, Lectures on the Origin and Growth of Religion (German trans., 1880), II, 398-99. 81 J. Hertel, Die Sonne und Mithra im Avesta (Leipzig, 1927), p. 2 and passim. 82 V. M. Apte, "All about 'vrata' in the Rgveda," Bulletin of the Deccan College Research Institute, III (1942), 407 ff. 83 C. D. Buck, A Dictionary of Selected Synonyms in the Principal Indo-European Languages (Chicago, 1949), p. 1464, is in contradistinction to Grace Sturtevant Hopkins ("Indo-European *deivos and Related Words" [Yale Univ. dissertation, 1932]), who questions the underlying notion of "brightness," inclined to ascribe to Zeus, luppiter, dyaus as well as Lat. deus, Skt. deva, etc., the common idea of "bright, shining." Cf. also the observation made by M. Eliade, Traite d'histoire des religions (Paris, 1949), p. 69: "Le simple fait que le nom du dieu aryen du ciel met l'accent sur le caractere brillant et serein n'exclut pas les autres th6ophanies ouraniennes de la personnalite de *Dieus." 84 I refer to C. W. J. van der Linden, The Concept of Deva (thesis, Utrecht, 1954), pp. 37-38. 86A. B. Keith, The Religion and Philosophy of the Veda and the Upanishads (Cambridge, Mass., 1925), pp. 75-76. 261
spriinglich als Vogel gedacht, und zwar war er eben der Vegetationsdamon (sowohl als Embryo wie als Vegetationsvogel) ...: griech. ist fast identisch mit Vi?vu."95 olwvos ["a large bird"< *6FLao-wos] Moreover, many etymologies, and especially those which connect a Sanskrit (or Greek, or Latin) word with a mere root-as is the case of the term vrata-must, from the semantic point of view, be hazy and indefinite, because the sense attributed to a root as a rule is a vague and abstract idea from which the senses of all derivatives are logically deducible. Similar remarks might be made with regard to other important names and terms. The "meanings" of the above term vratawere given in the order Gebot,Pflicht, Ordnungby those who subscribe to the view that this word etymologically belongs to Greek pr/jrp, "public speaker"; pTr pa, "verbal agreement," in the order Gewolltes, Gewdhltes, Geliibdeby those who derive it from var- "to choose."96Thus an "original" or "primary meaning" is not rarely adopted on account of etymological considerations. More generally speaking, many scholars are in some way or other inclined to consider those occurrences which are, or may be, in harmony with an etymological hypothesis as more "original": compare, e.g., Renou:97 a propos of RV. 3, 54, 5 "le sens (de vrata) est ici: 'domaine ou s'exerce la volont6 divine': cette analyse serait en faveur de l'6tymologie par vrt- zone de 'circulation. " It is, however, in my opinion incompatible with sound principles to suppose on the strength of etymological speculations, for instance, that, according to a prehistoric Indo-European view, the soul of the dead was a Schutzmacht, which made the crops grow or increase (the Vedic urvard "field yielding crop" explained as *urv-ald "growing by the souls": Avest. urvan, to be connected, then, with Vedic vr.oti in the sense of "warding off, keeping back").98 I cannot agree with V. Machek,99 who holds: uns stiitzend auf die Etymologie:Indraist ein Adjektivumindoeuropaischer Herkunftund bedeutete"stark,kraftig,"100 konnenwir ohne (ursprachlicher)
96 K. F. Johansson, Uberdie altindische Gottin Dhidnadund Verwandtes (Uppsala, 1917), pp. 47-48. 96 For the etymology of this word now see also Thieme, Indo-Iranian Journal, III (The Hague, 1959), 150. 97 L. Renou, "Les hymnes aux Visvedevah," Etudes vediqueset pdnindennes, IV (Paris, 1958), 46. 98P. Thieme, "Studien zur indogermanischenWortkunde und Religionsgeschichte," Akad. d. Wiss. Leipzig, Phil.-hist. Kl., XCVIII, No. 5 (1952), 55 ff. 99V. Machek, "Name und Herkunft des Gottes XII Indra,"ArchivOrientdlni, (Prague, 1941), 143 ff. 00I for one am not convinced by the author's argumentation. 263
or less complicated descriptions. Nor should we expect to find welldefined concepts or minutely circumscribed fields of action and influence'05without partial overlaps or vagueness of contours. Neither the Vedic poets nor the Aryan community, the popular beliefs and ideas of which they developed and tried to systematize,106had been submitted to philosophical training in a modern sense of the term. There is no good reason to take for granted that their ideas, concepts, and terminology were characterized by the precision and unambiguousness which are the goal of post-Socratic scientific argumentation.
Like the weltanschauliche terms of other peoples-Greek
61K7, 8cLus,
etc.-these concepts gradually developed, growing, enriching vo6uos, their contents and expanding the range of their applicability. They were symbols for Bewusstseinsinhalte, which were-in accordance with the experience, the views, convictions, and interpretations of those who attempted to penetrate into the ideas for which they stood and to speculate about their nature and relations-deepened and extended by a continual process of assimilation, association, identification, differentiation, and amplification.107In principle this process must have taken place like any process of semantic change, that is, either the "name" glides over to the "sense" of a satellitic idea or the "sense" glides over to the "name" of a closely associated idea.'08The direction of the cumulative results of the endless series of minor changes and semantic expansions'09 was no doubt largely influenced by the associations which prevailed in the minds of those who used these terms, by the sphere of their interests and their favorite trends of thought,"0 which can neither be reconstructed by means of the categories of traditional logico-rhetorical European semantics nor by reference to the phraseology and lines of thought of modern European poets.
106 The reader may for the sake of brevity be referred to my Die Religionen Indiens, I (Stuttgart, 1960), pp. 48 ff. 106 It may be remembered that, for instance, the ideas voiced with regard to the gods, etc., by the Homeric characters were considerably more vague and indefinite than those pronounced by the poet himself (E. Ehnmark, The Idea of Godin Homer [Uppsala, 1935], p. 102). 107 See, e.g., J. M. van Gelder, Der Atman in der Grossen-Wald-Geheimlehre (The Hague, 1957), p. 10; H. Vos, "OLAs"(thesis Utrecht, 1956), p. 29; and my Inleiding tot het Indische denken (Antwerp, 1948), pp. 9 ff., 23 ff. 108 Ullmann, Principles, pp. 216 ff.; L. Roudet, "Sur la classificationpsycholoXVIII (1921), 676 ff. gique des changementss6mantiques,"Journal depsychologie, 109Moreover: "Every word is a heritage from the past, and has derived its meaning from application to a countless number of particulars differing among themselves either much or little" (A. H. Gardiner, The Theory of Speech and Language[Oxford, 1932; 2d ed., 1951], p. 35).
110
265
described as "greatness" or "majesty," it also implies what we would call "distinction, importance, eminence in power, genius, or ability, possession of high qualities, superiority to the common human conditions of life, etc.," and "honor, reverence, homage to superiors, worship, adoration" occurring also to denote actions or occurrences generating this "greatness," such as worship, festivals, and sacrificial acts. Vague impressions and ideas, largely determined by emotions or aspirations, intuition, or speculation; views of events, phenomena, connections, backgrounds, causality; traditions and experience-all take the shape of more or less definite ideas, expressed by terms which are nowhere scientifically defined. Being symbols for the essentially incomprehensible aspects and factors of all important events in nature, society, and individual life, the investigations of their meaning were, however, for the ancients of the highest importance, because knowledge of the names meant control over the powers to which these referred. Hence also were the identifications, associations, and other terminological experiments of the poets and "philosophers" who attempted to penetrate into the mysteries behind fact and reality and to define the undefinable. And here is another source of difficulties for those who try to establish the semantically dominant elements. A study of the much discussed term brahmanll7led me to similar conclusions which, however, have been misunderstood by one of the reviewers of my publication.118Although I purposely refrained from any attempt at "translating" this name, Thieme believed me to regard it as an equivalent of our "power," and Mayrhofer1 still more incorrectly informs the readers of his etymological handbook that in my view brahman "urspriinglich 'Lebenskraft, Mana' gewesen sein soll." I would for the benefit of my superficial readers recall to memory that, while intending "to follow up the inquiry on problems which may be related to the riddle and to go on... bringing to the fore such aspects of the question as seem not to have attracted sufficient attention,"120I criticized the main views upheld by my predecessors, emphasizing the weakness of evolutionistic constructions and the difficulty of arranging the senses of ancient Vedic terms of outstanding importance, like brahman,in such a manner that a definite historical
117 118
Notes on Brahman (Utrecht, 1950). Thieme, "Brthman," Zeitschrift der deutschen morgenldndischen Gesellschaft,
454. The same author-who rightly rejects the above interpretationof brdhmanhad, in a review of my book (published in Anthropos,XLVII [1952], 319 ff.), not only adopted this "originalmeaning" but also enthusiastically subscribedto the etymological connection of the term with "brh-kraftigen, starken."
120 Notes
on Brahman, p. 3.
267
Ibid., p. 4.
It is my intention to return to some passages in Mayrhofer'slong discussion of brahma (op. cit., pp. 452-56) in another paper. I wish to emphasize that I am by no means an adversary of a sound historical method; we should, however, be aware of its limitations. 123 "En pr6sence de morphemesidentiques pourvus de sens differents, on doit se demander s'il existe un emploi of ces deux sens recouvrent leur unit6," E. Benveniste, "Problemess6mantiquesde la reconstruction,"in Word,X (New York, 1954), 251. 124 For some critical remarks on the etymology proposed by W. B. Henning (in Transactions of the Philological Society, 1944 [London, 1945], pp. 108 ff.) and adopted by Mayrhofer (loc. cit.), and as far as the formal side is concerned not combatted by the present author, see Notes on Brahman, pp. 69-70 (not mentioned by Mayrhofer).
126 Notes 126
127
268
less definite power" was on p. 70 specified as the "idea of 'inherent firmness,' supporting or fundamental principle." We should not, however, throw out the baby with the bath water. Even if brahman does not from the genetic point of view derive from the root brh-128 the agelong association of both words-that is to say also of their "senses"-in the heads, speculations, and weltanschauliche theories of the Indians129is of special interest and more worth studying than it is supposed to be by Thieme.130"It is quite possible that the features of a language... by means of which we link it to others in a stock or family are among the least important when we seek to connect it to the rest of the culture.""13 And, it may be added, of "popular etymology" may prove very often a successful case to be a source of welcome information of the important question as to how either traditionally or in a definite period, the Indians themselves thought about the basic, central or "original" sense of a "key word." The so-called popular etymology is an a posteriori motivation of a word revealing the associations into which it has entered. Those cases of this phenomenon which repeatedly occur in many texts may be regarded as reflecting more or less fixed opinions and convictions of the authors and the communities of which they form part and shed a peculiar light on their ways of interpreting There can be no doubt whatever nature, life, and spiritual world.132 that for the Indians brahman,which already in the R.gvedarepeatedly that is, "something that causes to increase, appears as a vardhanam,133 and strengthens, animates, grants prosperity" was to be connected with brh-,notwithstanding the possibility that this association was an "a posteriori etymology" and that this "popular etymology" may have contributed to a change in the meaning of the word.134 In the earliest texts in which it occurs, those of the R.gvedasamhita, which are the ancient products of Indian literature and Indian
128 Cases are, however, not wanting in which scholars while rejecting a "scientific" etymology which has been accepted for many decades return to the interpretation of the Ancients: see, e.g., P. Chantraine, in Festschrift-A. Debrunner (Bern, 1954), pp. 85 ff., on Gr. &-yos, "any matter of religious awe." 129 For a succinct survey see my Notes on Brahman,p. 18. 130 derdeutschen Thieme, "Brahman,"Zeitschrift morgenldndischen Gesellschaft,
CII, 95 f.
131
Hoijer, in Anthropology Today, p. 567. also my paper on the etymologies in the ancient Indian Brahmanas, in Lingua, V (Amsterdam,1955), 61 ff.,n. 54, and p. 83 containingsome remarkson the brahman controversy. 133 Notes on Brahman,p. 40; see also J. Charpentier,Brahman (Uppsala, 1932), pp. 4 and 85, n. 4. 134For "popular etymology" see, e.g., E. H. Sturtevant, Linguistic Change (New York, 1942), pp. 94 ff. 269
132See
of the Aryans intermarried or who had found a place on the fringes of Aryan society. Between the Indo-European or Indo-Iranian words, the existence of which may be hypothetically assumed, and their later forms which appear in the Vedic texts-the poetic diction of which was in prehistoric times evolved by authors whose work has been lost forever-is the usage of those who had in the Rgvedic period no access to literature but who may have influenced thought and vocabulary of the other bodies of ancient literature. Great motifs and symbols in religion and important thoughts in Weltanschauung are, even in one and the same period, different things to different men. It is therefore highly improbable that there has ever been a moment at which brahman only and exactly meant "formula" or "verse" or "sacred word." I cannot subscribe to the view formulated by Thieme'35 that we must attempt to find out the formal features of words and those traits of usage which are common to all the contexts in which it appears by linguistic procedures of analysis which are "quite independent of our views as to the religious and other ideas expressed by the text." It is in my opinion a mistaken belief that "the abstract content" of words such as rta, aramati, which stands for something like der rechteSinn, die gemdszeGesinnung, or puramdhi die Wunscherfiillung "is without relation to a possibly peculiar psychology of the Rigvedic poet." In principle, M. Bloomfieldl36 was no doubt right that "in the interpretation of a term that figures prominently in the mystichieratic sphere of the Veda [that is, Thieme'37 rightly adds: one of the "termes essentiels du R.V."138] it is peculiarly necessary to search for its uses outside that sphere." The difficulty, however, often is that the plain "prose central meaning" is not likely to appear frequently, or that we are not able to make out when a word is not enveloped in what Bloomfield'39 called "the Vedic haze," many words being always steeped in Weltanschauung and any reference to late Vedic or post-Vedic uses in "profane" texts being, of course, liable to introduce anachronisms. And even in those cases-which may be less in number than some Vedic scholars are nowadays inclined
135 P. Thieme, in a review of L. Renou, ttudes vediqueset pdnineennes, I, in Journal American OrientalSociety, LXXVII (New Haven, 1957), 51 ff., esp. p. 56. 136 M. Bloomfield, "The Vedic Word VidAtha," Journal American Oriental Society, XIX, 13 f. 137 Thieme, loc. cit., p. 54. 138 See Renou, op. cit., I, 22. 139 M. Bloomfield, Review of W. Neisser Zum Worterbuch des IRgveda,
271
branches of learning.l42These disciplines are not to supply deficiencies of our texts, or to replace facts which, though badly needed for the sake of an air-tight argument, are lacking in our sources. Resorting to them does not imply that the religion of Vedic man was in all or some respects practically the same as that of the ancient Germans or Babylonians or of present-day Eskimos or Polynesians or that the mental equipment of Vedic man was distinct from that of civilized man; neither does it express the conviction of the author that "he himself knows all about Vedic religion before consulting the texts." These sciences, for instance ethnology and phenomenology of religion, may offer us general notions and a wealth of information about special points and about features which are likely to occur in an archaic culture, about their backgrounds and interrelations, and this information can provide us with heuristic and illustrative principles for the study of Vedic religion. A knowledge of the types of religious communities may help us in understanding the social factors which have played a part in the formation of the same, an insight into the nature of myths and rites in general enables us to penetrate into the meaning of the mythico-ritual pattern of the ancient Indian culture. These disciplines may open our eyes to the characteristics of the culture toward an understanding of which we direct our efforts.'43A comparative study of the literary forms of the archaic religious poetry of other peoples is of service to those who desire to investigate the literary and linguistic structure of the Vedic hymns and the prose of the Brahmanas. But just as a comparative examination of "poetic devices" enables us to distinguish between their function in archaic literature and that in the works of modern poets and preserves us from viewing the Veda in the light of the art of Schiller and Goethe,144 so also may other disciplines make our minds alive to the possibilities and characteristics of archaic culture in general and to those of a special archaic culture in particular.
142 "Pour les societes qui ont, depuis plus ou moins longtemps, une litt6rature ou, du moins, des documents ecrits, l'6tude de l'histoire religieuse n'est qu'un cas particulier de l'histoire de la civilisation, ou de l'histoire tout court, et, dans la critique comme dans la construction, n'emploie pas d'autres procedes," G. Dumezil, in M. Eliade, Traite d'histoire des religions (Paris, 1949), p. 6. 143 It may, of course, be readily admitted that those who discuss the essence of religious phenomena in general could always derive greater advantage from a thorough knowledge of ancient Indian religion than they usually seem to care for. 144 As seems to be recommended by Thieme, Mitra and Aryaman (New Haven, 1957), p. 22, and in Review of J. Gonda, Some Observations on the Relations between "Gods" and "Powers" in the Veda, Indo-Iranian Journal, II, 233, whose views were already criticized by Kuiper, in Review of P. Thieme, Mitra and Aryaman, Indo-Iranian Journal, III, 211 ff. 273