Object-initial languages

[Hartmut Haberland]

I think Östen is right that we have to consider our notions of basicness
and markedness. An example I never had thought of but which I came
across in my teaching recently:

Languages like German, Danish, Swedish have a class of verbs that have
strong past tense endings in the present tense (modal verbs & the verb
"to know"), like

German     ich weiß 'I know' (same ending as ich kam 'I came', not as
ich sag-e 'I say')
Danish       jeg ved (same meaning) (as jeg kom, not as jeg sig-er), etc.

Now the usual explanation is: these verbs are really old IE perfects
that have acquired a present meaning (cf. Latin novi and Classical Greek
oida). Thus they are called preteritopresents.

But what is basic and what is marked here? I don't know of any relevant
studies of language acquisition, but I'd guess that children acquire ich
weiß before they acquire strong pasts. So maybe the past tenses of
strong verbs are modelled after the alternative present of these verbs.

Another case are languages like Modern Greek, where VS, VO, VSO, VOS,
VAdv etc. are very common constituent order patterns, and statistically
probably more frequent than SVO (Greek is pro-drop, of course). Also, OV
or OVS are very common if the object is co-indexed on the verb with a
clitic pronoun (some call it an agglutinative object marker). Still,
many people would maintain that Greek is SVO, simply because they assume
that some order must be basic. But what are the arguments for this? Does
Greek have to have a basic constituent order?

Hartmut Haberland


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/funknet/attachments/20021025/19be9a9c/attachment.htm>