IMAGE 01h2LexDiffFin01.gif

13

01h2LexDiffFin

(13)ajeLEsuis
"I'm that, I'm the one"(direct object reference)
(13)bj'ENsuis
"I'm one of them"(indirect object reference)
(13)cj'Ysuis
"I'm there"(loactive reference)

that would be agrammatical in Magoua and related dialects even if the underlying
complement were to be expressed overtly.

The set of agreement markers derived from earstwile subject pronouns agglutinated to
earstwile present tense stems of an earstwile copula alternating with the set of
agreement markers derived from unagglutinated older subject pronouns could have
grammaticalized to signal PRODUCTIVELY stativity inherent with verbs such as TO SEE
and extrinsic in the habitual use of verbs such as TO EAT, a salient parameter in the
semantics of Magoua and the French creoles. As it is, no overt alternation exists in
a number of circumstances (such as the subject marker /i/) and a productive process
similar to the le/i alternation in Reunionese Creole never came to be.

The demise of the copula and the rise of noun-class concord is clearly responsable
for the reanalysis of the flectional residue <tait> in (12) as a tense marker with
past reference. Northern varieties of koine French (such as Magoua) have two tense
markers with past reference, one non-negative and another in negative contexts, very
much like many languages "Out of Africa":

(14)a(mouén) ch-temàlàd<je_suis tait malade>
"(I) was sick"
(14)b(mouén) ch-pa-màlàd<je_suis pas (é)té malade>
"(I) wasn't sick"

though in the southern dialects the te/té alternation reduces to /té/. [In the
meridional varieties of koine French (Missouri, Louisiana, Saint-Barth), oral vowels
/i é e è/ shift to /i é è/. The opposition is maintained however in nasal front
vowels /én en/. Magoua has two unrounded front nasal vowels, just like HC. Both
also have two back nasal vowels, the most conspicuous one occurring in /mónn/
"person, people" (with a nasalized [u] and no following dental), a noteworthy lexical
item in Magoua of "African" origin. Cf. Wittmann 1996. The underlying <je_suis
été> is documented also in Bauche 1920:105, 111;cf. Wittmann 1998a:6-7. Bauche is
one of the few records we have of the koine French that survived in Paris in the
second half of the 19th century. Typologically, this surviving variety was more
closely related to the southern koines of the Americas than to the northern ones.]
[There are also a few important non-negative uses for /té/ such as is in statives
/ch-té pansé .../ "I thought that ...".]

Since AGR has zero morphology in creole French except for suppletive object agreement
as discussed under (8b), we would expect the noun-class clitics to survive as
noun-class prefixes of uncertain productivity which we find to be the case in ALL
creole varieties of French (Wittmann & Fournier 1982, Baker 1984, Manessy 1983,
Ndayiragije 1989). Most caracteristically, the inherited distribution of z-/l- in
creole French (with originally vowel-initial items), except for a very few instances
were n- SG is possible or z- PL is plausible, reflects the distribution of
masculine/feminine noun classes in the lexicon of koine French or an otherwise
attested distribution of older records (Wittmann 1984:"Exception faite de quelques
exemples où n- singulier est possible et où z- pluriel est plausible, les préfixes
nominalisants l- et z- du français créole apparaisent en distribution compémentaire:
les noms en l- correspondent en français populaire à des féminins ou à d'anciens
masculins féminisés, les noms en z- à des masculins ou d'anciens masculins."). Baker
(1984) and Ndayiragije (1989) suggest that "agglutination" phenomena of the French
creoles reflect a distribution derived from substrate input, most probably so because
the koine evidence wasn't available to them. Manessy (1983), confronting the
conflicting hypotheses of Baker (in a 1983 preprint version) and Wittmann & Fournier
(1982), qualifies substrate-induction as highly unlikely. Parkvall (2000:81-83) who
had all these and other facts at his disposal concludes notwithstanding that
"substrate factors were somehow at work. I have not been able to identify these,
however."

As this point, I must stress this:All traits of Magoua so far discussed are also

IMAGE 01h2LexDiffFin01.gif

13

Printed for Henri Wittmann <hwittman@cgocable.ca>

IMAGE 01h2LexDiffFin01.gif

14

01h2LexDiffFin

represented to varying degrees in all varieties of Québec French (for the Montreal
situation, see in particular Drapeau 1982). Outside of Québec, closely related koine
dialects survive in the Detroit area, the Prairies (Metis which is distinct from
Mitchif), Missouri, Louisiana and Saint-Barthelemy. The koine-creole continuum of
the 17th-18th century underlying the structural continuities of koine and creole
varieties of colonial French has surviving geographical continua relics in Louisiana
and Saint-Barthelemy isle. The latter situation in particular with its transitions
from a White creole in the east to a White koine in the west, by the very absence of
any African input, constitutes an unsurmountable enigma to all those to whom the
concept of "creolization" is a theoretical comfort they can't do without. The
Saint-Barth continuum of today captures the gradualness of diffusion at the very time
when, for socio-historical reason, the diffusion of the changes such as the deletion
of overt agreement markers froze in time.

The imaginary nature of theMYTHofFRILLS OF FRENCHas an impediment to a
natural derivation of creole French from French is a discredit to any of the current
hypotheses attempting to separate, in a scientifically meaningful way, koine and
creole varieties of colonial French on the basis of the concept of "creolization".

IV. SERIALIZATION IN FRENCH

That leaves as far as I can see just one of Parkvall's pronunciamientos on the frills
of French unattended for:his "There are [in French] suppletive adjectival
comparatives and a great number of Latin-derived prepositions."This most probably
refers to his (2000:23, 70-75) stance according to which koine French serials are
limited to lative heads and that benefactive, comparative and instrumental
serialization are lacking. My stand on this is that:

(15)a. Magoua as well as most related dialects have lative, comparative and
instrumental serialization but no benefactive serialization.

(15)b. HC as well as all the ba-varieties of creole French have lative and
comparative serialization but no benefactive or instrumental serialization.
[Ba-varieties of creole French are:Haitian Creole, the Creoles of a string of isles
extending from Saint-Bart to Trinidad, the Creole of French Guiana, and the various
varieties of Karipuna Creole spoken in the Amapá federal territory of northernmost
Brazil.]

(15)c. LC and the IO creoles would have instrumental serialization to the extent
that the verb can move into TENSE but no benefactive, lative, or comparative
serialization.

(15)d. Benefactive interpretations of /ba/ are recent developments based on
superstrate interferences reanalyzing the ba/pu alternation of Creole in terms of the
à/pour alternation of Standard French.

Since Parkvall admits non-creole French to having lative serialization, I'd content
myself with saying that lative serialization in Magoua combines verbs of movement
such as /àlé/ "go" and /vienn/ "come" with the invariable stems of the perfectly
antonymous verbs /mné/ "carry to" and /kri/ "look for" to form semantically
symmetrical pairs (àlé mné, àlé kri, vienn mné, vienn kri) such as I haven't seen so
far in any of the creoles discussed in current literature I'm aware of (but somebody
no doubt will be able to give evidence to the contrary).

Verb syncopation is the fertile ground on which grew instrumental serialization in
Magoua such as in:

(16)j-à-pong là-hàch j-à-kup là-branch
AGRS-AGRO-take CLASS-ax AGRS-AGRO-cut CLASS-branch
"I cut the branch with the ax."

Sentences such as (16) represent "past-present narratives", much like similar
instrumental serial contructions in many African languages. As has been noted
repeatedly, however, instrumental serialization is not a process open to full
grammaticalization in any language.

IMAGE 01h2LexDiffFin01.gif

14

Printed for Henri Wittmann <hwittman@cgocable.ca>

IMAGE 01h2LexDiffFin01.gif

15

01h2LexDiffFin

For "comparative" serialization in non-creole French, I'd refer Parkvall to Glossaire
(1930:499-500) where <passé> is glossed as "plus que, plus de" ("more than") and the
conjunction <passé que> as "dès que" ("as soon as"). Furthermore, because of the
rule shifting front low vowel /à/ + /r/ to /às/, /pàs (k)/ from earstwhile <parce
que> "because (of)" has been reinterpreted by popular etymology as deriving by
syncopation from /pàsé/ "pass, go/come past, surpass" as in:

(17)amon pti frér i tràvàyplus
[pàs(k) mouén (ch-tràvàj mwéns) dpu j-màlàd]
"My little brother worksmore
[thanI (work less) since I'm sick]"

The merger of "(more) than" and "because (of)" into /pase/ is a common basilectal
feature of all the ba-creoles.

The putative principle of substrate induction for /pase/ is to be rejected on
structural grounds as well. Comparative meaning results from the comparison of two
terms. The first term, introduced by the comparative marker, is the "compared term"
which is being rated against a "standard of comparison", the second term, introduced
by a correlating complementizer. HC and related varieties of creole French use a
marker derived from French forMORE<plus> to introduce the first term and a marker
derived from French forPASS<passé> to introduce the second term as can be seen
from (17b) when the second term has been elided:

(17)bJanpiro (paseMari)(Haitian Creole)
"John (is)MOREtall (PASS[=THAN] Mary)"

(17)ba *Jan ro pase
(17)bb *Jan pase ro

In (17b), /pase/ is a true complementizer requiring obligatorily the second term of
comparison. The same is true for the Magoua /pas/ of (17a). In Yoruba, Gbe and
other relevant African languages, analogues of "surpass" introduce the first term of
comparison in a manner identical to analogues of "more" and do not require the second
term:

(18)aó tóbi(mílo)(Yoruba)
"he (is) bigPASS[=MORE] (meGO[=THAN])"
"He is bigger (than me)."

(18)bàfì lè suè(dàdì)(Gbe)
"mouse is smallPASS[=MORE] (cat)"
"The mouse is smaller (than the cat)."

which makes the two constructions, (17) and (18), fundamentally dissimilar.

This leaves serializing with /ba/ as the utlimate residue. In all the relevant
varieties, /ba/ alternates with /pou/ as in:

(19)amwen te pote liv-la ba Mari(Haitian Creole)
"I PAST bring book DET SER Mary"
(19)bamwen te pote liv-la pou Mari
"I PAST bring book DET PREP Mary"
(19)bbmwen te pote Mari liv-la
"I PAST bring Mary book DET"

(19ba) being a non-synonymous structural alternate for (19a) and (19bb) being a
synonymous alternate for (19ba) though structurally dissimilar. As has been
demonstrated elsewhere (Wittmann 1995:322-25, Fournier 1996, Wittmann & Fournier
1996:261-64), only (19ba) and (19bb) can receive an unambiguously benefactive
reading. The alternation of the internal arguments in (19a) and (19b) reflects a
semantically constrained lative-benefactive dativity alternation universally
available in all languages (with data from English, Standard French, Magoua and Gbe;
cf. Larson 1988 for the theory):

(20)aI gave (brought) book to (for) Mary(English)
(20)bI gave (brought) Mary a book

IMAGE 01h2LexDiffFin01.gif

15

Printed for Henri Wittmann <hwittman@cgocable.ca>

[made with GoClick]