KAMBOJA, the name of a southeastern Iranian people (plur. OInd. Kambojāḥ) known by name only from Indo-Aryan epigraphic and literary sources since the Late Vedic period and living in the extreme northeastern area of Iranian tribes along the northern Indo-Iranian frontier. Their land likewise is called Kamboja-, but later also Kāmboja-. The name, the equivalent of which would be OIran. *Kambauǰa-, is etymologically unclear (Mayrhofer, Wörterbuch, I, p. 307). The former interpretation by Lévi 1923, 52–55 as an Austroasiatic prefix form to the ethnonym Bhoja- is generally rejected.

The name of the Kambojas is first attested in the Naighaṇṭukas (2.14), lists of rare (mostly dialectal or foreign) words compiled in Late Vedic times (being the earliest known work of Indian lexicography) and commented upon by Yāska in his treatise on “etymology” (Nirukta); here we read (2.2) śavatir gatikarmā Kambojev eva bhāyate “the word śavati as a verb of motion is used only by the Kambojas”; this is repeated nearly word-for-word by the grammarian Patañjali (2nd cent. BCE), Mahābhāya (I.9.25f.) and by later authors. This form śavati ‘goes’ phonetically is in full agreement with YAv. š́iiauua- (= OPers. šiyava-) < PIran. *ṣi̯au̯a- (in contrast to Ved. cyáva-, which in the Naighaṇṭukas appears in the same passage as śavati). From this we can conclude with certainty that the language of the Kambojas (“Kambōjī”) was an Iranian language with close relations to Younger Avestan (Kuhn; Grierson; Charpentier, pp. 144–45; Benveniste, pp. 46–47; Witzel, p. 92). The opinion of J. Wackernagel (p. 199; repr. p. 360, that the palatal ś shows the influence of some Northwest-Middle-Indo-Aryan language, is an unnecessary complication in view of YAv. š́- with its initial palatal. For Yāska, the word śavati obviously was a dialectal word he could use also for etymologically explaining other words like śū́ra- ‘hero’ or śván- ‘dog’. In any case, the Kambojas spoke a language different from that of the other Indo-Aryans. From the fact that Yāska and Patañjali here quote an example familiar to grammarians and lexicographers M. Witzel (p. 118 n. 108) concluded that this striking word must have become known to them about 500 BCE or even earlier.

The first attestations of the Kambojas that can be dated quite exactly are found in two of Aśoka’s (q.v.; mid-3rd cent. BCE) Great Rock Edicts (quoted here from the Kālsī version): In GRE 13.R yona-kabojeu ‘among Greeks and Kambojas’ (loc. plur.) and 5.J yona-kaboja-gadhālāna ‘of Greeks, Kambojas and Gandharians’ (gen. plur.). Here, the ethnonyms of two or three peoples are joined together, who lived at the border of the Mauryan Empire, were subject to its political, cultural and missionary influence and followed Aśoka’s ideology of “righteousness” (dhama), even if it is not quite clear what Aśoka did mean with Kaboja—only the Kambojas proper or all the Iranian tribes of his empire (cf. Benveniste, pp. 46ff.; Boyce and Grenet, Zoroastrianism, pp. 129ff., 148f.). Based on these passages, E. Benveniste drew a parallel between the Kambojas and Greeks mentioned there on the one hand, and the two non-Indo-Aryan languages of the Aśokan edicts found in Kandahar on the other: Greek and ‘Aramao-Iranian’ (see EPIGRAPHY i), the Iranian language underlying the text written in Aramaic script. These are the languages by which the Mauryan king intended to make known his religious message to the peoples in Eastern Afghanistan, somewhere between Kabul and Kandahar and in any case in the neighbourhood of Gandhāra. Therefore Benveniste (pp. 47f.) attributed the Iranian words found (as “eteograms”) in the Aramao-Iranian versions of Aśoka’s inscriptions also to the language of the Kambojas.

The same association of Greeks and Kambojas is also found in the Mahābhārata, the great Indian epic (see the references in Sörensen, p. 379), where such combinations of Kambojas and Yavanas ‘Greeks’ are attested, as well as those with the Gāndhāras, Bāhlīkas ‘Bactrians’ and Śakas ‘Indo-Scythians’. In Sanskrit geographical literature as well as in the Pāli Aguttaranikāya (1.213, 4.252, etc.) the Kambojas are listed last among the sixteen great peoples of the Indian subcontinent at the time of the Buddha. Moreover, we learn from Sanskrit and Pāli literary works of various kind (the earliest being the Late Vedic Vaśa-Brāhmaa, where a Sāmavedic teacher remarkably is characterized by his name as a Kambojan) several details about them: They shaved their heads completely, and they were ruled by a king (cf. Kamboja-in the sense of “Kambojan king” according to the rule of Pāṇini 4.1.175); the capital of that kingdom was called Rāja-pura- (lit. “King’s town”) in Mahābhārata (7.4.119), but cannot be located. The Kambojas were known as famous horse-breeders, and their horses are said to be the breed best suited for warfare. Those horses were imported to India in great number as well as tbeing delivered (like camels, asses, or various objects) as tribute.

Most interesting is a passage in the Pāli Jātaka (VI.208.27–30), from which it becomes clear that for the Indo-Aryans the Kambojas were a foreign, “non-Aryan” (anariya-) people with its own strange customs. For instance, they regarded it as their religious duty to kill insects, snakes, worms, frogs, and similar small animals. Scholars have long (see, e.g., Kuhn; Charpentier, pp. 145f.; etc.) connected this practice with the Avestan Vīdēdād. 14.5f. (and with Herodotus 1.140.3, where the same is said about the mágoi), and the conclusion has been drawn that the Kambojas were Mazdayasnians. In agreement with this view is the fact that Aśoka’s su-śrua- “good obedience” (thus GFE 4.C, Shāhbāzgaṛhī version) in the Kandahar bilingual (l. 6) is translated by hwptysty /hupatyāsti/ ‘good obedience’, which contains a typical word of the Mazdayasnian vocabulary, Av. paitiiāsti-, paiti.asti- ‘obedience’ (Benveniste, pp. 42, 47).

According to Fussman, pp. 33f. the unknown language of two rock-inscriptions from Dašt-e Nāwor (IDN 3 and 5), that are written in a script obviously descended from the Kharoṣṭhī, is a local Iranian idiom used there apart from Greek and Prākrit, namely that of the Kambojas, “Kambōjī” (cf. also Schmitt, 1994, pp. 185ff.), and something like “Proto-Ōrmuṛī“. If this hypothesis should prove to be true, we would be able to locate the Kambojas more precisely in the mountains around Ghaznī and on the Upper Arḡandāb.

Another hypothesis connected with the name Kamboja is the one set up most emphatically by Charpentier in 1923, that this ethnonym is related to the name of the Achaemenid kings Cambyses I and II (q.v.; OPers. k-b-u-j|-i-y-/Kambū̌jiya-/ or /-bauj-/), which simply would mean “king/ruler of the Kambojas” (p. 148). Again and again it was taken up and repeated (see, e.g., the references in Mayrhofer, 1979, p. 23), without, however, convincingly solving the formal problems: Part of the collateral tradition of the Great King’s name and in particular the Greek form of it, Kambýsēs, is not compatible with Kamboja (or OIran. *Kambauǰa-), since OIran. *-au- normally is rendered by Gk. -ō- (Schmitt, 2002, pp. 55f.). Besides, it is scarcely conceivable that a Persian noble was named “ruler of the Kambojas” in the time before Cyrus II (ibid., p. 56).


É. Benveniste, “Une bilingue gréco-araméenne d’Asoka. IV. Les données iraniennes,” JA 246, 1958, pp. 36-48.

Boyce and Grenet, Zoroastrianism, pp. 129ff., 148f.

J. Charpentier, “Der Name Kambyses (Kanbūǰiya),“ ZII 2, 1923, pp. 140-52.

G. Fussman, “Documents épigraphiques kouchans,” Bulletin de l’École Française d’Extrême-Orient 61, 1974, pp. 1-66.

G. A. Grierson, “The Language of the Kambojas,” JRAS 1911, pp. 801–2.

E. Kuhn, “Das Volk der Kambojas bei Yāska,” in Avesta, Pahlavi and Ancient Persian Studies in honour of the late Shams-ul-ulama Dastur Peshotanji Bahramji Sanjana, Strassburg and Leipzig, 1904, pp. 213-14.

B. Ch. Law, Tribes in Ancient India, 2nd ed., Poona, 1973, pp. 1-8.

S. Lévi, “Pré-aryen et pré-dravidien dans l’Inde,” JA 203, 1923, pp. 1-57.

Mayrhofer, Wörterbuch, I, p. 307.

M. Mayrhofer, Iranisches Personennamenbuch, vol. I, fasc. 2: Die altpersischen Namen, Vienna, 1979.

R. Schmitt, “Sprachzeugnisse alt- und mitteliranischer Sprachen aus Afghanistan,” in R. Bielmeier and R. Stempel, eds., Indogermanica et Caucasica. Festschrift für Karl Horst Schmidt, Berlin and New York, 1994, pp. 168-96.

Idem, Die iranischen und Iranier-Namen in den Schriften Xenophons, Iranica Graeca Vetustiora. II, Vienna, 2002.

S. Sörensen, An Index to the Names in the Mahābhārata, London 1904, repr. Delhi 1963, 1978.

J. Wackernagel, “Indoiranica,” ZVS 61, 1934, pp. 190-208; repr. in J. W., Kleine Schriften, vol. I, Göttingen, 1955, pp. 351–69.

M. Witzel, “Early Eastern Iran and the Atharvaveda,” Persica 9, 1980, pp. 86-128.

(Rüdiger Schmitt)

Cite this article: