Untitled Document

Pharyngealisation in Syriac

Dr.Abdell Hameed Jakhop Ghabreal

This phonetic phenomenon, particularly characteristic of Semitic languages, refers to any secondary involving a constriction of the pharynx, the tubular cavity which constitutes the throat above the larynx. For example, a pharyngealised /S/ is a secondary articulation produced by simultaneously constricting the pharynx while making the /s/ articulation. ''The auditory result would be a sound with a somewhat central and husky resonance.'' (Crystal, 1987:226). Pharyngealisation is a term usually applied to consonants other than pharyngeal consonants. It can be used with reference to vowels though such variations in vowel articulation are usually described in different terms ('centalised, retracted', etc.).
    Pharyngealisation may be an essential feature of the sound's identity, contrasting with other non- pharyngealised sounds. It is also a prosodic feature influencing neighbouring syllables.
The present paper deals with pharyngealisation in Syriac as spoken by the Christian community in Ankawa, a town just north of Erbil. Being a speaker of the Spoken Syriac of Ankawa (henceforth, the SSA), the writer is his own informant and the articulatory description in this study are based on his observations of his own articulations.
The SSA has three pharyngealised (emphatic) consonant phonemes, namely, /T/, /S/ and /R/. in the articulation of these pharyngealised consonants, and besides the basic tongue-tip/blade action, the upper part of the front of the tongue is slightly lowered and concave, "the tongue as a whole is tense and somewhat retracted" (Erwin, 1963:12) and there is some "contraction of the upper pharynx", (Jakobson, 1957: 106) whereas the articulation of their non- pharyngealised counterparts involve, among other things, the relative laxness of the tongue with a somewhat convex upper surface. A brief articulatory description of the said phonemes follows along with the establishment of their phonemic status. This may be a convenient point to specify the phonemic / phonetic notations used in this paper to represent Syriac segmentals. Here is the system of sumbols with a key word for each phoneme standardized symbols being excluded:
Vocalic Notation:
/i/ as in /mishxa/    'oil' (a sound very similar to a schwa)
/e:/ as in /ce:pa/         'rock'
/a:/ as in /na:sha/       'human being'
/o:/ as in /zo:ra/         'small'
Cononantal Notation:
/c/ as in /ce:pa/ 'rock' (the sound initiating and terminating English CHURCH).
/th/ as in /the:li/ 'came' (the sound initiating English THINK).
/sh/ as in /she:kar/ 'sugar' (the sound initiating and terminating English SUGAR).
/j/ as in /jo:g/ 'come close!' (The sound initiating and terminating English JUDGE).
/r/ as in /?arzan/  'cheap'  (a non- pharyngealised trill).
/y/ as in /yo:ma/  'day'  (the sound initiating English YES).
/g/ as in /Gam/  'sorrow' (voiced velar fricative).
1- /T/ The principal allophone of this phoneme is a voiceless pharyngealised aspirtated post-dental stop produced by the tip of the tongue against the ridge behind the upper front teeth at a more retracted point in the alveolar ridge than for its plain counterpart /t/, which is a voiceless aspirated dental stop. There is, moreover, a relatively bigger area of contact and a firmer one than for /t/, between the articulators.
What has just been said respecting the common feature of the secondary articulation of this and other pharyngealised consonants can now be summed up by saying "they are distinguished by lateral expansion of the tongue and by its flattening in the mouth in contrast with the lateral contraction and forward tongue-raising appropriate to 'non-emphatic' correlative consonants" and that " the 'flattening' referred to carries with it a concomitant reduction of the pharyngeal cavity in contrast with the open pharynx that accompanies 'forward-raising'." (Mitchell, 1975:31).
    The point of articulation for /T/ can be influenced by the following sound; it is further back, almost reaching the rear of the alveolar ridge, when followed by a back vowel, e.g /Ta:ma/ 'there'.
The phonemic status of /T/ can be established through the following minimal pairs which show the distinctive opposition of /T/ with /t/:
    /shiTli/  'singed' (M)  vs. /shitli/ 'nursery-plants'
    /Te:ni/   'loads'           vs  /te:ni/ 'figs'
    /tRa:ya/  'driving'       vs  /tra:ya/  'getting wet'
2- /S/ The main variant of this phoneme is a voiceless pharyngealised post-dental fricative, which is produced when the air-stream is forced out through a narrow groove formed between the tip and blade the upper front teeth, producing audible friction. The air seems to escape more freely by means of a wider groove in the center of the mouth and wider opening between the teeth than for its plain counterpart /s/, which is a voiceless dental fricative. Moreover, the side rims of the tongue make a firmer and closer contact with the upper side teeth than for /s/.
/S/ may have allophones articulated further back on the alveolar ridge when a back vowel follows, e.g /Rso:m/ 'Draw!'. The phoneme statue of /S/ can be recognized through the following contrastive pairs, which show its distinctive opposition with /s/:
/Swe:li/    'hardened' (M)  vs. /swe:li/  'got satisfied' (M)
/Spe:la/    'got clear' (F)    vs. /spe:la/  'filled'  (F)
/mSaloyi/ 'praying'        vs /msalo:yi/ 'amusement'
3- /R/ The principal allophone of this phoneme is a voiced pharyngealised alveolar trill, which is produced by a "stricture of intermittent closure." (Abercrombie, 1978: 49). The tip of the tongue approximates and produces rapid taps on the alveolar ridge, the tip being flattened into a wider area and the contact being firmer than for /r/, a voiced post-dental trill. A one-tap-trill also occurs in the SSA, especially intervocalically, e.g. /taRa/ 'door'.
/R/ may be partially or completely devoiced when it is in contact with voiceless stops or fricatives or finally before a pause, e.g. [Rpe:li]  'slackened' (M), [fRe:li] 'became abundant' [she:kaR] 'sugar'. The phonemic status of /R/ in the SSA can be established through these minimal pairs:
/xyaRta/  'cucumber'   vs. /xyarta/  'look' (n.)
/Pa:Ri/    'sheep' (pl.)   vs.  /pa:ri/   'money'
/buRka:tha/  'blessings'  vs.  /burka:tha/  'knees'
/go:Ra/    'man'              vs.  /go:ra/        'to marry' (F)
The minimum range of pharyngealisation in the SSA is one syllable. Once a pharyngealised phoneme occurs in a syllable, the nieghbouring syllables, preceding and following, can be pharyngealised. Indeed, the phenomenon of pharyngealisation is a "prosodic feature", to use Firth's term, that may extend even over the whole range of affixes attached to a pharyngealised base. This implies that each and every consonant, as well as vowel, phoneme has a pharyngealised allophone in the SSA. Accordingly, two categories of 'pharyngealised' consonant have to be distinguished:
i. pharyngealised phonemes, which are in parallel distribution vis-a-vis their corresponding non- pharyngealised consonant phonemes.
ii. Pharyngealised allophones, which are in complementary distribution with their non- pharyngealised counterparts. They are, in other words, variants allophonically conditioned in a pharyngealised environment.
Thus, for instance, in [SLI:WA] 'Cross', /S/ is a pharyngealised phoneme, whereas [I] [I:] [W] and [A] are only pharyngealised allophones of /i/, /i:/, /w/ and /a/, respectively.
    Although the possibility of the occurrence in the same word is actually difficult to decide auditorily (not through transcription though) which segment is to be considered primarily responsible for the prosody of pharyngealisation that may characterize the whole utterance. That is, it is difficult to determine which segment (or segments) is to be counted as a pharyngealised phoneme rather than a pharyngealised allophone or the other way round since pharyngealisation does not occur as a feature of a single segment. Nothing in the word 'saTaana'  'devil' (phonetically [saTa:na], for example, tells us whether [s] or [T] is responsible for pharyngealisation, or both.
     One way to solve such a problem is by tracing the etymology of certain words. Thus by referring to the written code of Syriac, we find that in       سطاًناً     [saTa:na], where  س stands for /s/and ط for /T/, it is the latter that is responsible for this prosody of pharyngealisation, [s] being merely a pharyngealised allophone of /s/. This sort of approach is, however, not valid synchronically.
If anyhow a certain segment occurs in a pharyngealised environment at a word junction, identification of its being a pharyngealised phoneme or allophone may sometimes be possible by analyzing the utterance in question into isolate forms. If the segment in question has a phonemic status in the isolate form, then it is to be regarded as a pharyngealised phoneme; if not, it should be only an allophonic variation of a non-pharyngealised consonant phoneme. For example, in [yo: mit So:ma] 'day of fasting' [-t] is only a phonologically conditioned allomorph, in terms of pharyngealisation of {-T}, a morphophoneme (=systematic phoneme) denoting possession or relationship. This allphone occurrence of  {-T} is actually the outcome of the juxtaposition of the two words /yo:mit/ 'day of' and /So:ma/ 'fasting', hence the resultant assimilation of {-t} to [-t] under the influence of the following /S/, which, on the other hand, is a pharyngealised phoneme since it has a genuinely phonemic status in the isolate form /So:ma/.
Similarly in [bir paTrus] 'peter's son' (/bir/ being a contracted form of /?u:brit/ 'son of'), the [r] of [bir], in this environment is actually a pharyngealised allophone of /r/ in anticipation of the next pharyngealised environment, for in the isolate from /bir/ it is originally non- pharyngealised.
By way of further contrast, in [maRpaTrus] 'Saint peter', the /R/ of [maR] is to be counted as a pharyngealised phoneme because it originally occurs as such as in the isolate form /maR/.
This instance of analysis ,however ,cannot account for certain segments occurring within certain pharyngealised roots since it is restricted to junctional positions .
    In may also be useful ,at this point ,to talk briefly about a different domain of pharyngealisation in the SSA, namely the tendency to arbitrarily pharyngealise loan –words even if certain such words may be devoid of pharyngealised phonemes ,e.g.,
[damar]            'blood vessel'
[pa:ke:t]           'packet'
[dama:g]          'brain '
[jizmi]              'boots'
Consequently, some very few such pharyngealised loan-words chance to form contrastive pairs with certain native expressions e.g. [ba:za] 'kind of cloth ' vs. [ba:za] 'will go' (F).It is also true that within vocabulary a few words do enter in to certain paradigms of minimal pairs, e.g. [zde:li] 'fell' (a star) vs. [zde:li] 'got frightened ' (M); [ma:ya] 'water' vs.[ma:ya] 'of that (F) sort'.
Such infrequent instances give rise to the emergence of what Erwin (1963:15) terms "the secondary emphatic", e.g. [B],[M],[Z] etc., whose independent status as distinctive sounds is too restricted indeed to be established as phonemes in the SSA phonology, since only a few pairs of words  are distinguished solely by the difference between a secondary emphatic and its plain equivalent .A sound should be given phonemic status in the system of a language if it forms some very frequent oppositions with other phonemes in the language .Contrarywise, /T/,/S/ and /R/ play a significant role in the phonological structure of the SSA in that their replacement by their non –pharyngealised  counterparts can result in a semantically different word or unintelligible utterance as has been demonstrated earlier .Syllables adjacent to a pharyngealised  syllable , and which may have been coloured by the prosody of pharyngealisation will lose this feature once such replacement takes place, e.g. [laTpe:li] 'didn't stick' vs. [latpe:li] 'didn't twist' (M), where the substitution of /T/ /for /t/leads to the automatic disappearance of the prosody from the whole utterance.
So far we have been looking at the effect of pharyngealisation mostly within words .perhaps the last example serves as a natural transition towards considering the prosody of pharyngealisation at word or morpheme boundaries both regressively and progressively. 
    The following phrases illustrate how the final segment of the first word may be allophonically affected by the pharyngealised phoneme /T/, /S/ and /R/ constituting the initial segment of the second word ,  hence regressive assimilaition in terms of pharyngealisation :
            /qam+TaRa/  →     (qam Ta Ra)  ‘in front of the door’                     
            /kam+Rrawi/ →    (kamSalwi:li)  ‘they crucified Him’
            /Ku:d+Rawi/  →     (KudRa:wi)   ‘when he grows up’
           In the following phrases, the initial  segment of the second word may de affected by the pharyngealised phonemes/ T /, /S/ and /R/ constituting the final segment of the first word, hence progressive assimilation in terms of pharyngealisation:
    /qa:T+blo: shin/ →  [qa: Tblo:shin]  ' I'll wear a suit'
       /xlo: S+ minna:/[xlo:Sm i nna:]  
 ' Leave her alone!'
The foregoing pages constitute a brief study of pharyngealisation in the spoken Syriac of Ankawa. Pharyngealisation plays two major roles in the phonology of this language. Firstly, it is responsible for the existence of three pharyngealised phonemes that contrast phonemically with their non- pharyngealised counterparts, namely, /T/ /S/ and /R/.
Secondly, pharyngealisation is a prosodic feature that can extend over neighboring syllables, or even whole utterance through processes of both regressive and progressive assimilation.
.1pharyngealised phonemes will be transcribed with a capital letter. Pharyngealised allophone of vowels and consonants will be transcribed with a subscript dash under the symbols for their non- pharyngealised counterpart.
.2since every segment of such utterances is normally pharyngealised ,they are transcribed phonetically in order to show their precise phonetic property .On the other hand ,in the phonetic transcription of the utterances that contain /T/,/S/or /R/ only the syllables in which these exist are found to have been pharyngealised ,e.g.[buRKa:tha] 'blessings'. This, however, does not rule out the possibility of the adjacent syllables being pharyngealised .
.1Abercrombie, D. Elements of General phonetics. Edinburgh:Edinburgh University Press,1978).
.2Crystal, D. A  Dictionary of linguistics and phonetics  .
(Oxford:basil Blackwell,1987).
.3Erwin, W.M  A Short Reference Grammar of Iraqi Arabic. (Washington:Georgetown University Press,1963).
.4Jakobson, R. 'Mufaxxama' "the emphatic phoneme in Arabic", in Studies presented to Joshua Whatmough. (The Hague: mouton and Co., 1957).
.5 Mitchell, T.F Principles of Firthian Linguistics.
(London: Longman Group, 1975).