“truth” in Avestan. The Indo-Iranian concept of truth is preserved in the Gāθās and in the younger Avesta unchanged.


AṦA, “truth” in Avestan, from Indo-Iranian *ṛtá-, a neuter noun having the same meaning. The word is attested in Old Persian as ạrta and in Old Indian as ṛtá-.

i. Avestan aṧa.

ii. Old Persian Aṛta.


i. Avestan Aša

Indo-Iranian inheritance. By reconstructing on the basis of the Avesta (esp. the Gāθās of Zarathustra) and the Vedas (esp. Rigveda) inherited formulas (phrases, figures of speech) it is possible to recover fragments of the religion and poetry of the common Indo-Iranian period (about 2000 B.C.) otherwise not directly attested. The most important formulas containing *ṛtá are the following (for the complete material see B. Schlerath, Awesta-Wörterbuch, Vorarbeiten II, Wiesbaden, 1968, pp. 168-82): (1) *ṛtám man- “to think (of) truth” (Y. 31.19 and Rigveda); (2) *ṛtám vaźh- “to drive the truth” (Y. 46.4 and Rigveda): The religious poet drives the true words of his hymn as the charioteer the horses; the metaphors of Indo-Iranian poetry are taken from the world of warriors. From that we can conclude that there were in that period no classes like those attested later in the Avesta and the Veda. Nomad feudalism will have prevailed with cattle breeding as the most important economic basis. G. Dumézil’s theory that there was a three-class society already in the Indo-European period remains unproven; (3) *ṛtásya path “the path of the truth” (Y. 51.13 and Rigveda): The hymn goes on that path to the gods; (4) *ṛtā sak- “to follow (= to be in alliance with) truth” (Y. 34.2 and Rigveda): The poet follows the truth as a follower a feudal lord. Sak- is a social technical term (cf. *sakhi “companion”). Compare also śárdhān ṛtaśya (Rigveda 8.7.21) “troops of truth” and the Av. personnal name aṧasarəδa- “combatant of truth;” (5) *ṛtā van- “to win with the help of truth” (Y. 53.5 and Rigveda; (6) *ṛtām dhar- “to uphold the truth” (Y. 43.1 and Rigveda; (7) *ṛtām yaź- “to venerate the truth” (Y. 51.20, Yasna Haptaŋhāiti, and Rigveda); (8) *ṛtām sap- “to serve truth” (Y. 31.22 and Rigveda); (9) *ṛtām vasu “good truth.”

The Indo-Iranian concept of truth is preserved in the Gāθās and in the younger Avesta unchanged. But the main topic of the Rigveda, “the truth and the gods,” is lacking in the Gāθās. In the younger Avesta only the conspicuous theme of Yt. 10 “Miθra (= god Contract; and truth” shows Vedic parallels.

The meaning of *ṛtā- and*sátya-. It is generally accepted that the original meaning of *ṛtā- (Vedic ṛtā-, Av. aṧa-) is “truth.” That is confirmed by the fact that aṧa- is opposed Av. drug- “lie.” We have also the testimony of Plutarch, De Iside et Osiride 47, where aṧa is translated by alētheia the “truth.” Moreover, in Vedic ṛtá- is opposed to ánṛta- “untrue, false,” the meaning of which is also clear from Skt., whereas ṛtá- disappears in post-Vedic literature. Lastly, the meaning of ṛtá- was closely connected with that of sátya- “true.” Many scholars understand aṧa- and Vedic ṛtá- as also meaning “order” (cosmic order, social order, moral order) or “righteousness,” maintaining that such a central concept “cannot be precisely rendered by some single word in another tongue” (Boyce, Zoroastrianism I, p. 27). Since in all religions, philosophical systems, etc., the central terms have a special individual meaning that can be determined only by examining all the contexts in which those terms are attested, it is methodically more correct to take aṧa- always as “truth” and to derive from the sum of its occurrences the precise meaning of “truth” in the Avesta.

Beside the substantive aṧa- “truth” we have the adj. haiθiia- “true” (Indo-Ir. *sátya-, Old Indian sátya-).

The connotation of *sātya- is clearly “true” in the sense of “really existing.” This is evident from the expression haiθīm varəz- “to make true, bring to realization” (also in the compounds haiθiiāuuarəz-, haiθiiāuuarəštā-, haiθiiā.vərəziia-). Since Indo-Ir. *varź- was lost in OInd., the Vedic equivalent is satyam kar- “to make come true,” also in the OInd. noun satya-kriyā (Pali saccakiriyā). To perform a Satyakriyā means to pronounce a truth and then to utter words meaning “just as that is true, so too will that and that become true.” In many cultures we find similar methods of compelling reality by parallelism (German Analogiezwang). The close parallels between Veda and Avesta are dealt with by H. Lüders, Varuṇa II, Göttingen, 1959, pp. 505-09. OInd. satya-kriyā- meant originally “making true (by the utterance of a true statement)” and not “pronouncing a true statement” as Lüders considered. Thus, the Indo-Ir. formula *sātyas mántras (Y. 31.6 and Rigveda) means not simply “true Word” but “formulated thought which is in conformity with the reality” or “poetic (religious) formula with inherent fulfillment (realization).”

The inherited formulas containing ṛtá- reveal also another concept of truth: *Ṛta′- is a poetical formulation of religious knowledge and judgment where the main stress is laid on the fact that the single parts of the utterance are properly joined together. The point is not the conformity with the reality but the internal correctness. Thus, Y. 47.2 ptā aṧahiia mazdǡ “(Lord) Wisdom (is) the father of truth:” The truth lies in the identification of “Wisdom” and “father of truth.” Hence it is clear why in India in post-Vedic times, when identification as a typical form of religious knowledge disappeared, the word ṛtá- was lost and only sātya- as adj. and substantive was preserved. Also in Iran aṧa- survived only as a Zoroastrian term. NPers. has rāst for “true” from OIr. (Old Pers.) rāsta- “prepared, made straight.”

In the Avesta the opposite of aṧa- and haiθiia- is drug- “lie” and in India the opposite of ṛtá- and sátya- is ánṛta-, which meant both “not internally in conformity” and “not in conformity with the reality.” Ánṛta-survived accordingly in classical Sanskrit whereas ṛtá- was lost because identification is of little importance in Hinduism.

These meanings of aṧa- and haiθiia- can be supported by etymologies: *ṛtá- past part. pass. from the IE. root *ar- “to join (properly) together” and *sátya- from *sat- “being, existing.”

Moreover, it is clear that the narrow sense of *ṛtá- “truth consisting in a correct joint (connection of thoughts, words)” overlaps partly with the concept of “order.” The *ṛtá- reflects structural items and relations of the cosmic order. But since *ṛtá- exists only when the appropriate words are uttered, only the translation “truth” is acceptable.

Aṧa and the other Aməṧa Spəṇtas. The Gāθās are hymns of praise for Ahura Mazdā “Lord Wisdom.” The literary form, in all probability the meter and many of the poetical formulas, is inherited from Indo

Iranian times. As a consequence of the glorification of the abstract concept of Wisdom, where no myths are handed down, the hymns are mainly devoted to the association of other abstract concepts with Wisdom. These concepts are the following (in brackets is given the number of times the respective word occurs in the 238 verses of the Gāθās): aṧa- “Truth” (157), vohumanah- “Good Thinking” (136), xšaθra- “Dominion” (56), ārmaiti- “Devotion” (40), aši- “Reward” (18), amərətatāt- “Life” (14), hauruuatāt- “Wholeness” (11), sraoša- “Obedience” (7). Virtually all these abstract concepts can be looked upon as persons. A central theme in the Gāθās in that Ahura Mazdā is identical with these abstract powers. They coincide and merge in Ahura Mazdā. Hence their differences are of secondary importance and are reflected only in their appellative meaning. The underlying aim is to make them “aspects” of Ahura Mazdā.

In the Younger Avesta the above-mentioned beings (except aši- and sraoša-) form together with Ahura Mazdā the group of the so-called seven Aməṧa Spəṇtas “Bounteous Immortals.” The limitation of the number and the concrete conception of a group of divine beings may be due to later theological systematizing but is undoubtedly in agreement with the thoughts of Zarathustra. In Yt. 13.83 it is said that all seven Aməṧa Spəṇtas have the same thoughts, words, and deeds. Such a leveling of the differences between them likewise accords with Zarathustra’s way of thinking.

Since the Aməṧa Spəṇtas represent the totality of good moral qualities, it is easy to understand why, by analogy with the inherited opposition between *ṛtá- “truth” and *drugh- “lie,” the other Aməṧa Spəṇtas were similarly assigned their evil counterparts. Already in the Gāθās we find vohu- manah- opposed to aka- manah-, xšaθra- to dušəxšaθra (not in YAv.), and ārmaiti- to tarə̄maiti-. In Yt. 19.96 we find further hauruuāt- “Wholeness” opposed to šud- “Hunger” and amərətāt- “Life” to taršna- “Thirst.”

The Aməṧa Spəṇta “Aṧa” and “fire.” In the Pahlavi texts (e.g., Bundahišn 3.12-19; Šāyest nē-šāyest, 15.5-6, 30 in F. M. P. Kotwal, The Supplementary Texts to the Šāyest nē-šāyest, Copenhagen, 1969) the seven Aməṧa Spəṇtas correspond to the seven elements. In that context Ardwahišt or Ašwahišt (Av. aṧa- vahišta-) “best truth” is connected to the fire or is the overlord over the fire. The complete system of correspondences is apparently late, but we have many passages in the Gāθās, and in the Younger Avesta which attest that the underlying ideas were present in some form throughout the history of Zoroastrianism. Of course the relation between truth and fire cannot be an arbitrary invention or mere association. Fire is the same in all its manifestations: fireplace, sacrificial fire, lightning, and sun. We can suppose that the Indians also knew about the (secret) identity of all light and fire phenomena because their identity plays an important role in the Rigveda. Since both Indians and Iranians knew a fire ordeal (cf. B. Schlerath, Das Königtum im Rig- und Atharvaveda, Wiesbaden, 1960, pp. 153-90) and since for the Iranians apparently every ordeal on earth was regarded as a microcosmic counterpart of the ordeal of the last judgment, we have one possible starting-point for the identification of truth and fire. Another may have been the fact that the place of the sacrificial fire was the place of the recitation of the highest (or “best” Av. vahišta-) truth (see H. Lommel, Die Religion Zarathustras, Tübingen, 1930, pp. 40-52, 120-25; Boyce, Zoroastrianism I, pp. 27-28, 204, 211-12). The identity of xšaθra- “Dominion” and aiiah- xšusta- “molten metal,” which denotes the last judgment, does not contradict this explanation of the relation between Aṧa and fire since (desirable) Dominion is also a form of truth and results from truth.

Bibliography: Given in the text.

(B. Schlerath)


ii. Old Persian Ata

In view of the frequent occurrence in the Achaemenid inscriptions of words for “lie” (drauga- “lie,” draujana- “liar, deceitful,” durujiya- “to lie”) it is surprising that the words aṛta- “truth” and aṛtāvan- “truthful” are found in only one inscription, Xerxes’ so-called daiva-inscription (XPh). The adjective aṛtāvan- is found in a phrase concerning the recompense of the god-fearing after death (see Ašavan), and aṛta- occurs three times in a much-discussed expression (XPh 40-54): Auramazdām yad- aṛtācā brazmaniy(a) “to worship Ahuramazdā a. b.” No explanation of the two words has so far been accepted by the communis opinio, a fact due to the morphological ambiguity of the words resulting from the nature of the Old Pers. script. The words are not translated in the Akkadian and Elamite versions of the inscription, merely transcribed, a fact which points to the expression being an old, inherited or borrowed, religious formula. The Akkadian form is arta-šá′-ʾ bi-ra-za-man-ni-i, the Elamite has ir-ta-ha-ci pīr-ra-ic-man-ni-ia (E. Herzfeld, Altpersische Inschriften, Berlin, 1938, pp. 29-31). Both the Akk. form bi-ra-za-man-ni-i and the Elam. pir-ra-ic-man-ni-ia indicate a reading brazman- rather than *bṛzman- or *barzman-.

There are two plausible ways of analyzing aṛtācā: 1. aṛtā instrumental sing. “with/through aṛtā” + “and.” Here, may connect aṛtā with the preceding Auramazdām (thus Kent, Old Persian, p. 82) or be a so-called “inverted ” (i.e., A- B for A B- “A and B”) and connect aṛtā with brazmaniy(a). 2.  AÂṟrtācā, or rather artā/ăacā (cf. the Elamite form) is for artā/ă-hacā “from/in accordance with aṛta,” cf. Gathic Av. ašāt hacā (AirWb., col. 231). (The normal word-order in Old Persian, different from Avestan, is for hacā to precede its noun, so if this analysis is the correct one, then we have a further indication that this is an old, possibly borrowed formula.) The loss of the intervocalic -h- would be similar to that in θāha¡y, θātiy “you say, he says” beside aθaha “he said” (cf. Kent, Old Persian, p. 46). A third proposal is that of J. Duchesne-Guillemin (BSOAS 25, 1962, pp. 336f.), who suggested derivation from *artānc- “turned towards aṛta.” A fourth analysis which deserves mention is that of R. Schmitt (Orientalia, N.S. 32, 1963, pp. 442ff.), who takes it to be the locative sing. of *aṛtu- (Av. ratu-, OInd. ṛtu-) “time, season.”

The second word admits of two principal analyses, either as brazmaniy, locative sing. of brazman-, or as brazmaniya- “connected with brazman.” The word may (most probably?) be connected with Av. barəsman Mid. Pers. barsom (OInd. barhíṣ) “bundle of sacred twigs,” thus either “at the barsom” or “with barsom.” Kent (Old Persian, p. 201 s.v.) equates brazmaniya with OInd. brahmaṇyà “religious” and translates “prayerful, reverent.” Another solution was advanced by W. B. Henning (in TPS, 1944, pp. 108-18), who connected the word with Man. Mid. Pers. and Parth. brahm “appearance, (correct) behavior, etc.”

Some of the proposed translations are the following: Kent (Old Persian, pp. 151-52) “worship Ahuramazda and Arta reverent(ly);” M. Boyce (Zoroastrianism II, pp. 174-76) “with due order and rites;” K. Hoffmann (in public lectures) “according to the (cosmic) truth (and order) with barsom (in the hands)” (with reference to the probable representations of magi (?) holding barsom in the treasure of the Oxus [see O. M. Dalton, The Treasure of the Oxus, London, 1964, p. xxvi, pls. XIII, XIV]); “in accordance with Ṛta and the ritual” (M. Schwartz in Camb. Hist. Iran II, 1985, pp. 689f.).

See also Ahlav; and Ardwahišt.



For further bibliographical references see Kent, Old Persian, pp. 170, 201.

W. Brandenstein and M. Mayrhofer, Handbuch des Altpersischen, Wiesbaden, 1964, p. 111.

M. Mayrhofer in M. Boyce and I. Gershevitch, eds., W. B. Henning Memorial Volume, London, 1970, p. 285.

(P. O. Skjærvø)

(B. Schlerath, P. O. Skjærvø)

Originally Published: December 15, 1987

Last Updated: August 16, 2011

This article is available in print.
Vol. II, Fasc. 7, pp. 694-696