Object-initial languages

[Bittor Hidalgo]

(A very long message, sorry!)

I studied in some detail the word-order of Basque, a more or less free word
order language, with very rich verbal morphology (besides semantic
information and time, mode and aspect features, it gives detail of
subject -ergative or absolutive-, direct object -accusative-, and indirect
object -dative, and at times, it also informs about the sex of the
interlocutor, and the even level of familiarity of the speaker to him/her;
all that, inside a single word -sometimes two, ...-). Of course, Basque is a
pro-drop language.

Reputedly, Basque is a non-rigid SOV language (Greenberg, De Rijk). But
statistically, in the conditions of Greenberg, the most frequent order in
Basque (Hidalgo, 1995) is clearly SVO (between 40 and 65% of the sentences),
and not SOV (between 15 and 30% of the sentences). And the object initial
orders (OSV/OVS) account for between 10 and 20% of the sentences. So are
both, oral and written language, to the exception of the written language of
some artificial purist syntactic school that, curiously, has triumphed in
the standard written style of the last 30 years (De Rijk).

Anyway, in the absence of an open subject -a very common instance in
Basque-, surely OV order is commoner than VO, although we don't have
statistical data.

The question, the hypothesis, I want to raise here (to test), is that, in my
opinion, the verb acts, concerning its position respect to its most close
complements -including S and O-, I think in all languages, but at least in
the languages I more or less know -besides Basque, the meriodional European
romances including French, and English or German-, it works as a grammatical
function word, that according to Wackernagel's law, goes clitic respect to
some complement (let's call it X). Two possibilities then (Venneman,
Hawkins): XV (verb second, somehow, inside this complex) / VX (verb first,
at least inside the complex).

This complex constitutes normally a minimal intonation unit (or accentual
phrase in generative terms). In the complex, presumably all languages in the
world, would accentuate the complement (unless pronominal, deaccentuated,
...). So we'll have VX languages, of rising accent (languages with proclitic
elements, prepositions, ... -all verb-initial languages, but also SVO
languages with rich verbal morphology like Spanish, Italian, Catalan, ...-),
and XV languages of descending accent (with enclitics, postpositions,
... -like Basque, and I suppose Turkish, Japanese at least-). But there are
languages in the between. Languages that historically had been XV(OV), and
conserve some of its features, but that in contact with other VX languages,
or in its process to become culture languages, had become also XV. So could
be the evolution of Latin (Wackernagel, Marouzeau, Panhuis), and of Latin to
romances, but also of the languages now called second-verb languages like
German, were any kind of complement can occupy the first position, or
languages like English or French, where the first position has been almost
completely grammaticalized historically to subject position, because of the
poor verbal morphology. It could also be the case of languages like
Hungarian, where the first position in the XV complex is reputed (like
erroneously in Basque) as focus position, although sometimes it could lodge
also focus (specially in short sentences, ...). I don't know if something
similar happens in Finnish. I would like to know.

Languages like French or English that have grammaticalized SV position as a
realization of XV, act, afterwards, as VX languages (rising accent) in
respect to other complements as O, adverbs, etc.. (V doesn't go alone, but
in the almost inseparable complex [SV]). Old French and English, where the
grammaticalization of SV is not completed, and the verb morphology was
richer, admitted also other complements before the verb (adverbs, even
objects, ...), just like still today are normal "there is ." or some "adverb
+ verb" or inversion structures in both languages. And German still works as
an XV language doubly: XV (verb second language), that in complex verbs
repeats the XV structure with the main verb (with a tonic complement before
it). I don't know if some reputedly rigid final language like Japanese
(although it seems that in conversation, Japanese is not so rigid -Clancy,
1982-) could accommodate somehow in this description.

I would like to know what you think about this point of view. I would also
like to know about the recent evolution of some culture languages, like
Hungarian, from a potentially XV structure, to a prescript obligatory SOV
order (in the XIX century), and a posterior evolution to ([XV] + X)
structure, or the artificial role of grammarians (Port Royal) in the
fixation of SVO order in French abandoning the so called inversions (even in
English), or the role of German grammarians in the artificial rigid an
obligatory final position of verb in subordinate sentences, ... The question
is if we can reduce the 6 type word order typology to 2 type word order
(with some mixed situations), forgetting, at last, the inaprehensible
concepts of "basic" or "neutral" word orders, and assigning the order of all
other complements depending exclusively of contextual features, information
structure organization, and intonational (accentual) segmentation of phrases
and sentences, and in cases of specific grammaticalization processes, like
those of French, English, ...).

Too long. Too many questions. Forgive me, and surtout forgive my macaronic
English.

Bittor

---------------------------------------------
Victor Hidalgo Eizagirre
k/ Buztintxulo, 72, behea
20015 Donostia / Euskal Herrria / (Spain)
Tfnoa eta faxa: 943-282192
posta-e: v.hidalgo at euskalnet.net