English Translation


The dictionary, which I hereby hand over to the public, is the result of my own needs. I saw that I couldn’t reach an understanding of the Vedas by writing down a translation without compiling a complete as possible glossary of the Rig-Veda, the main part and basis of the Vedas. The use of the compiled material added so much value to my linguistic – and in particular comparative – work, that I decided to process the material and make the material, that became such a help for me, generally accessible. I was well over halfway through the process of compiling and editing the dictionary, when I learned that Aufrecht also intended to publish such a dictionary. Aufrecht is an excellent linguist who, like very few others, has access to the entire literature commenting and explicating the Rig-Veda. Since I deem no one more qualified to undertake such an opus than Aufrecht, I decided to continue compiling and editing for my own benefit, but to make the publication dependent on whether Aufrecht’s plan would be realized or not. When the work was ready for printing, I learned through well-known scholars, to whom the newer literature in this area is most accurately known and who through their personal relationships with linguists working in England are better than others equipped to judge, whether the publication of such a dictionary will be realized in the near future, that there was nothing of the kind in preparation. I was asked by them to go forward with the publication of my dictionary. On these grounds I was convinced that I could not postpone the publication of the dictionary any longer.

Naturally, the Petersburger Wörterbuch constitutes the basis of my work. The Petersburger Wörterbuch initiated a new epoch of Sanksrit phililogy and in particular the appreciation of the Vedas. I consulted the information published therin and especially Roth’s groundbreaking work on the Vedas. Any deviation from the view expressed by Böhtlingk and Roth was only admitted after careful examination and after comparison of all concerned text passages allowed, as such an excellent work deserves. Specifically, I went back to the previous practice and avoided positing stems (roots) of inflecting words that do not occur in the language itself. Thus I posited that form of a declinable word as its stem form which it actually takes as the first part of a compound (apart from phonetic transformations); i.e. pitṙ, not pitar, bṙhat and not bṙhant, etc; furthermore, for verb forms which they posit as deverbal nouns (verbalia), I take verbal noun (verbale) as that declinable form of a verb, which either does not receive a suffix or only a “t” after a short vowel, that is vṙdh, not vardh because of the verbal noun vṙdh [cf. ṙtāvṙdh etc.], bhṙ, not bhar because of the verbal noune bhṙt, gir, not gar or gṝ because of the verbal gir, hu, not hvā or hve because of the verbal noun hû [deva-hû etc.]. Where two different verbal nouns (verbalia) are present, such as hvŕ-t and hrú-t, two root forms hvṙ and hru have to be posited. For every word, all forms occurring in the RV and for all forms, all text locations, where they occur, are given. Only for some very common forms and uninflectible words text locations are only given up to a certain song, unless later locations are of particular interest. The decision has been made explicit in the respective parts of the dictionary. The Sāmaveda (SV.), Atharva-veda (AV.), and Vājasaneyi­-Samhita (VS.) were only referenced where they were deemed necessary to determine form or meaning. Form of declinable words are given in the same order as in Panini: V. (Voc.), N. (Nom.), A. (Acc.), I. (Instr.), D. (Dat.), Ab. (Abl.), G. (Gen.), L. (Loc.), first the singular [s.], then the dual [d., du.] and the plural [p., pl.]. In respect to gender forms, N. and A. of the neutr. [n.] is always placed under the A. of masc. [m.] and the fem. [f.], where it has a separate form is placed after all forms of the other two genders. Each of these forms is given starting with the last vowel of the stem, and all preceding material is replaced by a preposed hyphen, e.g. for the stem aṅçumát, the form aṅçumátīm is given as -átīm. Only if two different stems (e.g. ábibhīvas, weak ábibhyus) are given, forms are given from the point where the two stems differ (e.g. -yuṣā for ábibhyuṣā). In entries of verbal forms, all finite forms precede the non-finite forms, and in particular the forms of the base verb precede those of the passives, causatives, intensives, and desideratives. Among these, the forms derived from the present tense stem (simply called stem) are placed at the beginning. The bare present tense stem or its Verstärkung [FR: meaning not clear to me, guna and vriddhi forms are sometimes called Verstärkung, literally: “its strengthening / intensification / enhancement / reinforcement”] is placed in front of these as a heading. After this the individual word forms in abbreviated form follow, similar to entries of nouns. Among these forms, all active voice forms precede the middle voice forms, the modal categories are given in the order ind., conj., opt., imperative; where there are multiple conjunctive forms [FR: subjunctive?], the present tense form will precede the imperfective form. In all moods, the finite forms are given in the familiar order. Following the present tense stems and under a new heading, the imperfect forms derived from the same stem are given, as long as the augment is preserved. Imperfect forms without the augment coincide with the second conjunctive [FR: subjunctive?] and are given under conjunctives. The perfect forms follow in the same manner, then the (rare) plusquamperfect and future forms (ending in -iṣyāmi), followed by the aorist. The non-finite forms start with the participles belonging to the different tenses, followed by the participles that do not belong to a particular tense and are not derived from tense forms. Then in no particular order, follow the participles ending in -ta or -na, which I call for reasons of brevity Part. II., then the participles ending in -tṙ as Part. III., and following that the participles -tva, -ya (-enya etc.) as Part. IV. The participles have been given under these names, because the usual names for them are completely useless. These are followed by the absolutives (-ya, -tvā etc.) and the infinitives, finally the verbale (verbal noun) is given (see above). In the citations, the corresponding noun is given for every adjective, for every genitive the noun or verb on which it depends is also given, for every verb its inflection is reported and for the nouns the particularly characteristic adjectives are added. What precedes the numerical reference refers to all immediately following passages. However, a word is placed in parentheses, if it does not occur in the cited form in the next instance. The material following the quotation relates only to these passages. In case an instance is given with its context, the character — replaces the reference word form. The text edition by Aufrecht forms the base of the dictionary. However, the corrections by M. Müller have been consistently incorporated. I do diverge from Aufrecht’s transcription where he represents a single sound by two letters, as these relations can be highly confusing in a dictionary. For this end, I write ṙ instead of ṛi, r̄ instead of ṙî, ṣ instead of sh, ē instead of ai, and ō instead of au. These last two spellings cannot give rise to confusion. For comparative works, these two as well as the characters e and o have to be avoided and e=ai, o=au, ē=āi, ō=āu should be used. The composite spelling has been preserved for aspirated stops. This is admissible because they are placed in the same position in the lexicographic order, regardless of whether they are treated as a single or two letters. I also preserved the compound sign ḷi, as it only occurs in the root kalp. In the representation of the accents, I deviate (from Aufrecht), in that I indicate the accent-less long vowel by a macron and the accented long vowel by a roof (^) – so ā instead of â, â instead of ấ – and in that I represent svarita by an accent on the preceding semi-vowel (y, v), e.g. asmadrýac instead of asmadryàc. Where semi-vowels are pronounced as vowels, I write them as such: there is no word mártya, ámartya in the RV, but only mártia, ámartia, and I could not bring myself to include those non-forms into the dictionary. However, for practical reasons, I added them in parentheses and posited them for the lexicographic order. Likewise, I have removed the phonological alterations between words (sandhi), which is particularly helpful for lexical transparency. Where the original text has sandhi beween vowels, I placed a ˘ between the vowels. How far the written representation deviates from the sandhis as required by the meter is particulary well illustrated by the interaction of a or ā with an ṙ of a following word. In the traditional text these two vowels are always separated. On the metrical side, they are only separated when -a, -ā correspond to -as, -e, -ās, -ē, or -ār, -ān (in the obsolete nominative form mātâr, hótār, víbhvān, which have to be posited in 399,6; 127,10; 329,3; 332,6; 564,3 and which correspond to Greek μήτηρ etc.) and where the complete orthographic form has to be restored or when ṙ is followed by a double consonant (ṙtvíya 275,2; ṙṣṭí 167,3; 169,3; 648,5) so that here an accumulation of three consonants is avoided or when the adjoining vowels belong to two metrically distinct lines (where sandhi metrically never occurs, yet where it always occurs in prose, as soon as the hyphen [FR: literally “sign of separation”] is missing) or finally when the joining vowels are separated by an incision of the verse. The latter occurs rarely (five times): 319,7; 357,9; 202,12; 906,7; 956,6. In all other cases a, ā with following ṙ result in ar. (The case of 925,2, where the sandhi does not occur, the two instances 688,4 und 913,15, where it does occur against the rules, are based on erroneous reading.) From these phenomena, we have to draw the conclusion that the suffixes -as, -e (= a+i), -ās, -ē (= ā+i) had not yet lost their final consonant (s, i or y) before ṙ and probably also not before any other vowel. This justifies the spelling I employed even further. This spelling can never give rise to any confusion as a comparison with the traditional texts, which justifiably form the base for the editions, remains always directly possible.

The etymology could not be missing, as it is often central to determining the meaning of a lemma. However, I kept the etymology as short as possible by referencing Curtius, Grundzüge der griechischen Etymologie (Cu.), Fick, Vergleichendes Wörterbuch der indogermanischen Sprachen 1870 (Fi.) , Kuhn, Zeitschrift für vergleichende Sprachforschung (Ku.), Böhtlingk und Roth, Sanskrit-Wörterbuch (BR.), Benfey, Glossar zum Sāma-Veda (Be. SV. gl.), as well as occassionally referencing other works. Compounds are indicated by a hyphen seperating the parts of the compound.

I have derived the meaning, where it seemed necessary, in its context from reconstructable basic meaning, but then simply by consecutive numbers for meaning and usage in a manner that seemed the most useful for the understanding of the cited passages, put in sequence and related to the referenced passeages, so that it becomes apparent which meaning or usage I attribute to the word in each referenced passage. Here, subjective understanding plays a crucial role and later work will find many reasons for corrections. In particular, since I distinguished meanings or usages that are only separated by soft shades of meaning. This was done to make my work as usable as possible.

The publishing house in Leipzig has acquired the services of two capable young correctors, for the extremely time-consuming work of proofreading. Accorinding to the nature of the matter, the main part of the work remains with me. I can assert with certainty that at least up to now all passages are referrenced correctly and that this stands to be the case in the future, as far as it is in my power. Thus, the remaining printing errors (which, with all due care, are unavoidable) will be easily controlled.

Since I reference the songs by consecutive numbers (as done by Aufrecht), so I may indicate a concordance between these consecutive numbers and the reference system that numbers the ten books of the RV separately:

1-191 = 1.1-1.191,
192-244 = 2,1-2,43
235-296 = 3.1- 3.62,
297–354 = 4.1-4.58,
355–441 = 5.1–5.87,
442–516 = 6.1-6.75,

517–620 = 7.1-7.104,
621–712 = 8,1–8,92,
713–826 = 9.1-9.114,
827–1017 = 10.1 to 10.191,
1018–1028 = Vālakhilya 1-11.

Stettin/Szczecin 10th August 1872.

The author.

English translation courtesy of Dr. Felix Rau, University of Cologne, 2018.