No Proto-Celtic?

Thomas McFadden tmcfadde at babel.ling.upenn.edu
Thu May 24 17:16:12 UTC 2001


i may regret bringing this up, but here goes anyway  (note that i havent
read Watkins' article yet).  'free word order language' as it is usally
used to describe languages like Latin doesn't mean that any order goes in
any situation, it just means that there's a lot of flexibility and a large
group of possible orders that can be used, each under the proper discourse
situations.  so while SOV, VSO and OSV might all be possible word orders
in Latin, for example, they aren't all equally possible, and they wouldn't
all be used in the same situation.  different orders are used to focus or
topicalize different elements in the clause, so e.g. if you were asked
who Brutus murdered, you might answer with an OSV or OVS sentence, but if
you were asked who murdered Caesar, you might use SOV or SVO.  this sort
of correlation between word orders and discourse contexts has been shown
to exist in `free word order' languages like Finnish and Russian, and i
expect Latin was no different.  crucially, in these languages there is a
clear sort of default word order that tends to be used in neutral
circumstances.  e.g. in Finnish the relative ordering of subject and
object is largely determined by whether the items referred to are old or
new to the discourse.  that is, a person or thing that was mentioned in
the previous sentence will tend to occur early in the sentence, while
something being mentioned for the first time that day will tend to occur
at the end.  so when the subject is old and the object is new, you get
SVO, and when the subject is new and the object is old, you get OVS.  but
what about when both are old or both are new?  this is one of those
neutral circumstances, and here we get SVO.  OVS sounds extremely odd in
such a situation.  so in spite of the great deal of variation in Finnish
word order (other orderings are also possible under other circumstances),
it makes a lot of sense to say that Finnish is basically SVO, but can have
other orders if the circumstances demand it.  so it's not the case at all
that the SOV/VSO controversy has no relevance for Latin.  in addition to
the possibility of all sorts of crazy word orders, every student of Latin
also knows that the 'typical' word order is SOV.  every intro. textbook of
Latin has a brief section on word order where SOV is given as 'typical' or
'normal' or something like that, followed by an explanation of how
commonly we find something deviating from the norm.  good textbooks
will go on to note that this isn't haphazzard, but that there are reasons
for different word orders to show up, like which word is considered most
important for some reason, which word a sentence is 'about' etc.  in fact,
i think studies have been done of word order patterns in the Latin texts
that show they do have discourse motivations.  i'll try to find info on
them when i get a chance.

>     And Calvert Warkins answered that both were probably wrong (his article
> on syntax reprinted in his "Selected Writings", IBS), PIE being a flexional
> language, with free word order (every student in Latin knows this) where the
> SOV / VSO controversy has no relevance.

>     XD



More information about the Indo-european mailing list