Notes
I am grateful to the Dutch Organisation for Scientific Research (N.W.O.) for their generous financial support of a four-year research into the bedouin dialects of the northern Sinai littoral.
For more details, cf. DE JONG, 1995. :
For the approximate distribution of the bedouin tribes in northern Sinai, cf. DE JONG, 1995, p. 105.
BLANC, 1970, p. 4, distinguishes three main styles: "plain colloquial", "koinized colloquial" and "poetic idiom", STEWART, 1990, preface, p. x, adds a fourth: "the style peculiar to pleadings and summings up".
Cf. for instance the poem in BAILEY, 1991, p. 140, which the author heard recited, among others, by Xlêf Mgêbil ilHirš, a chief of the Hrûš-clan of the Biyyâdiyya in Râb'a in the northwest of Sinai. The poem (11 lines) shows quite a few dialect features which are definitely not part of the Biyyâdi plain colloquial, such as: nunation (10 instances), non-emphatic interdentals (6 instances), the article al-, instead of proper Biyyâdi il-(25 instances), the preposition b followed by a suffix has a short base (bi+ - ib+), rather than the regular Biyyâdi long base with e (be+) (2 instances in l.1 and 6), and also a masc. 3rd pers. sg. pron. suffix -a(h), instead of Biyyâdi -u (1 instance in l. 10), The question remains how much of the poem, such as it appears here, was actually heard from the mouth of Xlêf. For a description of Biyyâdi Arabic (the dialect of the Hrûš-clan), cf. DE JONG (forthcoming).
Cf. TRUDGILL, 1983, p. 143; "As is well known, different situations, different topics, different genres require different linguistic styles and registers." This is not to say that such topics were not included, or even avoided in the research, but merely that extra caution was observed in interpreting forms occurring in texts where such topics were discussed.
Preface, p. x-xi.
In terms of geographical distance, the Masâ'îd and 'Ayâyda are nearer, but these tribes are still semi-nomadic.
One might say that only (sub-) conscious choices are involved during the first phase, i.e. the learning phase of this process (short-term adaptation).
Cf. TRUDGILL, 1986, p. 107.108 (paraphrased).
Cf. TRUDGILL, 1983, chapter 5.
Cf.TRUDGILL.1986, p. 62.
Cf. DE JONG, 1996, p. 61. The examples chosen here are loanwords from European languages: Fr. téléphone and direction, and Eng. guinea. We can be quite certain, therefore, that the Fayyûmi diphtongs in these examples are not remnants of older Arabic diphtongs, but are the result of phonological reinterpretation of the monophtongs e and ô in CaA.
Upper Egyptian gineh is another example of such an interdialect form; speakers "knew" that CaA /g/ corresponds to /ğ/ in their own dialect, and substituted /g/ for /ğ/ accordingly.
The short a's in the first two syllables of such forme as talafawn and daraksawn are to be interpreted in terms of the preferred palterns for loanwords from other languages a-a-vCC and a-a-vC, cf. WOlDICH, 1990a, pp. 147-148.
TRUDGILL, 1983, p. 45-46, draws our attention to an important aspect of the "observer's paradox": the production of "hyperdialectisms" as an unwanted effect of informant's well-disposedness towards what he considers (or would have to be) his original dialect, and towards the dialect survey itself, especially in unnatural language-situations created by the use of questionnaires (direct elicitation).
Often enough, during direct elicitation, informants would state that a certain form would not occur in their own dialect, but in one of the other dialects, and they were usually right. Cf. also MURRAY, 1936, p. 256-257 (about bedouins in southern Sinai): "Among themselves, they can distinguish each tribe and subtribe by their looks and dialects [...]". This holds for bedouins in northern Sinai as well.
The question was justified, as this informant had produced several instances of stressed articles on other occasions, and especially when answering to a questionnaire, which I had prepared for this research.
hadar: settled, as opposed to nomadic population: 'arab.
TRUDGILL, 1986, p. 10-11, speaks of imitation in this case, because English popsingers do not accommodate to the dialect of their (English) audience. On the other hand, we might as well speak of accommodation here, as the singers actually do accommodate to what is expected from them by their audience: to sing in an accent that goes with the genre.
To mention just a few of the many cultural manifestations, this is illustrated by customs pertaining to tribal law (cf. STEWART, 1987e, p. 480), wedding customs (cf. BAILEY, 1974b), and the tradition of oral poetry (cf. BAILEY, 1991). And also, more visibly in daily life, the traditional dress of the bedouins, their professional activities (raising goats and sheep), as well as their agricultural tools, such as the plough with the typical funnel-shaped implement mounted on the sole, or the hôgal, a threshing board, rather than the nôrag one finds in the Egyptian Delta. Clearly, all these are not of the Egyptian type, and neither are their dialects.
In a linguistic sense, the influence of a more Syro-Palestinian culture is already apparent in the Šarqiyya governorate. BEHNSTEDT/WOIDICH, 1988, p. 264, state "stehen die Dialekte der Šarqiyya denen Oberägyptens, ja sogar denen des syrisch-palästinensischen Raums, näher als denen des übrigen Deltas". A similar observation was made by ABUL FADL, 1961, p.5.
The dialects of northern Sinai, and perhaps all of Sinai (except that of the Dawâğra), form the western branch of the North West Arabian group as identified by PALVA, 1991.
Even a process of koineization affecting the dialects initially forming such a patchwork, which resulted eventually in a more homogeneous dialect area does not seem all that implausible. For instance, certain stress patterns found in the northwest today point towards an older stress type still present in the northeast (cf. below).
Cf. PALVA, 1991, p. 154 and 160, where a relative homogeneity of the NWA dialects in earlier times is hypothesized.
Cf. DE JONG, “A description of Biyyâdi Arabic” (forthcoming),
The Axârsa are (fully) settled along the same main road through northern Sinai as the Biyyâdiyya, and are actually geographically some fifteen to twenty-five kilometers closer to Egypt proper than the Biyyâdiyya. So the fact that their dialect trails Biyyâdi in this development towards a more sedentary Delta type is a bit puzzling. It must, in part at least, have to do with the high degree of education. AT-TAYYIB, 1993, p. 611, reports that the Axârsa are very well trained, and are second in this respect only to the Biyyâdiyya. He writes that both tribes have many relatives in Egypt proper, but one important difference is that many Biyyâdi families emigrated to the Tahrîr province in 1967,(p. 607), where they stayed during the Israeli occupation of Sinai, after which many families returned to Sinai. This type of exile is not reported for families of the Axârsa. It may also have to do with the long history of the Biyyâdiyya in Sinai; they are the but one oldest tribe in northern Sinai, and are reported to have arrived in Sinai somewhere between the 7th and 10th centuries, and to have been in the northwest (in the Gatya oasis) ever since the 13th century, which is longer than the Axârsa, who came between the 10th and 13th centuries (cf. BAILEY, 1985, p. 47 and 21). It is therefore fair to assume that the Biyyâdiyya have been exposed to Egyptian Delta type dialects longer than the Axârsa. AT-TAYYIB, however, documents (p. 610) that they came from Syria together with the Biyyâdiyya in the seventh century Higra, i.e. the period of the crusades against Salâh ad-Dîn.
Or more precisely, OD 3 (Eastern Dialect 3 spoken in the southwestem Šarqiyya and south and centre and southeast of Qalyûbiyya); cf. WOIDICH, "Rural Dialects of Egyptian Arabic" (appearing in this volume).
The setling for the interviews during this research was usually a mag'ad.
For difficulties involved in the imitation of phonology, cf. TRUDGILL, 1983, p. 11.
With this word 'urfi the speaker may have been refering to an older version of his own dialect, but in the Sinaitic context the word 'urfi much more refers to a cultural background which members of the different tribes feel to have in common. And clearly, in this case, the two notions overlap.
Although TRUDGILL, 1983, p. 31 -32, warns us to be careful with conclusions on dialect differences between generations of speakers in "apparent time", our conclusion seems justified here; the other possibility of the different reflexes of */ and *d, would be for the grandfather to have learned at a later stage in his life to substitute t with t, and d with d to, as it were, "make" his speech more 'urfi. But then one would expect imperfections, or maybe even hyperdialectisms (cf. ibid. p. 11-13 and 149-150). Such instances of "imperfect learning", however, were not observed, Where instances of t and d for *t and *d did occur in grandfathers speech, this could generally plausibly be ascribed to the koineizing influence of CaA.
Cf. BEHNSTEDT/WOIDICH, 1985b, maps 59-60.
Cf. ibid. map 59.
Cf. ibid.
Cf.WOIDICH, 1979, p. 81-83.
The point is that underlying forms need not be historical forms in generative phonology.
The reference is to primary word stress here, not to primary stress in word groups.
The rule here is that high vowels are not deleted when they occur between consonants C1 and C2 which are phonetically identical, or close, i.e. in vC1C1iC2v; Cf.WOIDICH, 1979, p.79.
The conclusion of a stress shift in eastern Šarqâwi was also drawn by ABUL FADL, 1961, p. 222, and GROTZFELD, 1969, p. 160.
Also elision of the high vowel in such positions would lead to forms which are not acceptable because they are no longer morphologically transparent; triple consonant-clusters in forms (after elision of i) like *yğadddu or *yhadddu, would probably be reduced to double consonants *yğaddu and *yhaddu (cf. WOIDICH, 1979, p. 79: *dann + na —> danna), which would make them look like 1st measure verbs, and that would, again, disturb morphological transparency.
The dialects spoken by the Dullâm (cf. BLANC, 1970), and the Rumelât and Sawârka form a homogeneous typological group; cf. DE JONG, 1995.
I.e. in the dialects of group 1, cf. ibid.
Cf. BLANC, 1970, p. 5 (ll6).
Cf. ibid., p. 10 (121).
Such forms are found in the dialects of the Axârsa, Samâ'na and 'Agâyla as well.
This velarized and labialized - *k, and the fem. - ik or - a, is certainly a feature of a dialect type spoken in the south of Sinai. NISHIO, 1992, p. 178-179, reports similar forms among the Ğbâliyya [-ok ~ -ku (masc. sg.), and -k ~ -ek (fem. Sg.)), and Manfred Woidich heard forms like 'alê'k, "on you (masc. sg.)" and ‘alê’kum, "on you (masc. pl.)" among the Garârša in Wâdi Ferân (personal communication). (BAILEY, 1991, map on p. 4, gives the position of the Garârša in Sinai at the beginning of this century). This vowelless *k (the superscript * is merely a notational device indicating labialization and velarization of he following k) must be the result of a back formation of the pl. forms: first the * in masc. pl. 'alêkum was labialized and velarized -> alê‘kum, as opposed to k in fem. pl. 'alêkin. Then -um and -in, like the verbal suffixes (e.g. dârabum/dârabin, they hit, masc./fem.) were interpreted as the pl. morphemes. Then, when k had become stable in this position, it acquired phonemic status in a new symmetric opposition 'alê*k - 'alêk.
I.e. battîx sêfi, which is grown in a so-called sefîyya, a "watermelon field".
Although younger speakers are aware that it is a feature of the speech of their elders, and they can reproduce it when specifically asked, it no longer occurs in their own spontaneous speech. In fact, in the majority of cases the grandfather also produced -ak, and it was only when he became tired that the older -*k suffix appeared.
The suffix is actually vowelless, thus creating a final consonant cluster, and attracting stress to the vowel immediately preceding, and in the case of 'ilbtiik, “your tin/packet” , and not *'ilibtik. And also the unrounded 2nd pers. sg. fem. suffix is vowelless: 'ağâbk. "he pleased you (fem.)", and ragabâtk, "your (fem.) neck".
Cf. BLANC, 1970, p. 130-131 (19-20), and DE JONG (1995), appendix, row 14. Although, in the village of asSama‘na in the Delta, the suffix -ik, when following a consonant, appears in the text of ABUL FADL, 1961, p. 134: 'allah yixrib bêtik ya ba'îda! (my transcription), and -ki when following a vowel; cf. ibid. p. 133: middîki; I assume that yixrib bêtik, "may He destroy your house", is too much of a koinized expression (one may hear this throughout Egypt) to draw any definitive conclusions here. Generally, we have invariable -ki in the eastern Šarqiyya as well; cf. BEHNSTEDT/WOlDICH, 1985b, map 152.
The youngest generation is increasingly being exposed to CaA, mainly in primary schools where teachers are often speakers of CaA. Furthermore, it should be noted that many members of the northwestern tribes (who are now in their mid thirties and forties) of Sinai, spent a considerable number of years in Egypt proper during the Israeli occupation, and often received an education there.
The grandfather tended to use the -um suffixes in sandhi, but -uw suffixes in pause, eg.: 'asân 'êh? yâthanum iw yâkluw: "to do what? (for them) to grind (wheat) and to eat".
In the System of orientation of the bedouins in northern Sinai, šarg is geographically south, and accordingly, gibli is west (although the gibla Js due south), ġarb is north, and šimâl is east. The explanation for this may be that many of the tribes in Sinai are originally from areas where the sea is to the west, i.e. the Gulf of 'Aqaba is west of the Higâz and the Mediterranean is west of Gaza. garrab, "going west", thus became synonymous with "going into the direction of the sea", and the opposite, šarrag, “going east”, with "going inland", and in this meaning the words moved with the tribes to nortern Sinai, where the Mediterranean is to the north, and where in daily speech the compass was turned 90 degrees clockwise. I am grateful to Fred Leemhuis for this suggestion. In addition, it must be remembered that many bedouins regularly traveled to Palestine to help out during harvest time. If they came from the Higâz in the old days, they were mšammlîn "going north'. Until the creation of the state of Israël many tribes would undertake the same journey (geographically going east) from northern Sinai. If then mšammil had already become synonymous with "going to Palestine", a similar 90 degrees turn of the compass in daily speech could have taken place, which leads to the same conclusion.
They have settled in a village by the same name in the eastern Delta; cf. village 92 (asSama'na) in ABUL FADL, 1961, p. 12, and the text on p. 132-138. Cf. also AT-TAYYIB, 1993, p. 600-604, who mentions the presence of numerous Samâ'na in this village (in the district of Fâqûs), and also in other villages.
During one session, when the old man was rather tired, I did not record any prefixes with i for a-type verbs.
raza’:. "compose rhymes" ; cf. BAILEY, 1981-1982, p. 139.
The z in (i)nzâm is due to the word being a loan from CaA or CA, therefore not *(i)ndâm. Typically, however, the syllabication is proper Smê'ni, therefore not (CaA) nizâm. Cf. TRUDGILL (1986), p. 16 (quoting Broselow, 1984), on constraints in accommodating syllable structure.
A mêfa is a "cylindrical clay oven dug into the ground, used to bake bread"; Cf. LANDBERG, 1920, mim -yâ '-fa'- 'alif maqsûra in vol. l ( 'alif- dâl), p. 81 (root * -f-y).
TRUDGILL, 1986, passim, points out that salient features are the first to be subject to change as a result of dialect contact, provided they are not "too salient, i.e. too much of a stereotype, in which case salience may inhibit change. Thus, when quoting other Bedouins, they will often successfully imitate salient features of the target dialect, but fail to include the less salient features of the target dialect.
This was only a tendency, and it by no means occurred in all possible instances.
I have also noticed that one of my best informants, a born and bred 'Arâyši, with a wealth of experience in recording and writing down bedouin oral poetry, and with whom I have spent lengthy hours working on recordings, was actually unable to produce an interdental t or d, replacing them with s and z, instead of the regular t and d reflexes of his own dialect.
These examples were recorded in the house of the Axrasi S, on an occasion where three members of the Axârsa, two members of the Biyyâdiyya. and one member of the Ahaywât were present. The storyteller directed the story to all men present. The fact that the Ahaywi (whose dialect was of a pure northeastern type) was among the audience, no doubt, contributed to the notably high incidence of B-forms in this story.
Cf. DE JONG, 1995, appendix, field N-17.
Cf. BLANC, 1970, p. 144 [33], STEWART, 1990, glossary, p. 249.
This (phonetically conditioned) suffix - ih is not unique for the Negev (cf. PALVA, 1991, p. 157), it occurs in the dialects of the Sawârka, Rumêlât (northern and southern), Tarâbîn, Ahaywât (cf. STEWART, 1990, passim), Bilî, Masâ'îd, 'Ayâyda, and of the Dawâġra as well, However, the 'Agâyla, Samâ'na, Axârsa. Biyyâdiyya, Tuwara (cf. BLANC, 1970, p. 21 [132], fint 37), and Ğbâliyya (cf. NISHIO, 1992, p. 179) have -u (own observations, unless otherwise indicated).
xalihin is not the apocopated imperative it may seem lo be, and which one would expect in northeastern dialects (cf. BLANC, 1970, p. 26 [137]), and STEWART, 1987a, p. 48), but rather a grammaticalized adhortative partide comparable to CaA xallî-, which is not declined either, eg. xallîk. xallîki, and xallîku, cf. WOIDICH, 1995, p. 263.
In two instances the emphatic interdental d was recorded for*d: idra “sorghum” and ma dugtiš, "I did not taste".
Numerous Arabic dialects manage quite well without interdentals, so in this sense, one might say that the opposition between t and t, and d and d is potentially redundant.
Cf. TRUDGILL, 1983, chapter 5. The implication of this rather short chapter is far-reaching. The majority of instances of linguistic simplification and reduction could then be attributed to sociolinguistic contact situations, that is, they are externally motivated, which makes the changes "non-natural" in character. Other changes, such as "an increase in redundancy, and even a possible move towards a more synthetic structure" (cf. ibid., p, 105), are then internally motivated, and the changes are for that reason to be labelled “natural”.
This is a general rule, to which there are exceptions, of course. For instance, in Bahârna Arabic, the dialect of Bahrayn, d became d, d became d, but f became ƒ! Thus hâda, “this”, duhr, “aftemoon”, and ƒalaƒa, "three". And in the Siiri-dialects of Anatolia the three interdentals all have labio-dental reflexes, whereas the dialects of Âzax (of the Mardin dialects), and of Bahzâni (of the Tigris group) both have sibilants, cf. JASTROW, 1978, p. 35-39.
This notion must, for a good part, contribute to the high degree of salience of this feature, cf. ftnt 60. In this case, d proved too salient, too much of an identity marker in a positive sense, to be subject to change.
Unfortunately, I have not come across instances of reflexes of either *t or *d cooccuring in one word with d, tike in MSA diġt, "bunch, mixture", which might have been Biyyâdi *diġt, and which would have been an interdialect form in the litteral sense of the term.
Similar pressure is apparent in the dialect of the Ğbâliyya in southern Sinai; cf. NISHIO, 1992. To cite a few examples: the imperative for "take" among older speakers is (in my transcription) 'uxud, but for younger speakers it is xud (cf.p. 91). "To you (masc. sg.)" for older speakers is luk, but for younger speakers it is lîk (cf. p. 191). Original a-type perfect verbs seem to have become i-type lor the younger generation, e.g. faham (p. 111-112), nasal (p. 107), hafad (p. 112), ->fihim, nizil, hifid. And also interdental emphatic in bardu, "also", has yielded to d in the koine form bardu in the speech of the younger generation (on p. 194).
And they were directly along the old railway to Palestine; hence Râb'a, the “fourth” station on this raiiway.
Many acquired university degrees in Egypt proper during the Israeli occupation. EUROCONSULT, 1992, E. 19 reports that in the village of Nagîla 80 of the male inhabitants hold a university degree, and AT-TAYYIB, 1994, p. 608, writes that the Biyyâdiyya have through history had the highest level of education. He mentions that there are more than 20 doctors, many engineers, 18 university professors, several teachers, and army- and police oflicers.
ABUL FADL, 1961, p. 3, mentions the village Nôba w-adDahašna in Šarqiyya, where they have settled. AT-TAYYIB, 1993, p. 607 lists a number of other villages in Šarqiyya where they live today.
Geographically, Biyyâdi territory is nearer to the Šarqiyya than it is to Palestine, or the Negev. Before the digging of the Suez Canal, traveling west by camel (to work as farmhands during harvest time, or for trade purposes) must have been easier and quicker than traveling east; going to the Šarqiyya mainly involved crossing wide plains, whereas traveling to the east meant traversing a difficult dune-landscape.
The date of their arrival in Sinai must have been between the 7th and 10th centuries A. D., after which they initially settled more to the east, "near al'Arîš", but in the 13th century their presence in the Gatya oasis is attested. Cf. BAILEY, 1985, p. 47 and 21.
The Dawâġra are Htêm (or Hutaym, cf. E.I.), who are Mtêr (or Mutayr). Cf. e.g. OPPENHEIM, 1943, p. 140: "Die Dewâghera werden zu den verachteten Mutêr gerechnet." The story told (by members of other tribes) concerning their pariah status is that during the earliest times of islam, a Duwêġri wornan, named Ğarâda, was caught trying to smuggle a message hidden in her hair to warn the unbelievers among the Qurayš against an oncoming raid by the followers of the Prophet. The Prophet, however, was warned by the archangel Gabriel, and she was caught in the act. After this act of betrayal the whole tribe was from then on excluded from any type of equal relationship with the other tribes. The Qur'ân verses mentioned with respect to this story are Surah 60:1, 9, 13 (oral information from several independent sources in the field). AT-TAYYIB, 1993, p. 743-744, gives the same story, and remarks in a footnote that the lady's name was actually Sârah. KENNETT, 1925, p. 23-24, gives another unflattering story concerning their lowly status. Whatever the true story may be, the social isolation of the Dawâġra has been the result, and their dialect, which is of a clear northeast Arabian type, must have remained largely unchanged ever since their arrival in Sinai.
Cf. TRUDGILL, 1986, p.39-40.
Cf. DE JONG, 1995, appendix, As (or feature 2 a): one may hear (stressed, if in eligible positions) gahawa-vowels in nominals, but not in verbs: e.g. bâhar, "sea", and naxâlha, "her datepalms" , but more typically mâhirha, "her dowry", and 'âhilha, "her family". As for feature 4): it is difficult to establish the productivity of measure IV, but I have certainly recorded more instances of measure IV in northeastern dialects. As for (eature 6): Speakers of Biyyâdi are very well aware of such lexical items when checked with questionnaires, but I have not recorded any in spontaneous speech. As for feature 9): the word for "mouth" in Biyyâdi is xašm, not 'afâm.
With the exception of the dialect of the Dawâġra, cf. DE JONG, 1995, p. 5, and appendix, field K-38. Cf. BLANC, 1970, p. 28 for tribes in the Negev (and Sinai). PALVA, 1984a, p. 16, writes: “The dialects of the Negev Bedouins display some traits (the b-imperfect in particular) typical of sedentary dialects. This fact reflects the old close connections of many tribes in the area with the sedentary population of South Palestine," As a whole the dialects belong to the same type as the Sinai dialects.
I.e. the dialect groups spoken (respectively) east and west of Wâdi 'Araba and the Gulf of 'Aqaba, cf. PALVA, 1990, p. 154 and 158.
Originally wasm at-Turayyâ, the rainy season of 75 days, roughly coinciding with the appearance of the Pleiades over the eastern horizon at the end of October; cf. BAILEY, 1974a, p.585.
The fem. suffix -hiy does not only occur in the Negev (cf. BLANC, 1970, p. 19-20(130-131) and PALVA, 1990, p. 157), but also in the tribal dialects spoken by the Rumelât and Sawârka (in the northeast), and (in the northwest) the Masâ'îd and 'Ayâyda (own observations).
And in the case of dialects in northeastern Sinai, we may also assume a strong influence from the dialect of the regional capital al'Arîš, where also a typically sedentary dialect is spoken, but more of the Palestinian type (cf. DE JONG, "A Description of 'Arâyši Arabic” (forthcoming))- For the dialects in the northwest, we must take into consideration the possibility of influence from the sedentary dialects of the Egyptian Delta.
Haut de page