A Fresh Analysis of the Origin and Diachronic Development of "Dialectal Tanwin" in Arabic.

AuthorStokes, Phillip W.
PositionReport
  1. INTRODUCTION

    Scholars of the history and development of Arabic are becoming increasingly aware of, and interested in, features attested in contemporary Arabic dialects that can illuminate the history and development of Arabic. One of the most discussed, and debated, is the morpheme called "dialectal tanwin" (henceforth DT). DT refers to the morpheme, typically realized as in or an, that is suffixed to a morphologically indefinite noun, primarily when followed by some type of adnominal adjective or clause, e.g., bint-in zena "a pretty girl" (Holes 2016: 131). This distribution of DT is attested historically in texts from Andalusia ([Corriente] 2013), as well as in Judaeo-Arabic texts, with attestations dating at least as early as the beginning of the ninth century (Blau and Hopkins 2017: 382). (1) In the Tihama region of SW Saudi Arabia and Yemen, a similar phenomenon, also referred to as DT, occurs, but with a different distribution (Behnstedt 2016: 64-67).

    Most scholars have believed that the morpheme is made up of a frozen case vowel and -n, equivalent with etymological tanwin, which was suffixed to syntactically unbound nouns that lacked the definite article. A few scholars have questioned such identification recently on the basis that the morpheme does not synchronically mark case, nor can it be reconstructed as having done so without relying on ClAr. Instead, they reconstruct it with its current function, sometimes dubbing it a "connecting morpheme" *-Vn (Owens 2006: 106; Holes 2011; Ferrando 2018: 111).

    Author's note: My sincere thanks to Na'ama Pat-El, Marijn van Putten, Ahmad Al-Jallad, Benjamin Suchard, Fokelien Kootstra, and three anonymous reviewers for their helpful comments, corrections, and suggestions on earlier versions of this paper. Any errors that remain are strictly mine. Abbreviations used through this text are: ADJ = adjective; ClAr = Classical Arabic: DT = dialectal tanwin; JA = Judaeo-Arabic; MAr = Middle Arabic; MSA = Modern Standard Arabic: N = noun; PP = prepositional phrase; QCT = Quranic Consonantal Text; RelCl = relative clause.

    I argue here for a new account of development of the morpheme. I will argue that the current distribution of DT in both non-Tihama and Tihama dialects is perfectly compatible with a derivation from etymological tanwin. We need not appeal to a separate, otherwise unattested morpheme to explain DT. However, I will argue that the prevalent reconstruction, laid out in detail by Blau (1981: 167-212), is incomplete and does not account for all available data in the most parsimonious and economic way. My proposal is that realization of DT represents a merger of the etymological case vowels following a loss of phonemic contrast, which itself possibly contributed directly to the breakdown of the case system. Further, I will make the case that the distribution of DT, which differs significantly between Tihama and non-Tihama varieties, should be understood against the background of the two patterns of pause attested in ClAr: prose and poetry. I conclude by contextualizing the foregoing discussion in the broader history of Arabic, including a discussion of the pre-Islamic epigraphic and early Islamic Arabic data.

    Section two reviews DT, relying on examples from a number of dialects, both historical and contemporary. The non-Tihama data are presented together first, broken down by syntactic function, followed by a brief review of the Tihama data. Section three reviews previous proposals in more detail, highlighting what are, to my mind, the weaknesses of each. Section four is dedicated to the presentation of a modified form of the traditional argument, which I hope will be more complete and, therefore, more convincing than previous attempts at explaining the origin and development of DT. The paper concludes with section five, which contextualizes the evidence historically.

  2. DATA

    DT is attested in both historical and contemporary data. In this section I will review the attested functions of DT, which, as noted above, are remarkably similar across time and space. In most contexts, the noun to which DT is suffixed is followed by an adnominal attribute. This attribute is virtually always an attributive adjective, and less commonly a prepositional phrase or a verbal clause. In a handful of dialects, DT occurs on nominal forms acting adverbially. A few other, more restricted functions will be discussed below.

    2.1. N-Vn +ADJ

    By far the most common context in which DT occurs in every available corpus is suffixed to a noun followed by an adnominal attributive adjective (N-Vn + ADJ). In Classical JA, in a few examples primarily from one text (High Ways), DT is written by means of the two kasras (-, )or two fathas (-' ) from Arabic orthography (Blau 1981: 173-74):

    (1) [phrase omitted] lam yadhab=ha iktisab=in 'ilmiyy NEG leave.IMPF.3ms=SUFF.3fs acquisition=DT scientific "And (an ignorance of) the acquisition of knowledge has not left it" Elsewhere, DT was written as a separate word [phrase omitted] /in/ or /an/ (Blau 1981: 175-76):

    (2) [phrase omitted] 'ila bilad an ba'ida to countries DT far "To faraway countries" The same practice of writing DT as a separate word is ubiquitous in the correspondence from Egypt and N. Africa found in the Cairo Geniza, dating from the eleventh to nineteenth centuries. The following examples are taken from Esther-Miriam Wagner's comprehensive study of these letters (2010: 178-79):

    (3) [phrase omitted] wa=nahnu taht hal an 'azim and=we under state DT terrible "While we (were) in a terrible state" (4) [phrase omitted] nabi' min=hu say(y) an axar buy.IMPF.lCP from=SUFF.3ms thing DT other "We will buy another thing from him" The pattern N-Vn + ADJ is the only use of DT that Federico Corriente lists for Andalusian Arabic, dating from the eleventh century (2013: [section]3.1.1.1.1):

    (5) 'ays=an dank life=DT miserable "Miserable life" (6) masayib=an 'izam disgraces=DT great "Great disgraces" (7) (Ibn Quzman, twelfth century, apud Ferrando 2018: 96) wajh=an malih wa=sarrab=an asfar face=DT nice and.drink=DT yellow "A beautiful face and a golden wine" This function of DT, marking a noun followed by attributive adjectives, is the most common one in the modern Arabic dialects. While especially concentrated in the Arabian peninsula, it is attested in dialects from across Anatolia and Mesopotamia, the Levant, sub-Saharan Africa, and even the isolated pockets of Arabic speakers in Central Asia:

    Afghanistan Arabic (Ingham 2006: 30)

    (8) fad gapp=in mahqul one speech=DT reasonable "Reasonable words" (9) zagir=id (2) darvis small=DT dervish "A dervish child" Uzbekistan Arabic (Zimmerman 2009: 621-22)

    (10) bayt=in kabir house=DT big "A big house" (11) mu=hin (3) ahmar water=DT red "Golden water" Bahraini Bahama Arabic (Holes 2016: 131-32)

    (12) bint=in zena girl=DT beautiful "A beautiful girl" (13) may=in barda water=DT cold "Cold water" Omani Arabic (Holes 1996: 47-48)

    (14) in kan naqt=in zena if to be.PERF.3ms female camel=DT good "If it's a good female camel" (15) u ida kan ksur=in katira and if to be.PERF.3ms broken bones=DT many "And if many bones are broken" Najdi Arabic (Ingham 1994: 48)

    (16) ja=na harbiyy=in tuwil come.PERF.3ms=1cp Harbi=DT tall "There came to us a tall Harbi" (17) beet=in kibir house=DT large "A large house" Dialects of SE Najd, such as the dialect of Al Murra, suffix the DT to the attributive adjective as well (Ingham 1986: 280):

    (18) rena b'ir=in 'od=in see.PERF.1cp male camel=DT large=DT "We saw a large male camel" Sudanese Arabic (Owens and Hassan 2009: 711-12)

    (19) rdjil abu watir=an hamra man father of car-DT red "A man with a red car" (20) rajl=an sum man=DT nasty "A nasty man" (21) saret sayyarat=in yadideh buy.PERF. 1cs car=DT new "I have bought a new car" (22) yabal=in aswad mountains=DT black "Black mountains" A few SW Saudi dialects, most notably that of Rijal Alma', pattern with SE Najdi dialects (see ex. 21 above), and attach the DT to the following attributive adjective as well (Asiri 2008: 72):

    (23) tahnah bint=in tayyibat=in that girl=DT good=DT "That is a good girl" 2.2. N-Vn + Attributive Verbal Clause

    DT is also attested suffixed to a noun followed by an attributive verbal clause. Although this pattern is decidedly less common across the contemporary dialects than N-Vn + ADJ, it is nevertheless attested both historically in JA as well as in contemporary dialects.

    Judaeo-Arabic

    (24) [phrase omitted] sahid an yahki witness DT tell.IMPF.3ms "A witness who is telling" (25) (Wagner 2010: 178-79) [phrase omitted] wa=dhud=ak ma'=l li=kull bayt and=take.IMPF.1cs=SUFF.2ms with=SUFF.1cs to=every house an aruh li=h DT go.IMPF.1cs to=it "I will take you with me to every house I go into" Andalusi Arabic (Ibn Quzman; Ferrando 2018: 101)

    (26) sufayfat=an yatul fi=hd l=i'tibar small lips=DT be long.IMPF.3ms in=3fs DEF-pondering "Small lips that would be pondered long" (27) ila yawm=an yulqa 'alayya t-turab until day=DT be thrown.IMPF.3ms upon=1cs DEF-dust "Until the day they cover me with earth" Bahraini Arabic (Holes 2016: 132)

    (28) arad=in bayya'=ha lands=DT sell.PERE3ms=3fs "Lands that he sold" Najdi Arabic (Ingham 1994: 52)

    (29) awwal siggit-in sift=aha first flat=DT see.PERF.1cs=3fs "The first flat I have seen" (30) ra'i siggit-in kallamt=ih owner flat=DT talk.PERF. 1 cs=3ms "The owner of a flat I have spoken to" Afghanistan Arabic (Ingham 2006: 34)

    (31) darwis=in suft dervish=DT see.PERE.1cs "I saw a dervish" (or "A dervish whom I saw") (32) wazir=in ko 'ind=u minister=DT to be.PERF.3ms at=3ms "He had a vizier" (or "a vizier whom he had") 2.3. N-Vn + PP

    The final usage, attested in both historical and contemporary Arabic varieties, is to mark a noun modified by an attributive prepositional phrase.

    Judaeo-Arabic

    (33) (Wagner 2010: 39) [phrase omitted] la ta'budun rabb an sawa=y(a) NEG worship.IMPF.2mp lord DT except=SUFF.1cs "Do not worship another God beside me" (34) (T-S 8J39.12/rm.16f; Blau 1981: 175) [phrase...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT