CMA Copula and similar phenomena

[Hans-Jürgen Sasse]

There is no doubt that Cypriot Maronite Arabic (CMA) has a regular
copula for the "present tense"  that has developed out of the personal
pronouns. It supplements the verb /ka:n/ 'be', which is used, as in all
other Arabic varieties, with nominal predicates in all tenses other than
the present, e.g. past /kund pxal/ 'I was sick' vs. /7ana pxal/ 'I am
sick'.

Although the example given above looks superficially identical to the
situation found elsewhere in Arabic (and in Semitic in general), use of
the "present tense copula" in CMA differs from other varieties of Arabic
(both classical and modern) in several respects.

First of all, it is obligatory in the third person even in the presence
of a full noun subject. Thus, we have /l-intsan 7o pxal/ (ART-man
COP:3sgm sick)  'the man is sick' but not */l-intsan pxal/. If you
compare this to the situation in, say, Egyptian (Cairo) Arabic, you
would find something like /ir-ra:gil 9ayya:n/ (ART-man sick) as the
normal construction there, i.e. the use of a "zero copula"
(juxtaposition of subject and nominal predicate) clearly prevails.
Egyptian Arabic (and most other varieties of Arabic as well) may use an
additional 3rd person pronoun in such cases, but it is then used as a
topic pronoun, reinforcing the topic character of the subject of the
nominal sentence (/ir-ra:gil huwwa 9ayya:n/). In the first and second
persons pronominal forms are used (Egyptian /7ana 9ayya:n/ 'I am sick'),
but these have to be interpreted as pronominal subjects rather than as
copula forms.

Secondly, the CMA copula forms of the 3rd person differ from the free
pronominal forms. Copula forms continue the original Arabic personal
pronouns 3sm /huwa/ (becoming CMA /7o/), 3sf /hiya/ (becoming CMA /7e/),
and 3pl */hinna/ - which typically replaces Classical /huma/ in that
area - (becoming CMA /7enne/). The corresponding independent pronouns
are CMA /7adha/, /7adhi/, and /7alli/ respectively, deriving from the
original demonstrative (Classical Arabic /ha:dha:, etc.). First and
second persons of the copula are identical with the independent
pronominal forms: 1sg /7ana/, 2sgm /7int/, 2sf /7inti/, 1pl /nahni/, 2pl
/7indu/. Incidentally, the derivation of the 3pl form /7enne/ from
Classical Arabic /7inna/ is totally implausible, given the entire
evidence.
(For more on CMA morphology the relevant manuals should be consulted,
e.g. Maria Tsiapera 1969, A Descriptive Analysis of Cypriote Maronite
Arabic. The Hague / Paris: Mouton.)

It stands to reason that the consolidation of the (erstwhile) pronominal
forms as indicators of the predicative link between the subject and the
predicate of a nominal sentence is not an immanent development of CMA,
but was driven by the structure of Greek, with which CMA has been in
close contact for centuries and whose influence is evident in all parts
of the grammar and the vocabulary of CMA. Greek has an obligatory copula
in the present tense (of verbal origin), and CMA simply assimilated its
structure to the Greek one by exploiting the already present (inherited)
device of using a topic pronoun. In other words, it made the optional
topic pronoun obligatory thereby creating a copula. The copula remained
with the old pronominal forms, while the independent pronouns were
renovated from the demonstratives.

Similar developments have been observed in other Arabic dialects which
are in close contact with copula languages. Most (if not all) Arabic
dialects spoken in Southeastern Anatolia have created enclitic copulas
out of truncated forms of the personal pronouns. That the copula is
enclitic here in contrast to CMA can be explained by the fact that all
the surrounding languages use enclitic copulas, whatever their origin
(Turkish, Kurdish, Armenian, Neo-Aramaic, ...). Thus, in Mhallami
(spoken in the Turkish provice of Mardin), the paradigms are as follows:

            Independent Pronouns                    Enclitic Copula
1sg        ana                                                -(a)na
2sgm     Ent                                                 -Ent
2sgf       Enti                                                -Enti
3sm       hu:we                                             -we
3sf         hi:ye                                               -ye
1pl         nEHne                                           -nEHne
2pl         EntEn                                            -EntEn
3pl         hEnne                                            -(E)nne

It is interesting to note that, even though the forms are very similar
to one another, the most obvious differentiation is again in the 3rd
person forms, as in CMA. Similar enclitic copulas have developed out of
pronouns in the Neo-Aramaic languages of that area (e.g. Turoyo), where
the forms are somewhat more differentiated, with the greatest
differentiation again in the 3rd person forms. Note that in Mhallami
(and the other Arabic dialects of that area as well), it is obligatory
to use the copula even in the presence of 1st and 2nd person pronouns.
So you cannot say /ana mEn bla:d me:rdi:n/ 'I am from the land of
Mardin' (as you would in other Arabic dialects), you have to say /ana
mEn bla:d me:rdi:n-ana/.

Let me stress, finally, that the older Semitic topic pronouns (as
described above for the "normal" Arabic type) are not yet "pro-copulas".
They simply serve to repeat the subject in order to stress its
topicality ("The man, he (is) sick"). But their reinterpretation as
something that fills the empty slot between the subject and the
predicate seems to be a very small step. As other discussants have
already pointed out, this is by no means confined to Semitic; it seems
to be a common option wherever there is a zero copula slot, be it in the
3rd person only or in the entire present paradigm. It comes as no
surprise that this is particularly often exploited in situations of
language contact where the neighboring language has a full-fledged
copula.

I hope this is enough to clarify the situation a little bit.

(Symbols used: 7 = glottal stop, 9 = voiced pharyngeal, H = voiceless
pharyngeal, E = schwa (central vowel))


Hans-Jürgen Sasse
Institut für Sprachwissenschaft
Universität zu Köln
D-50923 Köln