Review of Perspectives on Arabic Linguistics XVI
|
|
|
|
|
Review:
|
EDITOR: Boudelaa, Sami
TITLE: Perspectives on Arabic Linguistics XVI
SUBTITLE: Papers from the Sixteenth Annual Symposium on Arabic Linguistics,
Cambridge, March 2002
SERIES: Current Issues in Linguistic Theory 266
PUBLISHER: John Benjamins
YEAR: 2006
Rania Habib, Program in Linguistics, The University of Florida
SUMMARY
This volume of _Perspectives on Arabic Linguistics_ is a collection of
eight papers from among twenty-four papers presented at the Sixteenth
Annual Symposium on Arabic Linguistics that took place in March 2002, in
Cambridge University, UK. The eight papers were peer-reviewed and edited by
Sami Boudelaa.
The first paper, ''The organization of the lexicon in Arabic and other
Semitic languages,'' is by the guest speaker, Georges Bohas. In this paper,
Bohas argues that the traditional triliteral/triconsonantal root approach
(e.g. Fleisch 1961) does not explain the ''phonetic and semantic relations''
(p.1, 2) among words in Semitic languages; rather, a binary approach (e.g.
Renan 1855) to words in the lexicon better explains the homonymy, synonymy,
and antithetical polysemy observed in the Arabic lexicon.
In order to illustrate this point, Bohas first differentiates between the
''root'' of a word that consists of mere consonants (e.g. ktb 'write') with
no phonetic representation and the ''radical'' that exists as a ''phonetic
representation of the language'' (p. 4) (e.g. kataba 'he wrote'; katabtu 'I
wrote'). One can note that the same consonants (same root) but different
vowels are used, evoking various radicals and conveying different meanings.
Bohas proposes the etymon, ''a binary composition of the phonemes'' (p. 4),
to account for the phonetic-semantic relations. However, similarity in
meaning may exist among words that are derived from various roots that
mostly have two consonants in common; the third consonant is added or
mapped from the second consonant of the word. Bohas observed that words,
such as mataa, madda, and maTTa, consist of m and t, d, or T, all of which
are segments with the features [coronal] [-continuant] The combination of
{m x [coronal] [-continuant]}is a matrix that is linked to a semic nucleus
''to pull out, to stretch.'' (p. 5). It is such binary combination of feature
matrices and semantic nuclei that could provide explanation for the
phonetic-semantic relations present among words.
Bohas disagrees with previous views that hold that the root is a basic
reality and is the base for organization of the lexicon and speakers are
conscious of it and can extract it easily and spontaneously. Based on
empirical evidence, Bohas shows that the root cannot be detected easily;
rather it is very difficult to extract it from very common words even by
advanced native students of Arabic literature and languages. Though the 132
participants were able to read the given words correctly, they were not
able to analyze them and extract the correct roots from them, leading to
the following conclusion: ''root identification cannot be a precondition to
reading'' (p. 16).
This study has implications for other languages, particularly Afroasiatic
languages. The paper calls for extraction of the etymon and feature
matrices of each language before making comparisons between these languages.
In the second paper, Igancio Ferrando examines ''The plural of paucity in
Arabic and its actual scope on two claims by Siibawayhi and Al-farraa'''.
The plural of paucity (PP) is considered a fourth type of plural that is
peculiar in the form selected to represent it. Ferrando differentiates
between the notion of PP (e.g ʔaklub ''a few dogs (from three to ten)'') and
plural of abundance (PA) (e.g. kilaab ''dogs (more than ten)''). The former
is an '''individual, specific' plural'' and the later is a '''collective,
non-differentiatial' plural'' (p. 40). In this paper, Ferrando tries to
answer a number of questions. His main concerns are to find out if PP
constitutes a separate number category; whether its use and distribution in
present-day Arabic compares to the variety of Arabic described by
Siibawayhi; and if the PP has an effect on ''agreement variation in
Classical Arabic (full vs. deflected)'' (p. 39). Ferrando alludes to the
difference between Arabic and other Proto-Semitic languages that use
particular morphology to differentiate between PP and PA, such as Tigre (a
North Ethiopian living language) and Akkadian.
Siibawayhi contends that PP is ''a true component of Arabic'' (p. 45).
However, the notion of PP weakened with time and subsequent grammarians
paid very little attention to this notion; rather, they reflected ''loss of
discrimination'' (p. 47) between PP and PA. In the present day, very few
forms of PP are retained in Modern Standard Arabic and are considered as
''archaic and/or high-register variant[s]'' (p. 48); the PP forms are no
longer observed as distinct from PA forms in meaning or number. Ferrando
supports Siibawayhi's claim that a difference did exist in the past between
PP and PA. His point of view is supported by statistical evidence from the
Muʕallaqaat that are representative of the Pre-Islamic period and of the
variety described by Siibawayhi. The survey shows that a clear semantic
distinction was made by poets in their use of PP and PA forms. Comparing
these results with more recent poetic corpus by Belnap and Gee (1994), one
can observe that the distinction between PP and PA is disappearing
gradually as the Arabic language is evolving.
The second claim made by Al-farraa' is that the choice of agreement
corresponds with the type of plural form, that is, PP co-occurs with plural
verbs, adjectives, and pronouns (i.e. full/natural agreement), whereas PA
co-occurs with feminine singular forms: verbs, adjectives, and pronouns
(i.e. deflected agreement). The survey done by Ferrando also corresponds
with this view to some extent. However, no conclusive conclusions could be
reached from such preliminary data. However, the results raise the
following hypothesis: ''the distribution of PP-PA forms according to the
context is one of the primary criteria to be handled when trying to explain
the apparently unmotivated free variation of the agreement patterns'' (p.
59). Ferrando's final observation is that PP is ''an optional feature at the
disposal of Arabic writers,'' not ''a living productive notion'' (p. 59).
The third paper, ''Why there is no koiné in Sanʕaaʔ, Yemen'', is by Andrew
Freeman. From the title of the paper, one can observe that the purpose of
the study is to find out if there is a koiné in Sanʕaaʔ, Yemen. The
increased migration to Sanʕaaʔ led the author to predict a breakdown in the
social networks and the development of a new variety by the newly-formed
community. The data is based mainly on a questionnaire given to 72
participants on the use of 39 linguistic features: phonological,
morphological, and lexical; 36 hours of informal interviews; 38 hours of
three TV serials written in various dialects; six hours of the educational,
dialectal radio show ''Musʕad wa Musʕad''; and a text of a play that consists
of 18,000 words, written in various dialects. The questionnaires provided a
strong evidence for the lack of a koiné. Even TV serials showed that
writers of those serials had to know the dialect of each actor in order to
write his/her role in that dialect, indicating that people are aware of the
existence of a middle dialect and that there is no stable inter-community
variety or koiné.
Freeman lays out six factors that hinder the formation of a koine. He
arrives at the following conclusion from his analyses of the data and from
his observations. There is no koiné in Sanʕaaʔ; rather, there are two
distinctive dialect systems in Sanʕaaʔ. These two dialect systems are
moving in the direction of being more distinct. Freeman predicts from this
observation future formation of two koinés in Sanʕaaʔ: ''one spoken by the
Sanʕaanis and the migrants from the northern highlands and the other by the
migrants from the southern region of Yemen'' (p. 79). An example of this
split is the use of the future markers sha in the north and baa in the
southern region in Sanʕaaʔ. Another example is that there is no compromise
position between the use of the phonological variants of /q/ and /j/.
Speakers either only use [q] and [g] (the variants of one of the dialects
in Sanʕaaʔ) or [g] and [j] (the variants of one of another dialects in
Sanʕaaʔ) as variants of /q/ and /j/ respectively. A switch from [q] to [g]
must be accompanied by a switch from [g] to [j]; there is not intermediate
position.
In the fourth paper, ''Empty nuclei in Arabic speech patterns and the
diacritic sukuun'', Michael Ingleby and Fatmah A Baothman argue that the
diacritic sukuun in Arabic constitutes a phonological segment though it
orthographically represents the absence of a vowel. Direct evidence comes
from ''duration statistics of clusters and geminates in recorded speech'' (p.
83). Indirect evidence comes from ''the coarticulation phenomena associated
with clusters'' (p. 83). The proposal is that in the absence of a coda and
'empty vowel' that separates two onsets is a ''sukuun nucleus'' (p. 83). The
evidence is taken from Modern Standard Arabic (MSA) used in secular news
broadcasting and among educated Arab professionals. The NO CODA approach
the authors adopt is based on the assumption that the syllable is a CV or
CVV. In the CV syllable, the nucleus V could be either fatha (a), damma
(ʊ), kasra (ɪ) or sukuun (short vowel). In the CVV syllable, the VV nucleus
is either aa, ʊʊ, or ɪɪ.
The representation of sukuun as an active nucleus can account for devoicing
word-finally. The prime H is taken to denote voicelessness and is
representative of voiceless obstruents. The sukuun is considered as the
element H in isolation. The presence of the sukuun word-finally is
assimilated by the final consonant and thus it is devoiced (e.g. ʕutb -->
ʕutp). In this sense, the sukuun has a regressive assimilative effect on
the final consonant of the root. This NO CODA approach also accounts for
schwa epenthesis between clusters such as ʁð that are surrounded by vowels.
Evidence comes from a spectrogram investigation of the corpus of
recordings. Ingleby and Baothman take the weakened form of the sukuun to be
present in all consonant pairs including geminates. This is supported by a
spectrogram test that shows that geminates require more time of
articulation than their singleton consonants. In a comparison with the
duration of geminates in other languages, Arabic geminates are the longest
in articulation (extended 120%) whereas they are extended 60-70% in Italian
and 85-95% in English. The duration stability of sukuun is also expected to
have a role in determining, like vowels, stress, pitch, and length of
stressed segments. The study may have bearing on computational phonology
and may explain similar phenomena taking place in other languages, such as
German and Slavic languages, which are analyzed by some researchers to have
empty nuclei (e.g. Brockhaus 1995). The study suggests that the empty
nucleus should be treated as a phonological parameter that is ON in
languages that encounter devoicing (e.g. Arabic, German, and Dutch) and OFF
in languages that do not (e.g. English and Romance languages).
In the fifth paper, ''Representing coarticulation processes in Arabic'',
Fatmah A. Baothman and Michael Ingleby present an acoustic study of a
7000-word corpus from recordings of ten MSA speakers. The paper is
concerned with the directionality of nasal assimilation and pharyngeal
spreading in Arabic. In comparison with English, Baothman and Ingleby found
that nasal assimilation in Arabic is unidirectional and follows the same
regressive pattern in English (e.g. dˤank --> dˤaɳk (velarized); qanʂˤ -->
qaɲʂˤ (palatalized); janfuθha --> jamfuθha (labialized)). This type of
assimilation is local (i.e. from the immediate following consonant). On the
other hand, pharyngeal spreading is ''bi-directional and non-local'' (p. 98),
i.e. it can spread regressively and progressively and the emphatic source
does not have to be immediately following or preceding the pharyngealized
sound (e.g. tˤuquusˤ (progressive long-ranged); sˤiqtˤ (regressive
long-ranged)).
In the sixth paper, ''The textual component in Classical Arabic:
Investigating information structure'', Salwa A. Kamel investigates the
effect of disparate word orders on the transfer of meaning from one
language to another. Thus, she contrasts Information Structure in Arabic
with those in English to discover their effect on meaning and to ''establish
the limits of information equivalence in both languages'' (p. 103). Kamel
uses data from Aljurjani (1992) and draws on the similarity between
Halliday's (1994) Textual Level of Functional grammar and Aljurjani's
900-year-old theory of naðˤm (i.e. ordering constituents in sentences) in
order to show that there is a relationship between information structure
and meaning. Aljurjani's model contends that word order is motivated by
context and that ''textual arrangement dictated by background information
reflect[s] the speaker/writer's intentions'' (p.106). It is a functional
theory in which sentence structure, meaning and purpose are closely
related. For Aljurjani, ''[m]eaning is lexical and grammatical, and words
come to 'mean' only when combined with other words, i.e., contextual
meaning'' (p. 108). Like Aljurjani's naðˤm theory, Halliday's Functional
Grammar relates information structure with thematic structure in the clause.
The study shows that Classical Arabic makes use of preposing and initial
position (i.e. syntactic processes) to create ''focal emphasis in the
message'' (p. 127) in contrast to English that uses tonicity (tonic
syllable) (i.e. phonological processes) to establish focal prominence.
Thus, Arabic moves elements to initial and earlier position to achieve this
prominence. Definiteness plays a role in Arabic in determining 'New' and
'Given' information. Initial position is stipulated on intuitive sense to
host ''the verb, the scope of negation and interrogation, and nominal New
information'' (p. 129).
The seventh paper, ''The acoustic and auditory differences in the /t/-/ṭ/
opposition in male and female speakers of Jordanian Arabic'', is by Ghada
Khattab, Feda Al-Tamimi, and Barry Heselwood. The study aims at
investigating gender differences in implementing the phonemic opposition
between the plain /t/ and the emphatic /ṭ/ in Jordanian Arabic. The locus
of investigation is the differences in production and the acoustic signal,
not the perceptual role of those acoustic signals. 212 tokens of each
plosive (424 in total) were elicited and analyzed from five male and five
female speakers who were engaged in manipulated topics in the conversation.
The two Arabic authors rated the tokens on a point scale of emphaticness:
full emphatic, half-emphatic, plain. Agreement between the two raters was
100% for males and 97% for females.
The results of the acoustic analysis shows that all male speakers and the
two female speakers from Irbid pronounced /ṭ/ as full emphatic; no
difference with respect to /t/ that was pronounced plain by all speakers.
The three female speakers from the capital Amman varied their pronunciation
of /ṭ/ between plain, half emphatic, and full emphatic. This led to the
hesitation in concluding that gender plays the major role in emphaticness.
Rather, it could be locality and social class differences that play a major
role in determining emphaticness. The reason for this expectation is that
variation occurred only among Ammani females, one of which approximated /ṭ/
towards [t] in her speech much more than the others to give the impression
that she is from a modern urban culture and from a high-class area in Amman.
The results also show that in general the males' realization of vowels
immediately after /ṭ/ show more approximation of F1 and F2 than females'
realizations. The formant frequency data is also supported by a Voice Onset
Time (VOT) data that shows that VOT values for women are significantly
higher that those for males in their pronunciation of /ṭ/. However, this
does not give a conclusive conclusion that there is extreme opposition
between males and females in the pronunciation of /ṭ/ for the same reasons
mentioned above. Further research is needed where social class and degree
of modernization versus traditional culture should be controlled for as
well as locality because these social factors may play a significant role
in the degree of emphaticness observed among speakers more than gender
does. Thus, the findings of this study present /tt/ as a sociophonetic
feature.
The eighth paper, ''Pharyngealization effects in Maltese Arabic'', is by Mary
Ann Walter who argues that emphatics in Arabic involve a tongue root
feature. She gives description of Maltese Arabic in which emphatics no
longer exist on the surface but their effects may still be observed on
neighboring vowels, such as backing and lowering. For example, in saif
'summer', originally ṣaif, vowel raising is blocked in an emphatic
environment. On the other hand vowel raising is allowed in seif 'sword',
originally saif, in non-emphatic environment.
In order to investigate the effect of emphatics on neighboring vowels,
Walter draws words from Standard Maltese-English dictionaries, isolates
emphatics, and tries to determine their context by comparing them to Arabic
words. Walter found that emphasis spread induces the following vowel
changes: /a/ is retained as /a/ (e.g. tabiiba 'female doctor' from
ṭabiiba); /i/ is lowered to /e/ (e.g. dell 'shadow' from ẓill); and /aa/ is
shortened to /a/ (e.g. sawwar 'painter' from ṣawwaar). In contrast, a
non-emphatic environment induces different changes: /a/ --> /e/; /i/ is
retained as /i/; and /aa/ --> /ie/.
Emphasis spread could be rightward (e.g. sawwar 'painter' from ṣawwaar) or
leftward (e.g. rabbat 'person who ties' from rabbaaṭ). Leftward emphatic
spreading is more permissive than rightward spreading that could be blocked
by segments that are specified for the antagonistic features [+high] and
[+front] (e.g. the palatals /y/ and /š/, particularly a geminate high front
vowel) because emphatics are specified for [-ATR] and [RTR]. /w/ or /ww/ do
not block spreading. Furthermore, in Maltese Arabic, a process of
dissimilation of gutturals similar to the one that occurs in Palestinian
Arabic takes place. In her evaluation of the descriptive and explanatory
adequacy of recently proposed feature geometries, Walter finds the [+ATR]
feature proposed by other models to be unnecessary to explain guttural
dissimilation. Rather, most of these processes could only be explained and
motivated by Tongue Root node only.
EVALUATION
Most of the papers in this volume focus on phonetics, phonology, and
acoustic analysis. Though the third paper deals with sociolinguistic
variation and change, it deals with phonological segments. The first two
papers involve in addition to phonology, morphology and semantic
implications. Only the sixth paper diverts away from phonology and deals
with the relationship between syntax and semantics. Though the eighth paper
is concerned with phonological processes, it has sociolinguistic
implications. The topics are diverse and not closely related to each other.
The fourth and fifth paper complement each other as they are written by the
same authors and are based on the same recorded data. The diversity of the
papers and topics probably did not allow for organization of the papers
into different sections and topics within the volume. Nonetheless, most of
the topics of the papers are interesting and call for further investigation
and research in the future, which is a good departure for researchers in
general.
The first paper by Bohas about the organization of the lexicon brings about
a new innovative approach that overthrows the notion of the triconsonantal
root as the basic unit for organizing the lexicon.
The second paper by Ferrando revisits the notion of the plural of paucity
and shows the deterioration of its significance as a separate number
category in our present day in comparison to its important status in the
Pre-Islamic period.
As a sociolinguistic variation and change paper, the third paper by Freeman
presents a new consequence of social factors and regional contact. Instead
of the development of a koiné as a result of the migration to Sanaaʕaʔ, a
breakdown in the social setting takes place between the northern and the
southern regions, leading to two separate koinés in the north and south of
Sanaaʕaʔ.
Presenting the sukuun as an active nucleus in the fourth paper by Ingleby
and Baothman has very strong implications for other languages that deal
with empty nuclei and draws on the similarity between Arabic and other
languages, such as German, in this regard. The same authors deal in the
fifth paper with the directionality of nasal assimilation and pharyngeal
spreading, which are phenomena that are common in other languages,
particularly nasal assimilation. Pharyngeal spreading (called by some
uvularization (e.g. Zawaydeh (1998)) has been the concern of many
researchers of Arabic, but the acoustic analysis of this study adds support
to previous analyses.
The sixth paper by Kamel brings two approaches together: a 900-year old one
(i.e. Aljurjani's naðˤm theory) and Halliday's more recent Functional
Grammar theory in which structure is closely related to purpose and
meaning. The paper shows that the past analysis is not much different for
the present analysis by adopting data from Aljurjani and applying
Halliday's model.
The final two papers present empirical acoustic evidence for two different
aspects of the Arabic language. The seventh paper by Khattab, Al-Tamimi,
and Heselwood examines the sociolinguistic variable use of /ṭ/ and implies
that gender may not be the main factor behind this variation; rather, other
factors such as social class and location may play a more significant role
in such variation. The eighth paper by Walter examines the observed effect
of emphatics on neighboring vowels in Maltese Arabic despite the
disappearance of emphatics on the surface. This paper is somehow related to
the fifth paper in its reference to emphasis spread (i.e. RTR feature) that
seems to be still taking place in Maltese Arabic in both directions, like
in Arabic in which emphatics still exist, despite the absence of those
emphatics.
REFERENCES
Aljurjani, Abdul-qaahir. 1992. dalaaʔil ʔal-Iʕjaaz ed. By Mahmoud M.
Shaker. Cairo: Matbaʕat Al-madani.
Belnap, R. Kirk and John Gee. 1994. ''Classical Arabic in contact: The
transition to near categorical agreement patterns''. Perspectives on Arabic
Linguistics VI ed. By Mushira Eid, Vicente Cantarino, Kieth Walters,
121-49. Amsterdam: John Benjamins.
Brockhaus, Wiebke G. 1995. Final Devoicing in the Phonology of German.
Linguistische Arbeiten 336, Tübingen: Max Niemeyer Verlag.
Fleisch, Henri. 1961. Traité de philologie arabe, vol. I. Préliminaires,
phonétique,
morphologie nominale. Beyrouth: Imprimerie Catholique.
Halliday, M. A. K. 1994. An Introduction to Functional Grammar. London:
Edward Arnold.
Renan, Ernest. 1855. Historie générale et système comparé des langues
sémitiques. Paris: Imprimerie Catholique.
Zawaydeh, Bushra Adnan, 1998. ''Gradient uvularization spread in
Ammani-Jordanian Arabic''. Perspectives on Arabic linguistics XI ed. by
Elabbas Benmamoun, Mushira Eid, Niloofar Haeri, 117-142. Amsterdam: John
Benjamins.
|
| |
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Rania Habib is a Ph.D. Candidate in Linguistics at the University of
Florida. As a Fulbright student, she completed her master's degree in
linguistics at the University of Florida in 2005. Rania is interested in
sociolinguistic variation and change and her current research involves the
application of Optimality Theory and the Gradual Learning Algorithm to
sociolinguistic variation and change. She is also interested in Pragmatics,
Second Language Acquisition, and Syntax. Rania has research papers in these
linguistic subfields, some of which have been presented in conferences,
submitted for publication, and are about to appear in well-known journals.
|
|
|
|
|
|
