ARAKADRI- (or Arkadri-, Elam. Ḫa-rak-qatar-ri-iš, Akk. kurA-ra-ka-ad-ri-ʾ), name of uncertain meaning given in Darius I’s inscription (DB 1.37) to a mountain in the region of Pišiyāuvādā (or Paišiya(h)uvādā). It was the seat of Gaumāta, the false Bardiya (Smerdis), whence he revolted against Darius. Pišiyāuvādā has been identified with Pasargadae in Fārs (e.g., J. Marquart, Untersuchungen zur Geschichte von Eran, Göttingen, 1896, p. 154) but it is rather strange that the base of a mage and anti-Achaemenid pretender to the throne should be sought in the heart of Achaemenid homeland (see W. Eilers, “Der Name Demawend,” Archiv Orientalni 22, 1954, pp. 268, 310). Neither Fārs nor the rest of Iran has retained a name (e.g., *Argīr, *Rayīr) recognizably derived from Ar(a)kadriš. There is no substance either to the claim that it is identical with the Assyrian hursagḪa-li-ḫa-ad-ri, a mountain mentioned in Tiglathpileser’s Annals (line 33; see M. Streck, in ZA 15, 1900, pp. 369ff.; and F. W. König in Reallexicon der Assyriologie I, Berlin and Leipzig, 1932, p. 129a), especially since this name seems to contain pre-Iranian elements. Old Persian Ar(a)kadriš might be a modified proper name from an older Zagros language (cf. the mountain uru/ḫursagSa-al-ad-ri, the last bastion of Elamite resistance against Assurbanipal; also the region of Adri conquered by Salmanassar III and the mountain ḫursagMi-li-ad-ru-ni mentioned by Tiglathpileser I as in the region of Naïri. The old name of the land Urartu still appears, in the reign of Salmanassar I [13th cent. B.C.], as Uruadri).

C. Bartholomae gave the name an Iranian etymology (*arka-dris- “facing the sun”) by comparing OInd. arka′- “ray, sun” and the root dṛś-, OIr. dars-, “to see” (Handbuch der altiranischen Dialekte, Leipzig, 1883, p. 210). Thus he placed the name in the category of sunny or shady places (see Eilers, in Proceedings of the Twenty-Seventh International Congress of Orientalists [Ann Arbor, 1967], Wiesbaden, 1971, pp. 92f.), but OInd. arká- and the verb árcati (“shines,” IE. root erkw-) is clearly foreign to Iranian. Moreover, the Old Persian nom. sing. form Ar(a)kad(a)riš as well as the Elamite and Akkadian versions suggest a phonemic form Ārakadriš rather than Ārkadriš. But the analysis remains unclear (ā or privative a-plus rak-?). The older interpretation as “bearing a castle” (K. Foy, in Kuhn Zeitschrift 35, 1899, p. 66) has been rejected by O. Szemerényi, “Iranica V,” Monumentum H. S. Nyberg II, Acta Iranica 5, Tehran and Liège, 1975, pp. 368ff.). New Pers. ark/arg (citadel), if a borrowing from Latin arx, arc-em or Greek hē ákra, akrópolis, can not be used to explain Old Iranian toponymy. And the IE. root arek-/alek- (protect, repulse) has not yet been attested in Indo-Iranian. In view of the application of the name, one might suggest for the second part the word adri- (rock, stone) attested in Indic. The compound would then have to be Ar(a)kadri-, with an athematic prior element. The suggestion of an Iranian adri- is supported by Ptolemy’s Paryadres (OIr. “rock-strewn”?) Mountain (5.13.5, 9; also Strabo 5.319 and Pliny 5.27.98; see R. Schmitt in BNF 15, 1964, pp. 297f., and H. Treidel in Pauly-Wissowa, Supp. X, 1965, cols. 484ff.). Admittedly, this mountain is somewhat distant, in northeast Anatolia; E. Honigmann identifies it with Armenian Parxar (Die Ostgrenze des byzantinischen Reiches von 363 bis 1071, Brussels, 1935, p. 181).



C. Bartholomae, Zum altiranischen Wörterbuch, Strassburg 1906, pp. 105, 116f.

H. Herzfeld, “Zarathustra,” AMI 1 , 1929-30, p. 86 n. 2.

F. W. König, Der Falsche Bardija, Vienna, 1938, pp. 183-84.

M. Mayrhofer, Kurzgefasstes etymologisches Wörterbuch des Altindischen I, Heidelberg, 1956, pp. 30, 548.

J. Pokorny, Indogermanisches etymologisches Wörterbuch I, Berne, 1959, p. 66.

W. Brandenstein and M. Mayrhofer, Handbuch des Altpersischen, Wiesbaden, 1964, p. 104.

(W. Eilers)

Originally Published: December 15, 1986

Last Updated: August 10, 2011

This article is available in print.
Vol. II, Fasc. 3, pp. 248-249