Folk Etymology in the North-Eastern Neo-Aramaic Dialects

Aramaic Studies 16 (2018) 215–233 Aramaic Studies brill.com/arst Folk Etymology in the North-Eastern Neo-Aramaic Dialects Hezy Mutzafi Dept. of Hebrew Language and Semitic Linguistics, Tel Aviv University, Tel Aviv, Israel [email protected] Abstract Although folk etymology is a common linguistic phenomenon, it has hitherto hardly been touched upon in lexicological and other works related to varieties of NeoAramaic. The present article concerns twelve cases of folk etymology selected from some of the dialects of North-Eastern Neo-Aramaic (NENA), the largest and most variegated division of modern Aramaic. Among these are three folk-etymological interpretations that did not induce structural or other changes, as well as nine cases of folketymological processes that reshaped NENA lexical items. Keywords folk etymology – folk-etymological interpretation – folk-etymological alteration – NENA – lexical cognates 1 Preliminaries1 Folk etymology, also known as popular etymology or associative etymology, is a linguistic phenomenon in which a certain word or phrase is erroneously interpreted as being related to another due to the synchronic phonetic, structural and often semantic similarity of these lexemes, despite their distinct 1 I am grateful to Prof. Matthew Morgenstern and Dr. Samuel Fox for their valuable comments. Data on Neo-Aramaic regional varieties are fieldwork-based, unless a reference is adduced, © koninklijke brill nv, leiden, 2018 | doi:10.1163/17455227-01602007 216 mutzafi etymological backgrounds.2 In many cases such a spurious associative connection motivates an alteration of a certain word (or word constituent) so that it is brought closer to—or even replaced by—another word. More specifically, alteration induced by folk etymology is the phonological, structural, semantic or orthographic change and de-isolation of an opaque, isolated word under the influence of a semantically transparent word that is synchronically misconstrued as being related to the opaque one. Thus, for instance, in the case of the archaic and dialectal English compound sand-blind ‘half blind, dim-sighted’, linguists assume that the component *sam‘half’ in reconstructed Old English *samblind ‘half blind’ was replaced in later English with the phonetically similar and semantically transparent noun sand due to the opaqueness and isolation of sam- ‘half’, which became obsolete along the way towards modern English.3 and except for Western Neo-Aramaic, based on Arnold, forthcoming, a copy of which was kindly made available to me by the author. Notes on transcription: k̭ is an unaspirated phoneme whereas k is aspirated. Vowel length is indicated only where it is phonemic, i.e., for ā vs. a (in some dialects). Superscript + indicates word-emphasis. Stress is penultimate unless otherwise indicated (transcription of Neo-Aramaic words quoted from scholarly works is adapted to this method). Abbreviations: Ar. – Arabic, Aram. – Aramaic, C. – Christian (NENA dialect), CM – Classical Mandaic, Gk. – Greek, J. – Jewish (NENA dialect), J. Az. – Jewish Azerbaijan (dialect cluster), JBA – Jewish Babylonian Aramaic, Kurd. – Kurdish, lit. – literally, NENA – North-Eastern Neo-Aramaic, NM – Neo-Mandaic, pre-mod. – pre-modern, Syr. – Syriac. Main sources for pre-modern Aramaic are: E. Drower and R. Macuch, A Mandaic Dictionary (Oxford: Clarendon Press, 1963), M. Sokoloff, A Dictionary of Jewish Babylonian Aramaic of the Talmudic and Geonic Periods (Ramat-Gan: Bar-Ilan University Press., 2002) (henceforth: DJBA), and M. Sokoloff, A Syriac Lexicon (Winona Lake: Eisenbrauns, 2009); for Akkadian: W. von Soden, Akkadisches Handwörterbuch (3 vols.; Wiesbaden: Harrassowitz, 1965–1981); for Kurdish: M. Chyet, Kurdish-English Dictionary. Ferhenga Kurmancî-Inglîzî (New Haven, CT: Yale University Press, 2003) (henceforth: Chyet, Dictionary), D. İzoli, Ferheng Kurdi-Tırki (Istanbul: Deng Yayınları, 1992) (henceforth: İzoli, Dictionary); and for Persian: S. Hayyim, The New Persian-English Dictionary (Tehran: Farhang Moʾaser, 1997), and F. Steingass, Comprehensive Persian-English Dictionary (2 vols.; London: Routledge & K. Paul, 1892). 2 For selected literature related to folk etymology see S. Ullmann, The Principles of Semantics (Oxford: Blackwell, 2nd edn, 1957) pp. 91–92; idem, Semantics: An Introduction to the Science of Meaning (Oxford: Blackwell, 1962) (henceforth: Ullmann, Semantics), pp. 101–105, 164, 220– 222; G. Rundblad and D. Kronenfeld, ‘Folk-Etymology: Haphazard Perversion or Shrewd Analogy?’, in J. Coleman and C. Kay (eds.), Lexicology, Semantics and Lexicography (Amsterdam: John Benjamins, 2000) pp. 19–34; A. Shiler, Language History: An Introduction (Amsterdam: John Benjamins, 2000) pp. 86–89; H. Hock and B. Joseph, Language History, Language Change and Language Relationship: An Introduction to Historical and Comparative Linguistics (Berlin: Mouton de Gruyter, 2nd edn, 2009) pp. 169–170, 226; and S. Michel, ‘Word-Formation and Folk Etymology’, in P. Müller et al., Word-Formation: An International Handbook of the Languages of Europe (2 vols.; Berlin: De Gruyter Mouton, 2015), vol. 2, pp. 1002–1019. 3 See Ullmann, Semantics, pp. 102–103; and, similarly, H. Hock, Principles of Historical Linguistics (Berlin: Mouton de Gruyter, 2nd edn, 1991) pp. 202–203. Aramaic Studies 16 (2018) 215–233 folk etymology in the north-eastern neo-aramaic dialects 217 Structural and other innovations motivated by folk etymology are commonplace, as is shown by the many examples that have been adduced in linguistic literature related to European languages.4 Although the Semitic languages are naturally not immune to folk etymology and its effects, systematic studies of folk-etymological interpretations and alterations in Semitic languages appear to be exiguous. The only works dedicated entirely to this topic are, as far as I know, Ignác Goldziher’s article entitled ‘Arabische Beiträge zur Volksetymologie’, Gad Ben-Ami Sarfatti’s article ‘Popular Etymology in Modern Hebrew’, and Leonid Kogan’s article ‘Popular Etymology in the Semitic Languages’.5 Some further cases of folk etymology are scattered in scholarly works related to Semitic linguistics and various Semitic and other languages.6 With regard to Aramaic, dictionaries and other works mention a handful of interpretations and alterations induced by folk etymology in pre-modern Aramaic varieties, as well as a few such cases in Neo-Aramaic. Among the cases of folk etymology in pre-modern Aramaic is Syr. peṣḥā ‘Passover; paschal sacrifice’, from Hebrew pɛsaḥ and apparently influenced by, or directly borrowed from, Greek πάσχα.7 Assuming this Greek connection, the 4 For numerous examples in English see A. Palmer, Folk-Etymology: A Dictionary of Verbal Corruptions or Words Perverted in Form or Meaning, by False Derivation or Mistaken Analogy (London: George Bell, 1882). 5 See I. Goldziher, ‘Arabische Beiträge zur Volksetymologie’, Zeitschrift für Völkerpsychologie und Sprachwissenschaft 18 (1888), pp. 69–82; G. Ben-Ami Sarfatti, ‘Popular Etymology in Modern Hebrew’ [in Hebrew], Leš 39 (1975), pp. 236–262; idem, ‘Popular Etymology in Modern Hebrew—Ending’ [in Hebrew], Leš 40 (1976), pp. 117–141; and L. Kogan, ‘Popular Etymology in the Semitic Languages’, in L. Kogan (ed.), Studia Semitica (Moscow: Russian State University for the Humanities, 2003) pp. 120–140. I am grateful to Prof. Simon Hopkins for making Goldziher’s paper known to me. 6 To take a few examples, cases of folk etymology in Hebrew and Arabic (as well as Yiddish and some other languages) are discussed in G. Zuckermann, ‘“Etymythological Othering” and the Power of “Lexical Engineering” in Judaism, Islam and Christianity: A Socio(sopho)logical Perspective’, in T. Omoniyi and J.A. Fishman (eds.), Explorations in the Sociology of Language and Religion (Amsterdam: John Benjamins, 2006) pp. 237–258; some possible cases of faunal terms reinterpreted or reshaped by folk etymology in some of the Semitic languages are adduced in A. Militarev and L. Kogan, Semitic Etymological Dictionary, vol. 2: Animal Names (Münster: Ugarit-Verlag, 2005) p. liv; and a few cases of folk etymology in Harari are discussed in W. Leslau, ‘Gleanings from the Harari vocabulary’, Rassegna di Studi Etiopici 16 (1960), pp. 23– 37 (32). In addition, a few cases of folk etymology in the Babylonian Talmud are adduced in M. Grünbaum, ‘Assymilationen und Volksetymologien im Talmud’, ZDMG 42 (1888), pp. 248– 257, and a few in Egyptian Arabic are found in K. Vollers, ‘Beiträge zur Kenntnis der lebenden arabischen Sprache in Aegypten’, ZDMG 50 (1896), pp. 607–657. 7 Consider C. Brockelmann, Lexicon Syriacum (Halle: Max Niemeyer, 2nd edn, 1928) (henceforth: LS) p. 587a. Aramaic Studies 16 (2018) 215–233 218 mutzafi change s > ṣ in this word likely occurred by assimilation to *p̣ as a reflex of the unaspirated Greek consonant written with π,8 and the result, peṣḥā [p̣eṣḥā], was probably synchronically interpreted as if related to the native Syriac verbal root pṣḥ ‘to rejoice’, hence Passover as a ‘festival of joy’.9 Another case of folk etymology in classical Aramaic appears to be JBA ‫טייעא‬, ‫‘ טיאעא‬Arab’. The JBA by-form ‫ טייא‬reflects more closely the etymon ṭāʔiyy, the name of an Arab tribe, whence also Syr. ṭayyāyā ‘Arab, Muslim’ is derived. It seems to me that the non-etymological ‫ ע‬in JBA ‫טייעא‬, ‫( טיאעא‬obviously reflecting an orthographic rather than phonetic change) was introduced by a folketymological association with the JBA verb ‫‘ טעי‬to err, get lost, stray, go astray’, especially with this verb in the meaning ‘to stray’, which is very much akin to the meaning ‘to wander’.10 2 Cases of Folk Etymology in Neo-Aramaic Based on Scholarly Literature (1) Ṭuroyo ṭayo ‘Muslim’ is folk etymologised, at least by some speakers, as if related to ṭaʕyo ‘one in error’,11 the connection to the Arab tribe of ṭāʔiyy (see the etymology of JBA ‫ טייעא‬and Syr. ṭayyāyā above) being no longer remembered. (2) Ṭuroyo ṭawʕənto ‘bramble’ and ṭawʕune ‘bramble fruit (pl.)’ are, according to some informants, related to ṭawʕune ‘little loaves of the Host’ as a 8 9 10 11 For Gk. π > Syr. p̣ and the assimilation of adjacent consonants to p̣ see A. Butts, ‘The Integration of Consonants in Greek Loanwords in Syriac’, AS 14 (2016), pp. 1–35 (18–20). There is, however, also Syr. psk, pskʔ ‘Passover’ < πάσχα. This associative interpretation was first recognised by Th. Nöldeke, Neue Beiträge zur semitischen Sprachwissenschaft (Strassburg: Trübner, 1910) p. 37; see also LS, p. 587a. For a different explanation of this excrescent ‫ ע‬see J. Blau, On Pseudo-Corrections in Some Semitic Languages (Jerusalem: Israel Academy of Sciences and Humanities, 1970) pp. 50–51. For another, more established case of inserted ‫ ע‬by folk etymology see ‫בי‬ ‫עקתא‬, ‫ביעקתא‬, ‫‘ ביקתא < בעיקתא‬hut’ in J. Epstein, Studies in Talmudic Literature and Semitic Languages (2 vols; Jerusalem: Magnes, 1983–1988 [in Hebrew]), vol. 2, p. 835; DJBA 205a, 217; M. Morgenstern, Studies in Jewish Babylonian Aramaic Based on Early Eastern Manuscripts (Winona Lake: Eisenbrauns, 2011) p. 71. The origin of ‫ביקתא‬, however, remains unknown. See A. Tezel, Comparative Etymological Studies in the Western Neo-Syriac (Ṭūrōyo) Lexicon (Uppsala: Uppsala Universitet, 2003) (henceforth: Tezel, Comparative Studies) p. 24 n. 13. This folk-etymological interpretation is reflected in the idiosyncratic (and possibly idiolectal) neologism maṭʕiyono ‘Muslim’, lit.: ‘one who leads astray, leads to error’, in a text elicited from a speaker of the Ṭuroyo dialect of ʿIwardo. See Beṯ Șawoce, Sayfo b Țurcabdin 1914–1915 (Genocide in Ṭur ʿAbdin, 1914–1915 [in Ṭuroyo]) (Södertälje: Nisibin, 2006) p. 130/18. Aramaic Studies 16 (2018) 215–233 folk etymology in the north-eastern neo-aramaic dialects 219 diminutive of ṭawʕe, sg. ṭawʕo ‘Host’; but Tezel suggests that this might be a folk-etymological change of a form such as Syr. ṭʕūntō/ṭʕantō ‘crop of fruit’.12 (3) Midyat-Ṭuroyo (viz. the Ṭuroyo dialect of the town of Midyat) nuxroyo ~ nəxroyo ‘fiancé; foreigner’ as opposed to rural Ṭuroyo and 19th-century Midyat-Ṭuroyo nxiroyo ‘fiancé’ and nuxroyo ‘foreigner’, may well be a relevant case.13 The etyma of the two distinct words nxiroyo and nuxroyo are attested in Western Syriac as mḵīrō (root mkr) ‘betrothed, fiancé’ and nuḵrōyō (root nkr) ‘foreigner’. The change in the dialect of Midyat appears to have been induced by folk-etymological association of nxiroyo ‘fiancé’ with the phonetically similar word nuxroyo ~ nəxroyo ‘foreigner’ and is related to the fact that in Ṭur ʿAbdin betrothed couples were prohibited any intimate acquaintance.14 (4) J. Zakho qahwāna ‘music record’, which is borrowed from Iraqi Arabic qawāna with a later insertion of h probably due to a folk-etymological association with qahwa ‘coffee’, given that phonograph records were initially played in Kurdistan only in a qahwxāna ‘coffee-house’.15 Another case mentioned hesitantly in Sabar’s dictionary is, to my mind, hardly relevant: J. Zakho reškāsa ‘testicles’ is adduced under the entry (ʔe)škāṯa with a remark: ‘folk-etymology “on the belly”?’16 It is more likely, however, that the change *ʔəškāsa (cf. colloquial J. Nerwa ʔəškāsa ‘testicles’, sg. ʔəšǝksa) > reškāsa ‘testicles’ is a taboo-motivated lexical replacement of the expected J. Zakho reflex *ʔəškāsa with a rhyming compound literally meaning ‘on the belly’ (*rēš karsā). 12 13 14 15 16 Tezel, Comparative Studies, pp. 167–168. Cf. Tezel, Comparative Studies, p. 104. Cf. O. Jastrow, Laut- und Formenlehre des neuaramäischen Dialekts von Mîdin im Ṭûr ʿAbdîn. (4th edn.; Wiesbaden: Harrassowitz, 1993) (henceforth: Jastrow, Midin) p. 235 n. 6. The claim in Jastrow, Midin, p. 233 n. 15 that nəxroyo and nxiroyo share the same (historical) root is untenable, as the author himself seems to have realised later (Tezel, Comparative Studies, p. 104 n. 75). The separate etymologies of these nouns were first recognised in Tezel, Comparative Studies, pp. 104–105, where, however, nxiroyo > nəxroyo is not considered a case of folk etymology. See Y. Sabar, A Jewish Neo-Aramaic Dictionary: Dialects of Amidya, Dihok, Nerwa and Zakho, Northwestern Iraq (Wiesbaden: Harrassowitz, 2002) (henceforth: Sabar, Dictionary) p. 13 n. 54. Sabar, Dictionary, p. 101a. Aramaic Studies 16 (2018) 215–233 220 3 mutzafi New Observations on Folk Etymology in the NENA Dialects In what follows I shall discuss a dozen cases and possible cases of folk etymology in various NENA dialects: three cases of a folk-etymological interpretation without structural changes followed by nine cases of structural alterations induced by folk etymology. 3.1 Folk-Etymological Interpretation in NENA without Alteration 1. ʕezqəṯā > *ʕǝsəqṯa > səqlá > sqəltá ‘ring’ < sqəltá ‘beautiful’ The Proto-NENA word for ‘ring’ (as a jewel) was cognate with, inter alia, Syr. ʕezqəṯā ‘ring, signet’, JBA ‫‘ עיזקתא‬ring seal’ and CM ʿzqta, ʿsqta, ʿsaqta, ʿsiqta ‘ring, ring seal’. As in the majority of forms in CM, in all NENA dialectal cognates the sibilant consonant is s (by partial assimilation to the unvoiced q),17 and the inherited Proto-NENA antecedent of contemporary cognates could be reconstructed as *ʕǝsəqṯā (cf. Ṭur. ʕisaqṯo ‘ring’ < *ʕəsəqṯo). Among the more conservative dialectal NENA reflexes are ʔəsəqṯa (e.g. in J. Dohok and ʿAmidya) and ʔisəqṯa (e.g. in Alqosh and Ṭyare). Among the more progressive forms are those with aphaeresis of the initial syllable: Qaraqosh, Bariṭle and Betanure səqṯa (in the latter dialect alongside ʔəsəqṯa), and, similarly, Sandu səqsa. The form səqṯa is the direct precursor of səqlá ‘ring’ in the Trans-Zab dialect of Rustaqa, with the regular Trans-Zab shift ṯ >l.18 In all other Trans-Zab dialects the cognate form is sqəltá,19 with a feminine ending -ta appended to the erstwhile feminine ending lá (< ṯa). This suffixation with -ta took place after the l in səqlá had been synchronically interpreted as part of the root. The same formation occurs in some other cases in the Trans-Zab dialects, e.g. (1) təkṯa ‘trouser-cord’ (as, e.g., in Betanure) > Rustaqa təklá ‘id.’ > J. Sulemaniyya təklá ~ tkəltá ‘id.’, elsewhere in Trans-Zab tkəltá ‘id.’; and (2) *ʔakkəḏāyṯā ‘a breed of hen, Akkadian hen’20 > *ʔakṯeṯa > J. Sanandaj ʔaklelá ~ ʔakleltá ‘hen’,21 Saqiz ʔakleltá ‘id.’. 17 18 19 20 21 Forms with z (in Syriac script) in A. Maclean, A Dictionary of the Dialects of Vernacular Syriac (Oxford: Clarendon press, 1901) (henceforth: Maclean, Dictionary) p. 237b are based on etymology rather than genuine NENA speech. For the Jewish Neo-Aramaic dialect group of Trans-Zab see H. Mutzafi, ‘Trans-Zab Jewish Neo-Aramaic’, BSOAS 71/3 (2008), pp. 409–431. A similar form is Barzan squlta ‘ring’, which must be a borrowing from neighbouring TransZab, given that the regular reflex of ṯ in Barzan is Ø rather than l. For this etymon see H. Mutzafi, Comparative Lexical Studies in Neo-Mandaic (Leiden: Brill, 2014) pp. 38, 193. See the discussion in G. Khan, The Jewish Neo-Aramaic Dialect of Sanandaj (Piscataway, NJ: Gorgias Press, 2009) pp. 174–175, and further examples there. Aramaic Studies 16 (2018) 215–233 folk etymology in the north-eastern neo-aramaic dialects 221 In the Trans-Zab dialect cluster of Iranian Azerbaijan and adjacent areas (henceforth: J. Az., comprising dialects such as J. Urmi, J. Salmas and Naghada) the form sqəltá ‘ring’ is homophonous with the adjective sqəltá ‘beautiful, pretty’, which is a feminine derivative of sqilá ‘beautiful, handsome’, the sg.m. passive participle of the verb sql, sāqə́l ‘to become beautiful’. The Aramaic provenance of J. Az. sqilá and sāqə́l can be demonstrated by comparison to Syr. sqīlā ‘elegant, polished’ and sqal ‘to polish, embellish, adorn’, among other pre-modern Aramaic cognates. The homophony between (*ʕəsəqṯā >) sqəltá ‘ring’ and (*sqīltā >) sqəltá ‘beautiful, pretty’ in the J. Az. dialects triggered a synchronic folk-etymological interpretation of sqəltá ‘ring’ as basically denoting ‘beautiful’, as was expressed by my informants. 2. temrē > təmre ‘eyelids’ < təmre ‘dates’ Qaraqosh has two homophonous nouns pronounced təmre: təmre1 ‘eyelids’, sg. təmra, and təmre2 ‘dates’ (as fruit), sg. tmərṯa. The former is related to Syr. temrē, sg. temrā ‘eyelid’, which in all likelihood is a cognate of Biblical Hebrew šəmuroṯ ‘eyelids’,22 hence Qaraqosh təmre1 ‘eyelids’ harks back to a proto-form with ṯ. Qaraqosh təmre2 ‘dates’ is related to Syr. tamrē, sg. tmarṯā ‘date’, the Semitic cognates of which—such as Hebrew tåmår ‘date’, Ar. tamr ‘dates’—prove that its proto-form was with t. The homophony of təmre1 and təmre2 coupled with the slightly similar, oval shape of both referents, induced a folk-etymological association between the two, so that the common phrase təmrəd ʔene ‘eyelids’ (corresponding to Syr. temrē d-ʕaynē) is synchronically interpreted, at least by some speakers, as if literally meaning ‘dates of the eyes’.23 3. Kurd. ferman ‘decree; edict of massacre’ > parman ‘massacre’ < prm ‘to slaughter’ Northern Kurdish ferman, basically ‘order, decree, edict’ has a secondary meaning ‘edict of massacre or genocide’.24 These meanings are reflected in the loanword farman ‘decree; massacre (led by the authorities against a minority)’ in some NENA dialects (e.g. Tisqopa, Betanure and Barzan). Similar lexical items 22 23 24 See further possible cognates in LS, p. 828a. Consider also G. Khan, The Neo-Aramaic Dialect of Qaraqosh (Leiden: Brill, 2002) p. 746b, where the phrase təmrəd ʔena ‘eyelids’ is presented as related to təmre ‘dates’. See İzoli, Dictionary, p. 151a. Furthermore, Kurdish ferman in the meaning ‘genocide’ may well be attested in Chyet, Dictionary, p. 191b fermanê fila, if this is to be construed ‘the genocide of the Armenians’ rather than Chyet’s rendering ‘edict regarding the Armenians’. Aramaic Studies 16 (2018) 215–233 222 mutzafi are the Ṭuroyo loanwords farmā ́n ‘sultan’s order’ and fərmā ́n or fərman ‘massacre’,25 and Mlaḥsô farmán dá-mšiḥoye ‘the massacre of the Christians’.26 Various Christian NENA (C. NENA) varieties that exhibit the shift of foreign f to native p borrowed the Kurdish word as parman (e.g. Baz, Dez, Sat) or parmən (most Ṭyare dialects) ‘massacre, genocide’, as well as paṛman (Ṣara,27 with spontaneous r > ṛ), p̣aṛmǝn (MarBishu,28 with p > p̣ near ṛ) with the same meanings. It seems that the form parman has spread from these dialects, as well as from the Urmi-based koine spoken by many of the modern Assyrians, to some other C. NENA dialects where foreign f is regularly preserved, such as Tkhuma, Halmun and Barwar. The coincidental similarity of parman, parmən ‘massacre’ to the verb prm, pārəm (in Ṭyare: pɻm, pāɻəm) ‘to slaughter’ (also ‘to cut’) induced a folk-etymological connection between the two lexemes, as is evident from informants who consider the foreign noun to be related to the native verb.29 3.2 Folk-Etymological Alterations of NENA Words 4. sāḥōrā ‘beggar’/*sāḵōrā ‘blocker’ > saxora ‘stye’ > sixora ‘id.’ < sixora ‘porcupine; porcupine quill’ Various NENA dialects evince the term saxora (e.g. Alqosh, Betanure, Barashe), saxoɻa (most Ṭyare dialects) or saḥura (e.g. Jinet), denoting ‘stye (sty of the eye)’. I should like to propose two alternative etyma of saxora (and cognates) as worthy of consideration: (1) sāḥōrā ‘beggar’ or, less likely, (2) *sāḵōrā ‘blocker’. In detail: (1) Pre-mod. Aram. sāḥōrā is attested in Syriac in the sense of ‘beggar’. Reflexes with this meaning are found in some of the NENA regional varieties: saxora in Qaraqosh, Bariṭle, ʿAnkawa, C. Koy Sanjaq, C. Shaqlawa, C. Sulemaniyya and C. Sanandaj; and saḥura in Umra Ḥtaya, which also denotes ‘stye’.30 25 26 27 28 29 30 See H. Ritter, Ṭūrōyo: Die Volkssprache der syrischen Christen des Ṭūr ʿAbdîn, B. Wörterbuch (Beirut: Orient-Institut der Deutschen Morgenländischen Gesellschaft, 1979) pp. 151, 159. See Jastrow 1994: 74/1, 173. The author renders this phrase ‘the persecution of the Christians’, but the context is that of massacre or genocide. As in Sh. Talay, Neuaramäische Texte in den Dialekten der Khabur-Assyrer in Nordostsyrien (Wiesbaden: Harrassowitz, 2009) (henceforth: Talay, Texte) p. 384/3. As in Talay, Texte, pp. 408/10, 414/5 et passim. Consider also Talay, Texte, p. 158/48, where the two lexemes occur side by side in a narrative in the dialect of Tkhuma-Gawaya: kəllɛ iraqaye w kurdiyye primilɛ go simele, wíḏewa parman ‘the Iraqis and Kurds slaughtered them [the Assyrians] all in Simele, they perpetrated a massacre’. Cognates in other Neo-Aramaic languages are Neo-Mandaic sahurɔ ‘beggar’ and, similarly, Western Neo-Aramaic reflexes of the unattested antecedent *saḥḥārā, i.e. Jubbʿadin saḥḥōra and Maʿlula ṣaḥḥōra (with s > ṣ likely from partial assimilation to ḥ). Aramaic Studies 16 (2018) 215–233 folk etymology in the north-eastern neo-aramaic dialects 223 This proposed etymology presupposes a radical semantic shift ‘beggar’ > ‘stye’, which may seem, prima facie, far-fetched. Nonetheless, this semantic shift can be postulated based on the fact that at least in some NENA-speaking communities there was a tradition of giving alms to beggars and paupers as a means of curing a stye.31 Furthermore, the same semantic shift is strikingly evident in Levantine Arabic šaḥḥā ́ḏ or šaḥḥā ́d ‘beggar’ > ‘beggar, stye’;32 and, similarly, in post-classical Arabic šaḥḥā ́ḏ ‘(importunate) beggar’, šaḥḥā ́ḏ ʔal-ʕayn ‘stye’.33 The background of the semantic shift in Arabic is similar to that postulated for NENA, namely Levantine traditions of begging for alms as a means of curing a stye.34 Much farther afield, a striking independent parallel occurs in Japanese, where monomorai ‘beggar; stye’ initially denoted ‘beggar’ only, with the semantic expansion to include ‘stye’ motivated by a custom of begging a handful of rice in order to cure a stye.35 In all three cases (NENA, Arabic and Japanese) the background of the semantic shift ‘beggar’ > ‘stye’ is either giving alms or begging for alms in order to cure a stye. If this etymology is valid, the dialect of Umra Ḥtaya is, as far as is known to me, the only NENA regional variety that preserves both erstwhile and transferred denotations of the polysemic noun saḥura ‘beggar; stye’. 31 32 33 34 35 For a somewhat similar etymology and semantic shift involving ‘man’ > ‘lesion’ consider English furuncle < Latin furunculus ‘little thief’. See, inter alia, L. Bauer, Deutsch-arabisches Wörterbuch der Umgangssprache in Palästina und Libanon (Wiesbaden: Harrassowitz, 2nd edn, 1957) pp. 58a, s.v. Bettler, 132, s.v. Gerstenkorn; A. Barthélemy, Dictionnaire arabe-français: Dialectes des Syrie: Alep, Damas, Liban, Jérusalem (6 vols.; Paris: Librairie Orientaliste Paul Geuthner, 1935–1969), vol. 2, p. 379, s.v. šaḥḥā ́d; A. Barghouthi, Dictionary of Colloquial Palestinian Arabic Dialect (RamallahBireh: Jamʿīyyat ʾInʿāš ʾal-ʾUsrah, 2001 [in Arabic]) (henceforth: Barghouthi, Dictionary) p. 660, s.v. šaḥḥā ́ḏ. See H. Wehr, A Dictionary of Modern Written Arabic (Ithaca, NY: Spoken Language Services, 3rd edn, 1976) (henceforth: Wehr, Dictionary) p. 457a. See A. Frayha, A Dictionary of Non-Classical Vocables in the Spoken Arabic of Lebanon (Beirut: American University of Beirut, 1947 [in Arabic]) (henceforth: Frayha, Dictionary) p. 91a, where it is stated that some people believe that a stye cannot be cured unless the afflicted person begs for alms from seven women by the name of Maryam; and Barghouthi (Dictionary, p. 660) mentions a custom whereby a person afflicted with a stye goes out to some houses to beg for a mouthful of food, and after he eats that food his stye is cured. See W. Grootaers, ‘Folk Etymology in Japanese Dialects’, in R. Crespo et al. (eds.), Aspects of Language: Studies in Honour of Mario Alinei, vol. 2: Theoretical and Applied Semantics (Amsterdam: Rodopi, 1987) pp. 195–202 (198–199), where the background of the semantic shift ‘beggar’ > ‘stye’ in Japanese is elucidated in view of the fact that ‘in Japan there is a recurring belief that begging a handful of rice from three neighbours and eating the gruel cures a stye … Hence the standard language name for the stye is monomorai “beggar” ’. Aramaic Studies 16 (2018) 215–233 224 mutzafi (2) Alternatively, the origin of saxora ‘stye’ might have been *sāḵōrā ‘blocker’, derived from a verbal root that is still reflected in NENA as Qaraqosh sāxər ‘to block’ (also in Jilu sāxər ‘to dam, fill’, according to Maclean, Dictionary, p. 226a).36 The ḥ in Jinet and Umra Ḥtaya saḥura poses no obstacle to this etymology, given that in these dialects *ḵ [x] shifted to *ḥ.37 I am rather inclined, however, to prefer sāḥōrā ‘beggar’ as the etymon, since the semantic connection between beggar and stye is anchored in the local culture and has a striking parallel in Levantine Arabic. Whereas the etymology of saxora (or saḥura) ‘stye’ remains somewhat uncertain, what is clear is that various NENA dialects evince a folk-etymological association of saxora ‘stye’ with the Kurdish loanword sixora (and variant forms), denoting ‘porcupine’, and in some NENA dialects (e.g. J. Zakho) also ‘porcupine quill’,38 as if a stye pricks the eye(lid) like the sharp point of a porcupine’s quill.39 The result was a replacement of saxora with—or its restructuring into—sixora (e.g. in Bne Romta-Ṭyare, Borb-Ruma Bohtan), sixorra (in C. Aradhin, Tin), sixurra (e.g. in Tkhuma, J. Zakho),40 sixuṛṛa (e.g. in Harbole, J. Dohok), or sixərra (in Isnakh), all denoting ‘porcupine; stye’, and some also ‘porcupine quill’. 36 37 38 39 40 I do not know of any other NENA dialect that preserves this verb. The etymology *sāḵōrā was already implied by A. Oraham, Dictionary of the Stabilized and Enriched Assyrian Language and English (Chicago: Consolidated Press, 1943) p. 355, where the term in question is spelt sāḵōrā in Syriac script. Additionally, Maclean (Dictionary, p. 225b) listed the vocable spelt sākōrā (in Syriac script) and transcribed sâkûrâ as a term for ‘stye’ in some unspecified NENA dialects spoken by tribal Nestorians of Hakkâri. Such a form, however, could not be corroborated by my informants, nor is the etymon *sāḵōrā expected to end up as a word with k. An alternative etymon, *sakkūrā, would have indeed yielded sakura in Hakkâri-NENA, but such an etymon is less likely than *sāḵōrā, given the agentive function of the productive pattern qaṭola vis-à-vis the stative use of the much more limited pattern qaṭula in these dialects. As in *rakkīḵā > *rakixa > rakiḥa ‘soft’. To be precise, it is likely that the pre-NENA phonemes /*ḥ/ and /*ḵ/ merged into /*x/ in Proto-NENA, and that in some if not most dialects of the northwestern NENA region of Bohtan (e.g. Jinet, Umra Ḥtaya, Hertevin) the merged phoneme /*x/ shifted to /ḥ/, as in *prāḥā > *prāxa > prāḥa1 ‘to fly’, *prāḵā > *prāxa > prāḥa2 ‘to rub’. Cf. Sh. Talay, Die neuaramäischen Dialekte der Khabur-Assyrer in Nordostsyrien: Einführung, Phonologie und Morphologie (Wiesbaden: Harrassowitz, 2008) pp. 44–45. Some NENA dialects use parra ‘feather’ or kǝtwa ‘thorn’ rather than sixurra for ‘porcupine quill’. Cf. Telkepe zəqṭa ‘a goad; stye’, related to Syr. zeqṯā ‘goad, sting’ (< Akk. ziqtu ‘barb, point’). Another animal name that came to mean ‘stye’ in various NENA dialects is qaroya (Sat), qaruya (MarBishu), +qaruwa (Timur, Gawar), +k̭aruwwa (Jilu), +k̭aruvva (C. Urmi), all ‘cock’ > ‘cock; stye’, apparently by comparison of a stye to the red comb of a cock. The J. Zakho form is already attested in Sabar, Dictionary, p. 239b. Aramaic Studies 16 (2018) 215–233 folk etymology in the north-eastern neo-aramaic dialects 225 It should also be mentioned that in dialects that did not undergo a folketymological change to sixora (or a kindred form) there is a clear distinction between the vocables for ‘stye’ and ‘porcupine(’s quill)’, e.g. Jinet saḥura ‘stye’: sixurra ‘porcupine’, Betanure saxora ‘stye’: sixurra ‘porcupine(’s quill)’, J. Nerwa (extinct since 2012) saxora: sixura ‘ditto’, Lizin-Ṭyare saxoɻa ‘stye’: sixorra ‘porcupine’. 5. narqōs > marqos, marqus g(l)ilaya ‘narcissus’ < marqus ‘(St.) Mark(’s flower)’ The Syriac flower name narqōs,41 a by-form of narqīs ‘narcissus’ (borrowed from Greek), is the etymon of various Christian NENA cognates with the same meaning. Nearly all these cognates expanded the basic form with the singular ending a or ta, and nearly all exhibit m rather than n as the initial sound of the word. Among the expanded cognates are narqoza (Sat), marqosa (C. Aradhin), marqoza (Alqosh), marquza (e.g. Bebede, Tilla), marqusta (e.g. Mer, Ashitha) and marqoza ~ marquṣta (Barwar).42 The forms with z were plausibly influenced by the Kurdish parallel nêrgiz. The change n > m, exhibited by all cognates except Sat narqoza, is typologically common and is akin to the change naddālā > madāla ‘centipede; millipede’ in Ashitha, Tin, Iṣṣin and some other Christian NENA dialects.43 As for Barwar marquṣta, its ṣ has perhaps emerged by partial assimilation to q. Two dialectal cognates have arisen by folk-etymological conflation with the name of Saint Mark: Bne Belatha-Ṭyare marqus ‘Mark’, viz. ‘Saint Mark’s flower’, and Bariṭle marqus glilaya ~ marqus gilaya ‘(St.) Mark the Galilean’.44 The form marqus might have developed as an alteration or replacement of an expanded form such as the aforementioned marqosa rather than of narqōs, in view of the large distribution of forms expanded with a or ta. An unlikely diachronic scenario would be to assume that forms such as marqosa and marqusta developed from marqus rather than from narqōs. Such an assumption would mean that 41 42 43 44 Vocalisation is according to T. Audo, Sīmtā d-Leššānā Suryāyā (Treasure of the Syriac Language) (2 vols.; Mosul: Imprimerie des Pères Dominicains, 1897), vol. 2, p. 115a. See G. Khan, The Neo-Aramaic Dialect of Barwar (3 vols.; Leiden: Brill, 2008) p. 1068. The intermediate antecedent maddālā is attested mainly in the mediaeval lexicons by Bar ʿAli and Bar Bahlul, in the latter as a word in the (early NENA?) dialect of Tikrit. See G. Hoffmann, Syrisch-arabische Glossen: Autographie einer Gothaischen Handschrift enthaltend Bar Ali’s Lexikon von Alaf bis Mim (Kiel: Schwers’sche Buchhandlung, 1874) p. 212/5438; R. Duval, Lexicon syriacum auctore Hassano bar Bahlule (3 vols.; Paris: E Reipublicæ Typographæo, 1888–1891) (henceforth: Duval, Bar Bahlul), vol. 2, p. 836, s.v. yadyādā. St. Mark was not a Galilean by birth, however. Another plant name related to a Christian saint is Ashitha ləxme-máryəm ‘(St.) Mary’s bread’, referring to a kind of edible herb. Aramaic Studies 16 (2018) 215–233 226 mutzafi various C. NENA dialects restructured marqus ‘(St.) Mark’ into forms remote from the name referring to the narcissus as Saint Mark’s flower. 6. dlá-lwāṯi > dlá-nāṯi ‘unbeknownst to me; without my permission’ < nāṯa ‘ear’ Betanure dlá-nāṯi and J. Nerwa dlá-nāsi ‘unbeknownst to me; without my permission’ have cognates in two other regional varieties of the same Lishana Deni dialect cluster, namely ʿAmidya dlá-lāṯi and J. Zakho lá-lāsi ~ lá-lʔāsi, both of which denote ‘without my knowledge, while I was unaware’. Whereas the ʿAmidya and J. Zakho cognates are opaque, the Betanure and J. Nerwa ones are synchronically transparent, comprising the components dla ‘without’ and nāṯi, nāsi ‘my ear’.45 The etymology of these compounds is, however, not related to ‘ear’ at all. It can readily be clarified in the light of cognates in Christian NENA dialects spoken in the areas of Zakho and Bohtan, where the second component is clearly a reflex of the pre-modern Aramaic preposition ləwāṯ ‘to, towards, with’, French chez: (1) Birsive lwāṯi ‘chez moi’, lá-lwāṯi ‘unbeknownst to me’. (2) Borb-Ruma (Bohtan) lá-lwoti ‘unbeknownst to me’. (3) Hertevin lá-lwati ‘unbeknownst to me; in my absence; without me’, lálwated ‘unbeknownst to; in the absence of (someone), without (someone)’.46 These C. NENA cognates and cognate meanings may point to the following etymology of the ʿAmidya and J. Zakho forms: (1) dlá-lwāṯi or lá-lwāṯi ‘not with me’ or ‘in my absence’ > (2) dlá-lwāṯi or lá-lwāṯi ‘unbeknownst to me’ (< I was not there to know about it) > loss of w in ʿAmidya dlá-lāṯi ‘unbeknownst to me’ > J. Zakho lá-lāsi ‘id.’ > excrescent ʔ in J. Zakho lá-lʔāsi ‘id.’.47 45 46 47 The etymology of nāṯa ‘ear’ is ʔeḏnāhāṯā ‘ears’, attested in Syriac and see H. Mutzafi, ‘The Reflexes of the Word ‫‘( אדנא‬ear’) in Eastern Neo-Aramaic: Etymology, Diversification and Innovation’, in M. Bar-Asher and M. Florentin (eds.), Samaritan, Hebrew and Aramaic Studies Presented to Professor Abraham Tal (Jerusalem: Bialik Institute, 2005 [in Hebrew]) pp. 229–242. Based on O. Jastrow, Der neuaramäische Dialekt von Hertevin (Provinz Siirt) (Wiesbaden: Harrassowitz, 1988) pp. 158/525 lá-lwatew ‘[they put the goblet in his sack] in his absence’, 158/526 lá-lwated ʔaḥew Benyamen ‘[they put Joseph’s goblet in the sack] in the absence of his [Joseph’s] brother Benjamin’, 166/593 le ʔoden ʔidara lá-lwated dab-baḥta ‘I couldn’t make it without this woman’. The first two examples could easily be rendered instead as ‘unbeknownst to him’ and ‘unbeknownst to his brother Benjamin’, which would fit the context just as much as ‘in his absence’. Cf. Sabar, Dictionary, p. 203a, where lá-lʔāsi is considered as perhaps harking back to the verbal root lʔs ‘to chew’. Aramaic Studies 16 (2018) 215–233 folk etymology in the north-eastern neo-aramaic dialects 227 The forms in Betanure and J. Nerwa, dlá-nāṯi and dlá-nāsi, respectively, hark back to the antecedent dlá-lāṯi (attested in ʿAmidya). Although the change l > n is typologically common, the change dlá-lāṯi > dlá-nāṯi, dlá-nāsi is hardly a mere phonological process. It was, in all likelihood, rather induced by folketymological association with nāṯa, nāsa ‘ear’ (< *ʔeḏnāhāṯā ‘ears’) in these dialects, and indeed dlá-nāṯi and dlá-nāsi were construed by informants as literally meaning ‘without my ear’, as in Betanure: brāti pləṭla m-beṯa dlá-nāṯi ‘my daughter went out of the house without my knowledge’, or: ‘I wasn’t aware of her leaving the house’, lit.: ‘my daughter went out of the house without my ear’ (I didn’t hear about it). A later meaning of dlá-nāṯi and dlá-nāsi is ‘without my permission’, as in pləṭla m-beṯa dlá-nāṯi ‘she went out of the house without my permission [and had I known about her intention, I wouldn’t have let her do so].’ 7. *bar-zīwānā > baɻziwāna ‘firefly’ > baɻza-wāna ‘id.’ < baɻza wāna ‘the ewe dries up’ In the Ṭyare dialects of Walṭo, Bne Belatha and Bne Romta, as well as in the dialect of Ṭal, the noun denoting ‘firefly’ is baɻziwāna (Ṭyare) or barziwāna (Ṭal). The etymon of this term most likely comprised the components bar ‘son’ and zīwānā ‘radiant, glowing’ (< zīwā ‘glow’ with the nomen agentis suffix -ānā— consider Syr. zīwānā ‘splendid or bright object’ and CM ziuana ‘bright, radiant’). In some other Ṭyare dialects, such as Sarspidho, Darawa and Ko d-Chalwe, the form is baɻza-wāna; and in the Ṭyare dialects of Ashitha and Chamba d-Mallik the plural form baɻza-wāne is used for the singular as well. Similarly, in the Baz dialects of Maha Khtaya, Be-Selim and Shwawwa the cognate is barza+wāna. The striking change of i to a in the forms baɻza-wāna, baɻza-wāne, barza+wāna could hardly be accounted for as having any phonological motivation. Rather, a folk-etymological interpretation of the plural form baɻziwāne ‘fireflies’ has probably triggered the phonetic change: At least in some Ṭyare villages there was a saying ʔɛgət palṭi baɻziwāne, baɻzi wāne ‘when fireflies are seen (literally: ‘emerge’) the ewes become dried up [of their milk]’, referring to the period of early summer. Similarly, in the Assyrian tribe of Baz people used to say: Ɂaygət parxi barza-+wāne, +wāne ybarzi mən +xalwa ‘when fireflies fly, the ewes become dried up of [their] milk’. Furthermore, among speakers of Ashitha there is a popular belief that fireflies sleep on leaves during the day, and when ewes eat vegetation with fireflies they consequently cease lactating. Hence a folkloristic connection was established between fireflies and ewes (wāne) being dried up of their milk (baɻzi) around the period of time when Aramaic Studies 16 (2018) 215–233 228 mutzafi fireflies were seen.48 Since the singular correspondent of baɻzi wāne (‘the ewes stop lactating’) is baɻza wāna (‘the ewe stops lactating’), the result was that a single firefly ended up being called baɻza-wāna (pl. baɻza-wāne), or as a depluralised form baɻza-wāne in some Ṭyare dialects. The same folk-etymological association lies behind the Baz compound barza-+wāna, which synchronically consists of the verb barza ‘she (might) be dried up’ and the noun +wāna ‘ewe’. 8. *qarṣʕānā > qarəṣena, +qorasayna ‘nettle’ < qārəṣ Ɂena/+Ɂayna ‘eye-pincher’ The plant name qrṣʕnɁ ‘nettle’ is listed in Bar Bahlul’s mediaeval lexicon as a plural form.49 If the vocalisation qarṣʕānē, adduced in most modern Syriac dictionaries,50 is correct, its singular form would probably be *qarṣʕānā.51 Although the etymology of this plant name is obscure, it is not unlikely that its first syllable is a reduced form of the verb qāreṣ ‘it stings’. The form *qarṣʕānā, or a closely related kindred form, appears to be the antecedent of various NENA dialectal cognates. Gawar +qərsana and Jilu +k̭ərsana are still fairly close to *qarṣʕānā, whereas some other NENA varieties exhibit vowel changes that swept their respective reflexes farther away from the etymon. Among these varieties are J. Nerwa qarəṣena, Betanure qaraṣena, and Sat +qorasayna. The morpho-phonological changes in these three dialects can at least partly be accounted for as resulting from folk-etymological association with Ɂena (Betanure, J. Nerwa) or +Ɂayna (Sat) ‘eye’ (after all, there is no phonological motivation whatsoever for the change of ā to e or ay), and a folk-etymological 48 49 50 51 This folk etymology by association with ewes is reflected in the Syriac spelling barzāwāʕnē (for Ashitha baɻza-wāne ‘firefly, fireflies’) in O. Gewargis Ashitha, Hilqa de Leshana: Assyrian-Arabic Dictionary (Dohok: n.p., 2nd edn, 2018) p. 127a, with the component wāʕnē related to both the vernacular term wāne ‘ewes’ and the entry ʕānā ‘small cattle’ (ibid., p. 649a), derived from Syriac. See Duval, Bar Bahlul, vol. 2, p. 1850, n. 23. See, inter alia, R. Payne Smith et al., Thesaurus Syriacus (2 vols.; Oxford: Clarendon Press, 1868–1901) (henceforth: Thesaurus) col. 3757 and LS, p. 700a. The lemma is vocalised qarṣʕānā in Thesaurus, col. 3757. A different vocalisation, qarṣaʕnā, rendered in Arabic qarṣaʕna(t), is offered in G. Cardahi, Al-Lobab, seu Dictionarium Syro-Arabicum (2 vols.; Beirut: Typographia Catholica, 1887–1891), vol. 2, p. 440a; cf. post-classical Ar. qirṣaʕna(t), qirṣaʕanna(t) ‘Eryngium, a kind of thorny plant’ (inter alia in J. Hava, Al-Faraid: Arabic-English Dictionary for the Use of Students [Beirut: Catholic Press, 2nd edn, 1915] p. 590b); and similar forms with gemination of n in Levantine Arabic (inter alia in Frayha, Dictionary, p. 138a, s.v. qirṣaʕanne; C. Denizeau, Dictionnaire des parlers arabes de Syrie, Liban et Palestine [Paris: Éditions G.-P. Maisonneuve, 1960] p. 412; s.v. qurṣeʕanne; Barghouthi, Dictionary, p. 991, s.v. qurṣaʕanne), each being an Aramaic substrate word (Frayha, Dictionary, p. 138a). Aramaic Studies 16 (2018) 215–233 folk etymology in the north-eastern neo-aramaic dialects 229 interpretation of each cognate as ‘eye-pincher’, involving a form akin to the verb qrṣ, qārəṣ (J. Nerwa, Betanure) or +qrs, +qarəs (Sat) ‘to pinch’. Indeed, my J. Nerwa and Betanure informants offered a synchronic interpretation of qarəṣena, qaraṣena as ‘eye-pincher’ (although, as is well known, nettles do not ‘pinch’ the eye); and it is very likely that the changes that shaped +qorasayna in Sat were also initially induced by the same folk-etymological association. In Sat the form begins with +qoras rather than +qares, however, and my Sat informants could offer no synchronic interpretation of this word except for its denotation ‘nettle’. Nonetheless, it may well be that in Sat the earlier form was *+qārəs-ʔayna ‘eye-pincher’, and that the form +qorasayna, with unexplained vocalic changes, arose after the folk-etymological association of nettle as ‘eye-pincher’ had been forgotten. 9. xwar-garme ‘bone-eater’ > xwaɻ-jaɻme ‘Egyptian vulture’ < xwāɻa ‘white’ A case of a very plausible folk-etymological process is the Ṭyare compound xwaɻ-jaɻme ‘Egyptian vulture’. Its literal meaning, ‘white-boned’, is synchronically transparent, comprising xwaɻ ‘white’52 and jaɻme ‘bones’ (which is how the compound is construed by native speakers). However, it is questionable whether xwaɻ-jaɻme can historically be analysed as ‘white-boned’. After all, it is very doubtful whether Egyptian vultures are distinguished by particularly pale bones, and, at any rate, their bones could have been visible only on the rare occasions when a carcass of an Egyptian vulture was seen. It seems much more likely that xwaɻ- stems from Kurdish xwar ‘eater’, a derivative of xwarin ‘to eat’, and that the erstwhile compound was xwar-garme ‘bone-eater’, since Egyptian vultures, as scavenging carrion birds, scrape meat off bones of carcasses, which may seem like eating bones when seen from afar. Indeed, this hybrid compound already appears in Rhétoré’s grammar of Soureth as ḵwār garmē (in Syriac script), which is glossed: ‘a bird that eats (Pers. ‫ )خوار‬bones’.53 Assuming that *xwar-garme ‘bone-eater’ is the correct etymology, ‘whiteboned’ should be considered a meaning induced by folk etymology, based on 52 53 For the use of xwaɻ as an apocopated, neo-construct form (hardly the old construct state) of xwāɻa, consider Ṭyare xwaɻ-dəqna ‘an elder’, lit. ‘white-beard’. J. Rhétoré, Grammaire de la langue Soureth ou Chaldéen vulgaire (Mossoul: Imprimerie des pères Dominicains, 1912) p. 194. Other such hybrid compounds in Ṭyare involve the Kurdish word sar ‘head’: sar-ʔamoḏa ‘the first child to be baptised among children brought to church’ and sar-qənna ‘firstborn child’. Aramaic Studies 16 (2018) 215–233 230 mutzafi the white or pale colour of most of the bird’s plumage, as well as on the near homophony of xwar ‘eater’ and xwaɻ ‘white’. Note also that in the change from *xwar-jaɻme to xwaɻ-jaɻme the compound was fully nativised, as foreign xwar was replaced by native xwaɻ with a retroflex r, a sound virtually exclusive to inherited NENA words.54 10. Pers. gerd-bɔ̄ d > ɟarda-bulə ‘whirlwind’ < ɟarda bulə ‘scraper of ears of corn’ The literal synchronic denotation of C. Urmi ɟarda-bulə ‘whirlwind’ is ‘scraper of ears of corn’, it being a compound that consists of the verbal form ɟarda ‘it (3sg.f.) scrapes, tears off’ and the noun bulə ‘ears of corn’ (sg. bula).55 ɟarda-bulə is most likely a folk-etymological alteration of the slightly similar sounding Persian compound gerd-bɔ̄ d ‘whirlwind’, which comprises the components gerd ‘round, around’ and bɔ̄ d ‘wind’. Thus a foreign lexeme has been fully integrated into the native lexical system of C. Urmi. The credit for connecting the C. Urmi and Persian compounds belongs to G. Khan,56 and note also the remark in Maclean’s dictionary (1901: 56a) that ɟarda-bulə ‘whirlwind’ is said to be prevalent in harvest time (viz. when bulǝ ‘ears of corn’ are harvested). 54 55 56 For this and other rhotic phonemes in Ṭyare see H. Mutzafi, ‘The Three Rhotic Phonemes in Ṭyare Neo-Aramaic’, AS 12 (2014), pp. 168–184. The same ɻ occurs also in jaɻme ‘bones’, which, incidentally, evinces initial ga > ja as in a few other Ṭyare words, e.g. *gammūrā > jamuɻa ‘tart’, *gaḇrā > jawṛa ‘man, husband’. For bulə in context see G. Khan, The Neo-Aramaic Dialect of the Assyrian Christians of Urmi (4 vols.; Leiden: Brill, 2016) (henceforth: Khan, C. Urmi), vol. 4, p. 334/11, and for the full semantic scope of this noun see Khan, C. Urmi, vol. 3, p. 106. The direct antecedent of the sg. form bula is bola, which is preserved in the meaning ‘panicle’ (head of maize, sorghum, millet, reed) in various C. NENA dialects of northern Iraq and Turkey, such as C. Aradhin and Ṭyare. The form bola itself developed from bolla in the same meaning, preserved in some of the C. NENA dialects of the area Mosul; and bolla is presumably a reflex of mbolla, a participial form related to the verb mbalbole (inter alia) ‘to dishevel’, so that mbolla had initially the sense of ‘dishevelled’ or ‘fluffy’. See Khan, C. Urmi, vol. 3, p. 162; and cf. Maclean, Dictionary, p. 56a, where the C. Urmi lexeme is said to be either related to [gərd in] Neo-Aramaic gərdab ‘whirlpool’ < Persian and Turkish ‫‘ گرداب‬id.’ (read: Pers. gerdɔb, Turk. girdap) or derived from *gārdaṯ bule ‘scraper of ears of corn’. Aramaic Studies 16 (2018) 215–233 folk etymology in the north-eastern neo-aramaic dialects 3.3 231 Possible Further Cases of Folk-Etymological Alterations of NENA Words 11. Akk. eṣemṣēru, eṣenṣēru ‘backbone’ > sənsāra, sənsārət xāṣa ‘id.’ < nasāra ‘saw’ As was suggested earlier,57 the Akkadian noun eṣemṣēru ‘backbone, spine’, a compound of eṣem ‘bone of’ and ṣēru ‘back’, surfaces as the substrate word sənsāra ‘backbone, spine’ in various NENA dialects of northern Iraq, e.g. Iṣṣin, C. Aradhin, Tisqopa, Telkepe, Betanure, J. Dohok and J. Zakho. In all these dialects sənsāra is freely interchangeable with sənsārət xāṣa (in Betanure, J. Dohok and J. Zakho also with sənsā ́r xāṣa), lit. ‘spine of the back’. The Akkadian word may well have entered Aramaic through its by-form eṣenṣēru, whence, assumedly, the pre-NENA or proto-NENA antecedent *ṣenṣērā, with aphaeresis of e in an unstressed open syllable (cf. Akk. esittu ‘mortar’ > late Syr. settā ‘id.’, NENA sətta ‘stone mortar’). Postulating the precursor *ṣenṣērā, its NENA reflex sənsāra necessitates the elucidation of some phonological changes, most notably the highly irregular loss of pharyngealisation (*ṣ > s) and the vowel change *ē > ā. The other changes, *e > ə in a closed syllable and word-final *ā > a are trivial NENA processes. Since there appear to be no phonological motivations for the changes *ṣ > s (such as a contiguous y or i) and *ē > ā, I should like to suggest folk-etymological association with nasāra, nasarta, masāra (and other dialectal cognates) ‘saw’ as a plausible cause for these changes. Note, in this connection, that in some NENA dialects nouns denoting ‘saw’ are used in genitive phrases with xāṣa ‘back’ to designate ‘spine, backbone’, e.g. Bebede nasārət xāṣa, Ashitha masāɻət xāṣa, Harbole masārət xāṣa, Marga māsārət xāṣa, Isnakh māsorət xāṣa. It follows that, quite possibly, *ṣenṣērā changed into sənsāra by folk-etymological association with nasāra (and cognates) ‘saw’,58 and that the phrase sənsārət xāṣa was synchronically interpreted as ‘saw of the back’. An alternative etymology would be sənsāra as a borrowing of a colloquial Arabic form related to Ar. silsila(t) ‘chain’, conflated with nasāra ‘saw’. Such a form is manifest in Jewish Baghdadi Arabic as sənsūl ‘spine’,59 which is probably 57 58 59 H. Mutzafi, ‘Akkadian Substrate Words and Meanings Surfacing in Neo-Aramaic’, Brill’s Annual of Afroasiatic Languages and Linguistics 10 (2018), pp. 24–51 (33–34). The change *ē > ā was perhaps reinforced, moreover, by analogy of *ṣenṣērā with words in the pattern CəCCāCa, e.g. Telkepe qənyāna ‘equid or bovine farm animal’, xəmyāna ‘father in-law’, məlxāwa ‘winnowing fork’, qərjāla ‘crab’ (< Kurd.). See Y. Avishur, A Dictionary of the New Judeo-Arabic Written and Spoken in Iraq (1600– Aramaic Studies 16 (2018) 215–233 232 mutzafi etymologically related to sə́nsəla ‘chain’ in the same Arabic variety.60 Consider also the fact that in some C. NENA varieties the term for ‘spine’ is literally ‘chain of the back’, viz. šišəltət xāṣa (e.g. in Alqosh, Karimlash, Tkhuma-Gawaya), šišəltət +xāsa (e.g. in Baz, C. Salmas, C. Urmi), šišəltət +xesa (in Sat). However, in NENA dialects that exhibit a clear borrowing of Ar. silsila(t) the meaning is ‘progeny, descendants’, as in J. Zakho sənsəlta, or, similarly, ‘lineage, lineal descent’, as in Tisqopa sə́nsəla. 12. Kurd. mir̄emir̄ ‘muttering, mumbling’ > miɻi-miɻe ‘gossip, rumour’ < ʔmɻ ‘to say’ Various NENA dialects evince reduplicative compounds that comprise two preterite forms of the verb ʔmr, ʔāmər ‘to say’: miɻi-miɻe (Ashitha) ‘gossip; rumour(s)’, lit.: ‘I said-he said’, mərri-mərru (Betanure, J. Nerwa) ‘id.’, lit.: ‘I said-they said’, məre-məre (Naghada) ‘id.’, lit.: ‘he said-he said’, mərri-mərrox (J. Zakho) ‘argument, hostile exchange of words; gossip, rumour(s)’, lit.: ‘I saidyou said’, məri-mərox (J. Urmi) ‘gossip; argument, debate’, lit.: ‘I said-you said’. It may well be that the origin of all these compounds is Kurdish mir̄emir̄ ‘muttering, mumbling’ (also ‘growling, snarling; purring, humming, grunting’) and possibly also the Kurdish verb mir̄mir̄în ‘to mutter, to mumble’. The Kurdish verb was likely initially altered (semantically and structurally) by folketymology in some NENA dialect or dialect cluster, and the compound that emerged diffused to other dialects. Consider, in this connection, Maclean’s entry ‘mirâmir’, defined as ‘growling, murmuring; whispering, esp. talking scandal’.61 It seems that this entry corresponds to C. Urmi +mərra-+mər, rendered ‘whining sound’ in Khan’s description of this dialect,62 and that the etymology is Kurdish mir̄emir̄. However, one cannot rule out the possibility that miɻi-miɻe and similar compounds have emerged in NENA without any connection to Kurdish mir̄emir̄ and mir̄mir̄în. Compare, in this connection, Ar. qāl wa-qīl ~ qīl wa-qāl ‘long palaver; idle talk, prattle, gossip’ (lit.: ‘he said and it was said’ ~ ‘it was said and he said’).63 60 61 62 63 2000) (3 vols.; Tel Aviv-Jaffa: Archaeological Center Publications, 2008–2010 [in Hebrew]) (henceforth: Avishur, Dictionary), vol. 3, p. 148. See Avishur, Dictionary, vol. 3, p. 148. See Maclean, Dictionary, p. 199b, with a Turkish etymology ‫ مرمر‬that I could not establish. See Khan, C. Urmi, vol. 3, p. 236. See Wehr, Dictionary, p. 797b. Aramaic Studies 16 (2018) 215–233 folk etymology in the north-eastern neo-aramaic dialects 4 233 Conclusion The foregoing passages discuss twelve selected cases of folk etymology in some of the NENA dialects. Among these are three folk-etymological interpretations that did not trigger structural or other changes (see nos. 1–3 in the synopsis below) and nine are cases (and possible cases) of changes induced by folk etymology (nos. 4–12 below). Most lexical items involved are inherited NENA words, including one case of an Akkadian substrate word surfacing in NENA (no. 11 below). The rest are of Kurdish or Persian origin (nos. 3, 10, 12 below), as well as one case of a Kurdish-NENA hybrid compound (no. 9 below). The aforementioned cases are summarised in the following synopsis: Etymon 1. 2. 3. 4. 5. 6. 7. NENA words affected by folk etymology ʕezqəṯā temrē Kurd. ferman sāḥōrā/*sāḵōrā narqōs dlá-lwāṯi *bar-zīwānā sqəltá ‘ring’ < sqəltá ‘beautiful’ təmre ‘eyelids’ < təmre ‘dates’ parman ‘massacre’ < prm ‘to slaughter’ sixora ‘stye’ < sixora ‘porcupine(’s quill)’ marqus ‘narcissus’ < marqus ‘St. Mark’ dlá-nāṯi ‘unbeknownst to me’ < nāṯa ‘ear’ baɻza-wāna ‘firefly’ < baɻza wāna ‘the ewe dries up (stops lactating)’ 8. *qarṣʕānā qarəṣena ‘nettle’ < qārəṣ ‘it pinches’, Ɂena ‘eye’ 9. Kurd. xwar + Aram. garmē xwaɻ-jaɻme ‘Egyptian vulture’ < xwāɻa ‘white’ 10. Pers. gerdbɔ̄ d ɟarda-bulə ‘whirlwind’, lit.: ‘scraper of ears of corn’ Possibly also: 11. Akk. eṣemṣēru 12. Kurd. mir̄emir̄ sənsāra, sənsārət xāṣa ‘backbone’ < nasāra ‘saw’ miɻi-miɻe ‘gossip, rumour’, lit. ‘I said-he said’ < ʔmɻ ‘to say’ Aramaic Studies 16 (2018) 215–233