We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 26 April 2021
This article discusses the language of the Xūz mentioned in Arabic sources, endorsing the view that it is the latest attestation of the Elamite language. Drawing on models from historical sociolinguistics, it also studies the problem of mutual acculturation between speakers of Elamite and Persian in antiquity.
1 The syllable Xūz, Hūz, is a representation in Arabic script of the Persian name for the region or city of Susa (Susa being the Greek rendering of the name). In Old Persian, Susa is written with the signs u-v-j (Schmitt, R., Wörterbuch der altpersischen Königsinschriften (Wiesbaden, 2014), p. 263CrossRefGoogle Scholar), the initial h- being evanescent in the Persian dialect of the royal inscriptions and the j possibly representing ž. The name was likely Hūž. The Arabic plural form Ahwāz (modern Persian Ahvāz, also the name of capital of the contemporary Iranian province of Xūzestān) translates the Aramaic Beṯ Huzāye, “the land of the Hūzes,” and referred to the same region. In early Arabic, Ahwāz was also the common abbreviated name for a city in the region properly called Sūq al-Ahwāz, “the Market of the Xūzīs.”
2 On the name Elam, which comes to English ultimately from Akkadian and perhaps, via Akkadian, from Elamite itself, see Álvarez-Mon, J., ‘Elam: Iran's First Empire’, in A Companion to the Archaeology of the Ancient Near East, (ed.) Potts, D. T. (Malden, 2012), p. 740Google Scholar. See further Potts, D. T., The Archaeology of Elam: Formation and Transformation of an Ancient Iranian State (Cambridge, 1999), pp. 1–9CrossRefGoogle Scholar on the name Elam and the imprecision of the geography associated with this term, Szemerényi, O., ‘Iranica II’, Die Sprache 12 (1966), pp. 190–194Google Scholar, as well as the pioneering study of Nöldeke, Th., ‘Griechische Namen Susiana's’, Nachrichten von der Königleichen Gesellschaft der Wissenschaften und der Georg-August-Universität zu Göttingen 8 (1874)Google Scholar on the name of Susa.
3 Surveyed by Potts ‘The Archaeology of Elam’ and by several contributions in the new volume J. Álvarez-Mon, G. P. Basello, and Y. Wicks (eds.), ‘The Elamite World’ (Abingdon, 2018).
4 Grammatical summaries, with references to older scholarship, can be found in Khačikjan, M., The Elamite Language (Rome, 1998)Google Scholar, Grillot-Susini, F., ‘ELAM v. Elamite Language’, Encyclopaedia Iranica VIII/3 (1998)Google Scholar, Stolper, M., ‘Elamite’, in The Cambridge Encyclopedia of the World's Ancient Languages, (ed.) Woodard, R. (Cambridge, 2004)Google Scholar, and Tavernier, J., ‘Élamite: Analyse grammatical et lecture de textes’, Res Antiquae 8 (2011)Google Scholar.
5 See Stolper, ‘Elamite’, pp. 64–65 on just what aspects of the language remain poorly understood. The hypothesis, most recently articulated by McAlpin, D., ‘Brahui and the Zagrosian Hypothesis’, Journal of the American Oriental Society 135.3 (2015)CrossRefGoogle Scholar, possibly valid but not yet adequately reviewed, that it is related distantly to the Brahui language of Balochistan and to the Dravidian languages of southern India, has not contributed meaningfully to the elucidation of Elamite grammar or lexicon.
6 See Romaine, S., ‘Contact and Language Death’, in The Handbook of Language Contact, (ed.) Hickey, Raymond (Malden, 2010)Google Scholar and Winford, D., An Introduction to Contact Linguistics (Malden, 2003), pp. 256–267Google Scholar on first language attrition and death. The posthumously published paper on the end of Elamite by Black, J., ‘The Obsolescence and Demise of Cuneiform Writing in Elam’, in The Disappearance of Writing Systems: Perspectives on Literacy and Communication, (eds.) Baines, J., Bennet, J., and Houston, S. (London, 2008), p. 65Google Scholar confuses the demise of Elamite writing with the demise of Elamite language. His hypothesis that the growing adoption of Aramaic led to the attrition of Elamite is more likely, at least for Xūzistān.
7 The name “Susian” (“susien” and “susiaque” in French, and “susisch,” in German) was a strong early contender for the modern name of Elamite. See Potts, D. T., ‘Ælam Regio: Elam in Western Scholarship from the Renaissance to the Late 19th Century’, in The Elamite World, (eds.) Álvarez-Mon, J., Basello, G. P., and Wicks, Yasmina (Abingdon, 2018), pp. 16–17Google Scholar.
8 O. G. Bolshakov, “Eṣṭakri,” Encyclopaedia Iranica.
9 al-Iṣṭaxrī 91.10–11: wa-ammā lisānuhum fa-inna ʿāmmatahum yatakallamūna bi-l-Fārisīyati wa-l-ʿArabīyati ġayra anna lahum lisānan āxara Xūzīyan laysa bi-ʿIbrānīyin wa-lā Suryānīyin wa-lā Fārsīy. These words are cited verbatim by Ibn Ḥawqal, writing in 977 (354.15–16), and by Yāqūt (2.497.4–6) in the early thirteenth century.
10 F. von Spiegel, Erânische Alterthumskunde (Leipzig, 1871–1878), 3 vols, 3.753, though scholars had not yet agreed then that the name of the ancient language should be Elamite; Huart, ‘Khūzistān’, Encyclopaedia of Islam, 1st (ed.), vol. 4, p. 985; Spuler, B., Iran in the Early Islamic Period: Politics, Culture, Administration and Public Life between the Arab and the Seljuk Conquests, 633–1055 (Leiden, 2015), p. 231CrossRefGoogle Scholar, “We are unlikely to be wrong if we assume that this language was the last offshoot of the Elamite language”; G. Lazard, ‘Pahlavi, Pârsi, Dari: les langues de l'Iran d'après Ibn al-Muqaffa‘’, in Iran and Islam, (ed.) C. E. Bosworth (Edinburgh,1971), p. 363; Potts ‘The Archaeology of Elam’, p. 415; Orsatti, P., Appunti per una Storia della Lingua Neopersiana. Parte I: parte generale, fonologia, la più antica documentazione (Rome, 2007), pp. 27–28Google Scholar; Tavernier, J., ‘Elamites and Iranians’, in The Elamite World, (eds.) Álvarez-Mon, J., Basello, G. P., and Wicks, Y. (Abingdon, 2018), pp. 421–422Google Scholar. Khačikjan, The Elamite Language, p. 1, also countenances the possibility that this later language is Elamite.
11 Potts, The Archaeology of Elam, p. 430.
12 Cited by Ibn an-Nadīm on the authority of Ibn al-Muqaffaʿ, Fihrist, (ed.) G. Flügel, Kitâb al-Fihrist (Leipzig, 1871–1872), 1.13.1–8; (ed.) Muḥammad Riḍā Taǧaddud, Kitāb al-Fihrist li-n-Nadīm (Tehran, 1971), 15.20–21; (ed.) Ayman Fuʾād Sayyid, Kitāb al-Fihrist (London, 2009), 1/1.32.5–6: wa-ammā l-Xūzīyatu fa-bihā kāna yatakallamu l-mulūku wa-l-ašrāfu fī l-xalwati wa-mawāḍiʿi l-laʿbi wa-l-laḏḏati wa-maʿa l-ḥāšiya. The passage was cited by other authors, too: Ḥamza al-Iṣfāhānī (circa 893–circa 970) Kitāb at-Tanbīh ʿalā ḥudūṯ at-taṣḥīf, ed. Muḥammad Asʿad Ṭalas (Damascus, 1968), pp. 23–24; Abū ʿAbdallāh Muḥammad ibn Aḥmad al-Xwārazmī, Kitāb Mafātīḥ al-ʿulūm, (ed.) G. van Vloten (Leiden, 1895), pp. 116–117 = maqāla 1, faṣl 6, written circa 977–980.
13 Yāqūt 3.925.17–19, with the significant variant, for “places of recreation, etc.”: fī l-xalāʾi wa-mawḍiʿi l-istifrāġi wa-ʿinda t-taʿarrī li-l-ḥammāmi wa-l-ābzani wa-l-muġtasal, “in the privy and toilet, and at the time of disrobing for the bath, the wash-basin, and the bathtub”. It is worth mentioning, although its context is not entirely clear, a fragmentary passage from Mani's Book of Giants, a third-century text, edited by Henning, W., ‘The Book of the Giants’, Bulletin of the School of Oriental and African Studies 11.1 (1943), pp. 58 and 62CrossRefGoogle Scholar. Here the giants of the Manichaean myth, in ancient times, assign different tasks to people of different lands. While the Mesenians “prepare (?),” and the Persians perform a task not given because of a break in the text, the Xūzians (Hūžīgān) “sweep (and) water,” or in other words, clean the dirty ground of residences. Is this an aetiology for a stereotypical social role of Xūzian labour?
14 ʿAbdassalām Muḥammad Hārūn (ed.), al-Ǧāḥiẓ, Kitāb al-Bayān wa-t-tabyīn (Cairo, 1968), 1.34.8–9. In speaking about the consonants affected by lisps, he addresses “the š-sound,” aš-šīn al-muʿǧama [s with diacritic dots, i.e. š]. Here he remarks, “It is only one of many points of articulation—points of articulation being innumerable and a subject about which there is no agreement. The same consideration applies to the consonants in foreign languages, and it is true in no respect more than it is in the language of the Xūz.” innamā huwa maxraǧun mina l-maxāriǧa wa-l-maxāriǧu lā tuḥṣā wa-lā yūqafu ʿalayhā wa-kaḏālika l-qawlu fī ḥurūfi luġāti l-ʿaǧami wa-laysa ḏālika fī šayʾin akṯaru minhu fī luġati l-Xūz. On the problems of Elamite sibilant phonology, see Khačikjan, ‘The Elamite Language’, pp. 7–9, Stolper, ‘Elamite’, p. 71, Tavernier, ‘Élamite’, pp. 319–320, and above all J. Tavernier, ‘On the Sounds Rendered by the s-, š- and ṣ/z-Series in Elamite’, in Language in the Ancient Near East: Proceedings of the 53e Rencontre Assyriologique Internationale, vol. 1, part 2, (eds.) L. Kogan, N. Koslova, S. Loesov, and S. Tischchenko (Winona Lake, 2010), where no fewer than six or seven alveolar fricatives, palatal-alveolar fricatives, and affricates are posited for Elamite phonology. In any case, al-Ǧāḥiẓ clearly indicates that Xūzī phonology was complex.
15 ʿAbdassalām Muḥammad Hārūn (ed.), al-Ǧāḥiẓ, Kitāb al-Ḥayawān, 8 vols., (Cairo, 1938–1945), 5.289.9–14; cf. Miller, J., ‘Man Is Not the Only Speaking Animal: Thresholds and Idiom in al-Jāḥiẓ’, in Arabic Humanities, Islamic Thought: Essays in Honor of Everett K. Rowson, (eds.) Lowry, J. E. and Toorawa, Sh. M. (Leiden, 2017), p. 99Google Scholar. His comparandum to draw a contrast with the high difficulty of Xūzī is the language of the Zanǧ, slaves mostly of African origin put to work in southern Iraq. Their language, al-Ǧāḥiẓ states, can be learned, by one buying and selling them, in a single month. This must be a pidgin or creole language.
16 McWhorter, J., Language Interrupted: Signs of Non-Native Acquisition in Standard Language Grammars (Oxford, 2007)CrossRefGoogle Scholar; Wray, A. and Grace, G. W., ‘The Consequences of Talking to Strangers: Evolutionary Corollaries of Socio-Cultural Influences on Linguistic Form’, Lingua 117 (2007)CrossRefGoogle Scholar; Trudgill, P., Sociolinguistic Typology: Social Determinants of Linguistic Complexity (Oxford, 2011)Google Scholar.
17 M. J. De Goeje (ed.), al-Yaʿqūbī, Kitāb al-Buldān, Bibliotheca Geographorum Arabicorum vol. VII, (Leiden, 1892), 361.15–17. wa-li-ahli hāḏā s-suqʿi lisānun xāṣṣun bihim yušbihu r-raṭānata illā anna l-ġāliba ʿalayhimu l-luġatu l-Fārsīya. De Goeje found this passage attributed to al-Yaʿqūbī, from the lost portion of that author's geography, which dealt with Xūzistān, cited in a manuscript of the Manāhiǧ al-fikar wa-mabāhiǧ al-ʿibar of Ǧamāladdīn al-Waṭwāṭ (d. 1318). De Goeje doubted whether this particular sentence was original to al-Yaʿqūbī. The recent translation of the works of al-Yaʿqūbī likewise supposes that these words “may not belong to al-Yaʿqūbī's original text. See M. S. Gordon, Ch. F. Robinson, E. K. Rowson, and M. Fishbein, The Works of Ibn Wāḍiḥ al-Yaʿqūbī: An English Translation, 3 vols., (Leiden, 2018), 203.1. In fact, the reference to the special language of Xūzistān and the omission of Arabic as a major language in the region are testimonies to the early origin of the sentence. There can be no question that these were al-Yaʿqūbī's words.
18 M. J. de Goeje (ed.), al-Muqaddasī, Kitāb Aḥsan at-taqāsīm fī maʿrifat al-aqālīm (Leiden, 1887), 418.6: wa-lahum lisānun lā yufham.
19 V. Minorsky and C. E. Bosworth, ‘Rām-Hurmuz’, Encyclopaedia of Islam, 2nd edition.; Dénes Gazsi, ‘Rāmhormoz’, Encyclopaedia Iranica.
20 Hārūn, Ḥayawān, 4.68.1–3: al-Aṣmaʿīyu ʿan Abī Ẓubyān … qāla: al-Xūzu … wa-smuhum muštaqqun mina l-xinzīri, ḏahaba ilā smihī bi-l-Fārsīyati xūk fa-ǧaʿalati l-ʿArabu xūk xūzan. For the correct etymology of xūk, see Horn, P., Grundriss der neupersischen Etymologie (Strassbourg, 1893), p. 113 #510CrossRefGoogle Scholar.
21 Goeje, al-Muqaddasī, 418.6–11: abġaḍu l-kalāmi ilā llāhi l-Fārisīyatu wa-kalāmu š-šayāṭīni l-Xūzīyatu wa-kalāmu ahli n-nāri l-Buxārīyatu wa-kalāmu ahli l-ǧannati l-ʿArabīya.
22 M. J. de Goeje (ed.), al-Iṣṭaxrī, Abū Isḥāq Ibrāhīm ibn Muḥammad al-Fārisī, Kitāb al-Masālik wa-l-mamālik (Leiden, 1887), 314.9–10: wa-ammā lisānu Buxārā fa-innahā lisānu s-Suġdi, “As for the language of Buxārā, it is Sogdian.”
23 de Goeje, al-Muqaddasī, 403.
24 Amir Harrak (edited and translated), The Acts of Mār Mārī the Apostle (Atlanta, 2005), pp. 70–75. The text does not mention a distinct language.
25 See further the notes of Schwarz, P., Iran im Mittelalter nach den arabischen Geographen IV (Leipzig, 1921), pp. 407–409Google Scholar.
26 Thus K. Potowski, ‘Language Maintenance and Shift’, in The Oxford Handbook of Sociolinguistics, (eds.) R. Bayley, R. Cameron, and C. Lucas (Oxford, 2013), p. 323, summarises research indicating that “negative attitudes lead to rapid shift” in language use among a younger generation. As Romaine, ‘Contact and Language Death’, p. 323, puts it, with references to modern instances of language shift, “When a language is highly stigmatised, many are reluctant to admit that they speak it”.
27 Blois, F. de, ‘Elamite Survivals in Western Iranian. A Preliminary Survey’, Studia Iranica, Mesopotamica, et Anatolica 1 (1994)Google Scholar.
28 Thomason and Kaufman 1988, pp. 77–78; van Coetsem 2000, pp. 31–32.
29 de Blois, ‘Elamite Survivals’, p. 16; W. F. M. Henkelman, ‘Elamite Administrative and Religious Heritage’, in The Elamite World, (eds.) J. Álvarez-Mon, G. P. Basello, and Y. Wicks (Abingdon, 2018), p. 813.
30 There is general agreement that the first part of these words comes from this Elamite term for cuneiform inscriptions, but scholars are divided about the origin of the second part of these two words. This is not the place to attempt a resolution, but the reader may consult Mohammad Hassandoust (Muḥammad Ḥasan-dūst), An Etymological Dictionary of the Persian Language (Farhang-i rīša-šināxtī-i zabān-i Fārsī), 5 volumes, (Tehran, 2016 (1395)), 2.1267–1268 §2254 and 2.1405 §2504, for a starting bibliography of prior explanations.
31 Tavernier, J., Iranica in the Achaemenid Period (ca. 550–330 B.C.): Lexicon of Old Iranian Proper Names and Loanwords, Attested in Non-Iranian Texts (Leuven, 2007)Google Scholar presents the most comprehensive study of Old Persian words and names written in Elamite.
32 I. Yakubovich, review of Seth L. Sanders (ed.), Margins of Writing, Origins of Cultures, Chicago: Oriental Institute 2005, in Journal of Indo-European Studies 36 (2008), p. 207, proposes that “several centuries of Elamite-Iranian bilingualism imposed the Iranian structural features upon Elamite,” as a hypothesis. Cf. W. F. M. Henkelman, ‘Cyrus the Persian and Darius the Elamite: A Case of Mistaken Identity’, in Herodot und das Persische Weltreich / Herodotus and the Persian Empire, (eds.) R. Rollinger, B. Truschnegg, and R. Bichler (Wiesbaden, 2011), p. 588.
33 Van Coetsem's model, described with idiosyncratic technical terms he coined himself, is outlined primarily in Coetsem, F. van, Loan Phonology and the Two Transfer Types in Language Contact (Dordrecht, 1988)CrossRefGoogle Scholar, Coetsem, F. van, ‘Outlining a Model of the Transmission Phenomenon in Language Contact’, Leuvense Bijdragen 84 (1995)Google Scholar, and van Coetsem, F., A General and Unified Theory of the Transmission Process in Language Contact (Heidelberg, 2000)Google Scholar; D. Winford, ‘Contact-induced Changes: Classification and Processes’, Diachronica 22.2 (2005) has presented it more clearly, as has Butts, A. M., Language Change in the Wake of Empire: Syriac in Its Greco-Roman Context (Winona Lake, 2016), pp. 16–20Google Scholar. It is summarised, as relevant to his argument, by Henkelman, ‘Cyrus the Persian’, pp. 589–591. Note that van Coetsem's terms have been adopted by some specialists in contact linguistics, but not widely. The concept of using regularly co-occurring language contact settings and outcomes in language change to illuminate each other is discussed by P. Muysken, ‘Using Scenarios in Language Contact Studies’, in Greek Influence on Egyptian-Coptic: Contact-Induced Change in an Ancient African Language, (eds.) E. Grossman, P. Dils, T. S. Richter, and W. Schenkel (Hamburg, 2017).
34 Henkelman, ‘Cyrus the Persian’, pp. 588–595 and Henkelman, ‘Elamite Administrative’, pp. 813–815.
35 W. F. M. Henkelman, ‘The Achaemenid Heartland: An Archaeological-Historical Perspective’, in A Companion to the Archaeology of the Ancient Near East, 2 vols., (ed.) D. T. Potts (Malden, 2012), p. 933.
36 Henkelman, ‘Cyrus the Persian’, p. 592.
37 Miroschedji, P. de, ‘La fin du royaume d'Anšan et de Suse et la naissance de l'empire Perse’, Zeitschrift für Assyriologie 75 (1985)CrossRefGoogle Scholar; Henkelman, W. F. M., The Other Gods Who Are: Studies in Elamite-Iranian Acculturation Based on the Persepolis Fortification Texts (Leiden, 2008), pp. 47–49Google Scholar, Henkelman, ‘Cyrus the Persian’, pp. 582–584.
38 Henkelman, ‘Cyrus the Persian’, p. 595, and Henkelman, ‘Elamite Administrative’, pp. 804–805; thus also M. Stolper, ‘Elamite Sources’, in A Companion to the Achaemenid Empire, (eds.) B. Jacobs and R. Rollinger (London, forthcoming) and Tavernier, ‘Elamites and Iranians’, p. 171, with less conviction. It is more difficult to accept Henkelman's notion, ‘Cyrus the Persian’, p. 612, without qualification that “(t)he ethnogenesis of the Persians may actually be seen as the great dynamic behind the rise of the Achaemenid Empire”. Much lurks, potentially, in this phrase “the great dynamic behind the rise”, for historians construe causes and effects differently. It would seem to be an anomaly of history that an empire should arise primarily because of a new identity in a certain population, rather than the identity taking shape in the wake of prior material processes.
39 W. F. M. Henkelman, ‘Humban & Auramazdā: Royal Gods in a Persian Landscape’, in Persian Religion in the Achaemenid Period / La religion pers à l’époque achéménide, (eds.) W. F. M. Henkelman and C. Redard (Wiesbaden, 2017), pp. 293–295; Henkelman, ‘Elamite Administrative’, pp. 804–805.
40 Henkelman, ‘The Achaemenid Heartland’, p. 933.
41 Henkelman, The Other Gods Who Are, pp. 160–163.
42 Winford, An Introduction, pp. 220–223; Holm, J., Languages in Contact: The Partial Restructuring of Vernaculars (Cambridge, 2004)Google Scholar; Trudgill, Sociolinguistic Typology, pp. 1–15. The generally accepted theory of pidgin and creole genesis rests on this model of non-native learning. For van Coetsem, A General and Unified Theory, p. 52, native language use is subsumed under “linguistic dominance.”
43 van Coetsem, A General and Unified Theory, pp. 84–87.
44 Muysken, P., Bilingual Speech: A Typology of Code-Mixing (Cambridge, 2000), pp. 272–273Google Scholar.
45 Henkelman, ‘Cyrus the Persian’, p. 622 (parentheses in original).
46 Henkelman, ‘Cyrus the Persian’, pp. 591–593. A more comprehensive study of the Iranian features replicated in Achaemenian Elamite, along with criteria for identifying them, would be welcome.
47 See Winford, ‘Contact-induced Changes’, pp. 385–388, on structural borrowing in the context of van Coetsem's theory, as well as Heine, B. and Kuteva, T., Language Contact and Grammaticalization (Cambridge, 2005)CrossRefGoogle Scholar, Y. Matras, Language Contact (Cambridge, 2009), pp. 235–265, and Butts, Language Change, pp. 139–152 on grammatical replication.
48 Winford, An Introduction, pp. 259–261; Winford, ‘Contact-induced Changes’, pp. 402–409. Compare the situation of present-day Irish, in which English features are pervasively imported by the almost entirely bilingual population, whereas English at large is not receptive to features transferred from Irish. See R. Hickey, ‘Contact and Language Shift’, in The Handbook of Language Contact, (ed.) R. Hickey (Malden, 2010), pp. 163–166.
49 Heine and Kuteva, Language Contact, pp. 237–239.
50 van Coetsem, ‘Outlining a Model’, p. 63, van Coetsem, A General and Unified Theory, p. 49; Winford, ‘Contact-induced Changes’, pp. 374–378.
51 Henkelman, ‘Elamite Administrative’, p. 814.
52 van Coetsem, A General and Unified Theory, pp. 215–236.
53 Hock, H. H., Principles of Historical Linguistics (Berlin, 1991), p. 492CrossRefGoogle Scholar: “unidirectional (in an unequal prestige relationship)”; Matras, Language Contact, pp. 57–60: “Unidirectional bilingualism usually arises in circumstances where group A dominates certain activity domains to which group B members require access, but this relationship is not reciprocal. As a result, group B speakers will import into their own language word-forms acquired through interaction with group A in the relevant domains. … The word-forms that are imported by the minority language from the dominant language are not typically limited to domain-specific vocabulary that is associated with the domains in which language A is dominant… Borrowings may also occur in the domain of grammatical word-forms and even morphology.” Cf. Winford, An Introduction, p. 64, “one-way bilingualism,” and Myers-Scotton, C., Multiple Voices (Malden, 2006), pp. 48–49Google Scholar, on non-reciprocal bilingualism. For Henkelman, ‘Elamite Administrative’, p. 814, “asymmetrical bilingualism” occurs in the Achaemenian Elamite texts but in the reverse: he assumes that it is Iranophones who are capable of “expressing themselves reasonably well in Elamite in administrative matters but probably less so in other settings”.
54 de Blois, ‘Elamite Survivals’, pp. 13–15. This is more likely to be imposition of an Elamite feature on Old Persian by Elamophones.
55 This accords with a hypothesis of Thomason, S. G. and Kaufman, T., Language Contact, Creolization, and Genetic Linguistics (Berkeley, 1988), p. 68CrossRefGoogle Scholar: “a politically superordinate group is unlikely to become bilingual in a non-prestigious subordinate group's language unless the superordinate group is much the smaller of the two”. See also Winford, An Introduction, pp. 33–37, on “‘unequal’ bilingualism”.
56 van Coetsem, A General and Unified Theory, p. 216.
57 Yakubovich, I., ‘Persian Ezāfe as a Contact-Induced Feature’, Voprosy Jazykoznanija 5 (2020), pp. 92–98Google Scholar.
58 No stage of Persian exhibits the right-branching pattern exclusively. Users of Middle Persian frequently used left-branching noun phrases (as in weh dēn, “the good religion”) resorted to the construction “ān ī [attribute] [noun],” which P. O. Skjærvø, ‘Middle West Iranian’, in The Iranian Languages, (ed.) G. Windfuhr (London, 2009), pp. 221–224, calls periphrastic adnominal constructions. Likewise, the left-branching New Persian iḍāfat-i maqlūbī is not strange (see Phillott, D. C., Higher Persian Grammar (Calcutta, 1919), pp. 149–151, 442Google Scholar) not to mention New Persian noun compounds. All these specific phenomena deserve stringent syntactical studies along the lines suggested implicitly by Yakubovich.
59 Yakubovich, ‘Persian Ezāfe’, p. 101. This takes the premise of de Blois's inquiry of 1994 considerably further, using typological data only recently established when he investigated the problem.
60 Winford, ‘Contact-induced Changes’.
61 Butts, Language Change, pp. 43–63.
62 Butts, Language Change, pp. 30–40.
63 E. Grossman, ‘Greek Loanwords in Coptic’, in Encyclopedia of Ancient Greek Language and Linguistics, (ed.) G. K. Giannakis (Leiden, 2014). S. Torallas Tovar, ‘The Reverse Case: Egyptian Borrowing in Greek’, in Greek Influence on Egyptian-Coptic: Contact-Induced Change in an Ancient African Language, (eds.) E. Grossman, P. Dils, Tonio S. Richter, and W. Schenkel (Hamburg, 2017), p. 97, relates that “only about 140 Egyptian words have been found in all known Greek texts,” “often hapax legomena,” or nonce borrowings in the jargon of contact linguistics.
64 Henkelman, ‘Elamite Administrative’, p. 804.
