Of the entire corpus of the Avesta, the Gathas have been translated far more frequently than any of its other divisions. The reason for their popularity is the challenge they pose to any Iranist who has been drawn into their study. They are formidable in their linguistic and ideological obscurity, yet seductive in the invitation they hold out to enter the conceptual world of one of the world’s greatest prophets. With the exception of Jean Kellens and Eric Pirart, translators have assumed that Zoroaster was the Gathas’ author, with a consensus that the last in the collection, Yasna 53, was not his composition, but issued from his circle of followers. The present review is restricted to complete translations and editions of the Gathas.

A translation of any text, modern or ancient, is already an interpretation, yet, probably nowhere in the world’s ancient literatures is this truism more apparent than in the translations of the Gathas. All the 20th century translations discussed here were produced by, at the least, competent scholars whose interpretations were informed by solid philology. Nevertheless, a comparison of various translations will reveal a variety of renderings of the underlying text whose divergence one from the other can be astonishing. Even within the same work multiple options may be entertained. Thus, for example, in his commentary to Yasna 28.1, Helmut Humbach (II, 1991) offered, beside his favored translation, no less than three alternatives. In a review of Mrs. Henning’s English translation of Jacques Duchesne-Guillemin’s French translation, Ilya Gershevitch (JRAS, 1952, p. 174), with typical humor, wrote: “Anybody can now see what the author thinks Zoroaster said. To expect more would not be fair, since out of the 238 surviving Gathic stanzas scarcely less than 190 are partly or completely incomprehensible. For each of these the author has selected one of the numerous guesses proposed by his predecessors, sometimes preferring a guess of his own.” Whether or not Gershevitch’s remarks are altogether accurate, they have a ring of truth, even if one accepts that this number will have been somewhat reduced through the following half-century of research.

The obscurities of the underlying text are often simply transposed into the translations. That is, even when one has what one believes to be an impeccable translation, that translation may not convey any obvious meaning. In the preface to his translation, Christian Bartholomae proposed to regard the Gathas as verse sermons which served as cryptic encapsulations of doctrine. Easy to remember, they would have been accompanied by extensive prose explanations which would have fully amplified the terse verse statements. Such prose commentaries could be imagined to have been inserted between verses, or, what Bartholomae regarded as more likely, placed at the end of a prose discourse. Whatever the contextual setting of the Gathas may have been in Zoroaster’s day, most modern translators have thought it necessary to provide prose synopses of whole hymns and/or of individual stanzas. In many cases the translations form part of larger studies dealing with philology and religio-historical interpretation. Such commentaries are often indispensable aids to understanding what translators think a Gathic passage ultimately means, as a translation standing alone will not always suffice to convey their intention.

A review of translations of a well-known passage will serve to illustrate some of the problems one can expect to encounter. Yasna 30.3 speaks of the twin spirits who in this Gatha choose between good and evil. The first two lines of the stanza read:


aṱ tā mainyū paouruyēyā yə̄mā xᵛafnā asrwātəm

manahicā wacahicāšyaoθanōi hī wahyō akəmcā


Bartholomae (Smith, Hinz): “The two Spirits in the beginning, who revealed themselves in a dream-vision as twins (are) the Better and the Evil in thought, word and deed” (Die beiden Geister zu Anfang, die sich durch ein Traumgesicht als Zwillingspaar offenbarten, [sind] das Bessere und das Böse in Gedanken, Wort und Tat”).

Lommel: “In these two first Spirits, which were perceived through a dream as twins, are, to be sure, in thought, speech, and action the Better and the Evil” (Under diese beiden ersten Geister, welche als Zwillinge durch einen Traum vernommen wurden, sind ja im Denken, Reden und Handeln das Bessere und das Schlechte).

Nyberg: “The two primordial Spirits, the twin Sleep [and his brother], were, as it has been transmitted, the Better and the Evil in mind, word and action” (Die beiden Ur-Mainyu, die Zwillinge Schlaf [und sein Bruder], waren, wie überliefert worden ist, das Bessere und das Schlechte in Sinn, Wort und Handeln).

Duchesne-Guillemin: “Now, in the beginning, the two spirits who are known (…) as twins, are, the one, the better, the evil in thought, speech and action” (Or, à l’origine, les deux esprits qui sont connus […] comme jumeaux, sont, l’un, le mieux, l’autre, le mal en pensée, parole, action).

Insler: “Yes, there are two fundamental spirits, twins which are renown to be in conflict. In thought and in word, in action they are two: the good and the bad.”

Humbach (1959): “These are both the fundamental efforts, the twins, who are recognized as two sorts of thoughts and two sorts of words and two sorts of action, the better and the evil” (Dies sind die beiden grundlegenden Bestrebungen, die Zwillinge, die als beiderlei Träume bekanntgeworden sind, als beiderlei Gendanken und beiderlei Worte, als beiderlei Werke, das bessere und das schechte).

Humbach (1991): “These (are) the two spirits (present) in the primal (stage of one’s existence), twins who have become famed (manifesting themselves as) the two (kinds) of dreams, the two (kinds of) thoughts and words, (and) the two (kinds of) actions, the better and the evil.”

Kellens and Pirart: “(I shall mention also) the two fundumental states of mind which are known as being twin dreams in regard to thought and speech. In regard to (ritual) action, they are the better (action and the bad (action)” ([Je vais dire aussi] les deux états d’esprit fondamentaux qui sont connus pour être des songes jumeaux lors de la pensée et de la parole. Lors de l’acte [rituel], ce sont le meilleur [acte] et le mauvais [acte]).

The greatest problems are posed by the words paouruyē and xᵛafnā. While a general sense of the twin spirits comes through in almost all translations, one would hope for more common consensus on just what this verse, so central to an understanding of Zoroaster’s dualistic theology, really means. Opinions differ over whether the mainyū are two primordial spirits, impulses or states of mind, or whether the activities they are involved in took place “at the beginning” or takes place in the ever present. Are they themselves “two (twin) dreams” or are they revealed (to whom?) “by means of a dream.” And how are we to understand “dream”: simply as a “dream” or a “vision” or, as Henrik Samuel Nyberg elaborated in his commentary, a shamanistic “trance” personified? Only one translator, Duchesne -Guillemin,had the humility to leave the word untranslated.

In spite of the uncertainty, often severe, which attends the interpretation of so many Gathic stanzas, it is a wonder that translators are so loathe to signal to those of their readers unfamiliar with Avestan, when the translation of a passage rests on pure conjecture or is the best guess among options. Related to this shortcoming is the tendency for each translator to appear, at least overtly, to be working in a scholarly vacuum.

There are two Gatha translations which, although they predate Bartholomae’s dictionary, must be mentioned, namely those of James Darmesteter and of Lawrence H. Mills. In fact, Mills published two translations. The first to appear was in the Sacred Books of the East (1887) and is still readily available today, while the second is contained in his monumental A Study of the Five Zarathushtrian (Zoroastrian) Gâthâs (1894). In the latter, generally ignored in 20th-century studies of the Gathas, Mills produced beside the Avestan text in both original script and roman transliteration, a literal translation into Latin and a free translation in English verse, the transliterated Pahlavi gloss with its English translation, the transliterated Sanskrit gloss of Neryosangh with its English translation, a Parsi-Persian translation in transliteration (not translated), and accompanying all this over two-hundred pages of notes. Its utility remains in its convenient arrangement of the original text with the Pahlavi and Sanskrit glosses.

In Gathic studies the name of James Darmesteter is always associated with his unusual, and universally rejected, ideas about Zoroaster. Often neglected, however, is his defense of the Zoroastrian tradition itself as a repository of the valid understanding of the Gathas, against the Vedicists employing the methods of comparative linguistics. Darmesteter argued forcefully that the rejection of the Pahlavi and Parsi traditions was based on an inadequate appreciation of the accuracy of the Pahlavi translation in terms of its glosses of Gathic vocabulary and of the larger tradition in terms of its fidelity to the ideas expressed in the Gathas. For Darmesteter, as for Mills, the Zoroastrian tradition could alone serve as a sure guide, since the comparative method would only lead the way to unverifiable speculation. The entire history of Gathic scholarship in the 20th century has tended to vindicate the comparative method at the expense of the Zoroastrian tradition’s authority. It is worth noting that, in her writings on Zoroastrianism, Mary Boyce has been a consistent advocate of the primacy of the Zoroastrian tradition in understanding almost all matters pertaining to Zoroaster, although she has not published a translation of the Gathas. Also noteworthy is the work in progress of a new edition and translation of the Pahlavi Yasna by Humbach and Pallan Ichaporia.

Although one can find useful insights, especially into the relationship of the later traditions to the Avestan text, the works of both Mills and Darmesteter on the Gathas were almost immediately rendered obsolete through the publication of Bartholomae’s Wörterbuch (1904) and his Die Gatha’s des Awesta (1905), suggestively subtitled “Zarathustras Verspredigten übersetzt.” For nearly a century the Wörterbuch has endured as the reference point for all serious philological work on the Gathas, as well as the Avesta generally. In it, Mills is mostly ignored, while Darmesteter’s ideas are frequently noted and almost as frequently dismissed with the recurring phrase “falsch Dst.” As one might expect, Bartholomae’s translation is based on the interpretations given in his dictionary. With its extensive notes referring to the dictionary, the translation together with the dictionary can be read as companion volumes, much like Fritz Wolff’s translation of the Avesta(1910).

Bartholomae’s Wörterbuch and Gatha’s directly inspired two English translations. The first was by James Hope Moulton, appearing as an appendix to his Early Zoroastrianism. It is actually a translation (“en une langue plus claire”; Duchesne-Guillemin, p. 13) of Bartholomae’s translation, in which Moulton included changes and improvements based primarily on Karl Friedrich Geldner’s ideas expressed in the Grundriss der iranischen Philologie and in Alfred Bertholet’s Religionsgeschichtliches Lesebuch, but also according to his own lights. The second translation formed part of an immense, lifelong study of the Gathas by Irach Jahangir Sohrabji Taraporewala (1951). Taraporewala’s work stands as an important bridge between the scientific goals of European scholarship and the needs of the community of Zoroastrians, particularly of Parsis. By providing both a literal translation keyed to the Wörterbuch and to other standard reference works such as Grassmann (Wörterbuch zum Rig-Veda), A. A. Macdonell (Vedic Grammar), and A. V. Williams Jackson (Avesta Grammar), and a free translation capturing “the thought contained in the verse according to my own bent,” Taraporewala bequeathed to posterity an eminently useful tool for students taking their first steps in the grammatical and lexical mysteries of the Gathas, no less than an inspiring portrayal of the Prophet’s thought for his co religionists. In this connection it is worth mentioning the 1929 doctoral dissertation of Maria Wilkins Smith, Studies in the Syntax of the Gathas, a work which contributed to the understanding of some problems of syntax, but which did not advance much beyond Bartholomae.

It is somewhat difficult to place Herman Lommel’s Die Gathas des Zarathustra, published posthumously in 1971, within the history of Gatha translations, as this final version is the product of study spanning nearly half a century. Begun as an extension of the partial translations by Friedrich Carl Andreas (q.v.) and Jacob Wackernagel, Lommel’s work continued until his death in 1968. Although this final publication of his complete translation is provided with notes and prose summary of all the hymns, it is most profitably read with his Die Religion Zarathustras, which remains a standard work in the field.

Two translations may be characterized as being informed by a history of religions approach. The first to appear was that of Henrik Samuel Nyberg in his Die Religion des alten Iran, in which he sought to demonstrate that Zoroaster was to be understood as belonging to the north Eurasian religious complex known as shamanism. Though not collected in one place, the book, nevertheless, contains a nearly complete inventory of translations for which a fairly thorough shamanistic interpretation is zealously promoted by the author. The second was Jacques Duchesne-Guillemin’s Zoroastre: Étude critique (1948). Written in clear French, the translations are situated in their religious context through a lengthy introduction and each accompanied by a prose summary. Duchesne-Guillemin’s approach was heavily influenced by George Dumézil, especially his Naissance d’archanges (1945), and his theories of an Indo-European tripartite ideology, though the latter’s ideas do not intrude overtly in the translation itself. Duchesne broke with the traditional order of the Gathas by arranging them in what he contended was their compositional order.

The first truly innovative departure from Bartholomae from a philological perspective was Helmut Humbach’s Die Gathas des Zarathustra (1959). Building on a long series of articles, Humbach devoted much of his introduction to text-critical and sylistic studies which provided the basis for an extensive corrective to Bartholomae’s Wörterbuch. The commentary also offered many new interpretations of lexical items. All of this contributed to a fresh translation which, perhaps to an even greater degree than previous German translations, is composed in a dense style that often does not move beyond the obscurity of the original Avestan. In order to lead the reader to Zoroaster’s intended meaning, Humbach supplied a stanza by stanza descriptive paraphrase. Frequently astounding in its departure from what appears to be the surface meaning, the paraphrase is dominated by the idea that the prophet was concerned almost exclusively with ritual in his professional role as priest.

Humbach’s translation was followed in 1961 by Walter Hinz’s Zarathustra. Like Duchesne-Guillemin’s translation, Hinz’s comes at the end of a monograph on Zoroaster and his religious background. Since Hinz places Zoroaster’s death on 1 May 553 B.C.E., he is drawn to the Near East, rather than to the Vedas, for analogies to understand the Gathas. Especially prominent are the Israelite prophets and Moḥammad. One should observe that the extensive philological notes to the translations are scrupulous in their citations of other scholars’ work. Although Hinz cites Humbach’s work in the notes, it was not until 1975 that Stanley Insler contested the extreme ritualistic interpretation given by Humbach. In his The Gāthās of Zarathustra, Insler used his introductions to each of the hymns, among other matters, to emphasize the moral and ethical character of Zoroaster’s thought neglected in the ritualist approach. While it is true that all modern translators have made use of comparative materials from the Vedas, Insler, an outstanding Vedacist himself, probed the limits of what the ancient Indian texts can contribute.

A significant contribution of Humbach’s Die Gathas was the reproduction of an improved text over both Geldner and Bartholomae, and further improvements were made by Insler. Both scholars, though, produced editions of the Vulgate text. For an appreciation of the original poetic diction of Zoroaster, a hypothetical reconstruction would be a desideratum. In 1978, Maria C. Monna published her dissertation, The Gathas of Zarathustra,which placed on facing pages Humbach’s Vulgate edition and her own reconstructed text, though without translation.

A decade later, the first volume appeared of a three volume exhaustive study of the Gathas by Jean Kellens and Eric Pirart, Les textes vieil-avestiques. In many ways these volumes carry forward the work of Humbach, especially in the unrelenting insistence on the Gathas’ overriding concern for ritual, an insistence which goes well beyond Humbach’s 1958 translation and paraphrase. The translators also came to the remarkable conclusion that Zoroaster did not compose the Gathas at all, but rather that they are the collective expression of a total religious group rendered poetically by certain artisans under the surveillance of the group. Be that as it may, the three volumes contain extensive text, critical and grammatical notes, vocabulary, and commentary.

Finally, in 1991, Humbach, in collaboration with Josef Elfenbein and Prods Oktor Skjærvø, published a second, totally revised edition of his earlier work under the title The Gāthās of Zarathushtra, and the Other Old Avestan Texts, in two volumes. The new translations often differ substantially from the 1958 edition, and, in all cases, they gain immensely in clarity of expression. Although the introductory matter dealing with cultural and religious history is amply supported by references to the scientific literature, the philological discussions and notes almost completely ignore this literature.

In taking stock of all the translations produced over the span of this century, and, having himself struggled with these difficult texts for many years, the present writer must concur with the pessimistic view of Gershevitch cited above. These have been good translations produced by eminent scholars, who have provided a wealth of insights into these marvelous poetic creations. To be sure, progress has been made. Yet, as so much remains conjecture and selection among posssible options, it is not reasonable to expect any substantial advances in the next century unless a radically novel methodology is developed which can stand the test of scientific review. The fascinating project of Martin Schwartz tentatively put forth in the Humbach Festschrift (1986), in which the compositional structure of Zoroaster’s oral poetry is subjected to radically novel analyses, may provide the hermeneutical key. However, it will need to be tested systematically and comprehensively in the entire corpus of the Gathas, a massive project which Schwartz expects to complete within the first years of the twenty-first century.



Chr. Bartholomae, tr., Die Gatha’s des Awesta: Zarathustra’s Verspredigten, Strasbourg, 1905.

A. Bertholet, Religionsgeschichtliches Lesebuch, Tübingen, 1908.

J. Duchesne-Guillemin, Zoroastre: Étude critique avec une traduction commentée des Gâthâ , Paris, 1948; tr. M. Henning as The Hymns of Zarathustra, London, 1952, repr., London, 1963; reviewed by I. Gershevitch, JRAS, 1952, p. 174.

W. Hinz, Zarathustra, Stuttgart, 1961; reviewed by J. Duchesne-Guillemin, ZDMG 113, 1963, pp. 327-28.

H. Humbach Die Gathas des Zarathustra, 2 vols., Heidelberg, 1959; reviewed by H.-P. Schmidt, DZMG 111, 1961, pp. 211-22; by I. Gershevitch, BSOAS 25, 1962, pp. 367-70; and by O. Klíma, Archiv Orientální 29, 1961, pp. 136-41.

Idem, The Gathas of Zarathushtra and the Other Old Avestan Texts, 2 vols., Heidelberg, 1991 (with the collaboration of J. Elfenbein and P. O. Skjærvø); reviewed by Gh. Gnoli, East and West 42, 1992, pp. 519-27; by W. Sundermann, Kratylos 38, 1993, pp. 50-66.

I. Ichaporia, The Heritage of Zarathushtra, Heidelberg, 1994 (popular translation based on the preceding). S. Insler, The Gāthās of Zarathustra, Acta Iranica 8, Tehran and Liège, 1975; reviewed by S. Zimmer, OLZ 75, 1980, pp. 64-65.

J. Kellens and E. Pirart, Les textes vieil-avestiques, 3 vols., Wiesbaden, 1988-91; reviewed by N. Oettinger, Kraytylos 38, 1993, pp. 43-49; and by S. Zimmer, OLZ 85, 1990, pp. 586-91.

H. Lommel, Die Religion Zarathustras nach dem Awesta dargestellt, Tübingen, 1930; repr. Hildesheim and New York, 1971.

Idem, Die Gathas des Zarathustra, ed., B. Schlerath, Basel and Stuttgart, 1971.

L. H. Mills, A Study of the Five Zarathushtrian (Zoroastrian) Gâthâs, with Texts and Translations, Erlangen, 1894, repr. 1977.

M. C. Monna, The Gathas of Zarathustra: A Reconstruction of the Text, Amsterdam, 1978; reviewed by S. Zimmer, OLZ 76, 1981, pp. 335-36 (also incorporates review of F. B. J. Kyiper, On Zarathustra’s Language, Amsterdam, 1978).

J. H. Moulton, Early Zoroastrianism, London, 1913.

H. S. Nyberg, Die Religionen des alten Iran, Leipzig, 1938, repr., Leiden, 1966.

B. Schlerath, “Die Gathas des Zarathustra,” OLZ 57, 1962, pp. 565-89 (review of Gatha scholarship with special attention to translations of Humbach and Hinz).

M. Schwartz, “Coded Sound Patterns, Acrostics, and Anagrams in Zoroaster’s Oral Poetry,” in R. Schmitt and P. O. Skjærvø, eds., Studia grammatica: Festschrift für Helmut Humbach, Munich, 1986, pp. 327-92.

M. W. Smith, Studies in the Syntax of the Gathas of Zarathushtra together with Text, Translation, and Notes, Philadelphia, 1929; reviewed by A. V. William Jackson, Language 6, 1930, pp. 329-32.

I. J. S. Taraporewala, The Divine Songs of Zarathushtra: A Philological Study of the Gathas of Zarathushtra, Containing the Text with Literal Translation, a Free English Rendering and Full Critical and Grammatical Notes, Metrical Index and Glossary, Bombay, 1951; also pub. posthumously, without the copious notes, in The Religion of Zarathushtra, Bombay, ca. 1979.

F. Wolff, Avesta: Die heiligen Bücher der Parsen übersetzt auf der Grundlage von Chr. Bartholomae’s Altiranischem Wörterbuch, Strassburg, 1910, repr., Berlin, 1924 and 1961.

(William W. Malandra)

Originally Published: December 15, 2000

Last Updated: February 2, 2012

This article is available in print.
Vol. X, Fasc. 3, pp. 327-330