No Proto-Celtic?

[Thomas McFadden]

>> Only that is not true.  Even a free word order language has a preferred,
>> *neutral* order.  In Latin that order was SOV.  Word order variations in
>> such languages are used to encode variations in emphasis and attitude -
>> they are *not* meaningless.

>     But in a flexional language, especialy of the old IE variety, they are
> of secondary importance. They belong to stylistics, not to grammar.

>   Even in modern French the difference of meaning between 'une tres
> grande maison' and 'une maison tres grande' is minimal ; you can use
> both sentences freely to express exactly the same thing (no difference
> in style or emphasis). Modern Slavic languages could provide many more
> examples.

of course it depends on what you mean by grammar, but under my conception
of the term i have to disagree with you strongly on this.  flexibility in
word order is actually regulated rather tightly by the grammatical system
of a given language, it seems.  take the French example you gave.  French
grammar allows the two possibilities that you gave, but not, e.g. `une
maison grande tres', or any number of others that are possible e.g. in
Greek.  German is another great example.  there is a great deal of
flexibility in what element comes first in a sentence, but once that has
been decided, the order of the rest of the elements is pretty rigidly
determined.  note also in comparison to French that German allows 'ein
sehr grosses Haus' but not 'ein Haus sehr grosses'.  it seems to me you
have to encode these possibilities in the grammar, rather than leaving it
up to something called stylistics outside the grammar.  if stylistics is
outside the grammar then it shouldn't be specific to a language, it should
be the same for all people, or maybe it could be culturally determined or
something like that, but those possibilities are clearly falsified by the
word order possibilities of world's languages.  each language has its own
distinct pattern of word order flexibility, so we can't say that such
flexibility results from a common human `stylistics'.  furthermore, a
person who grows up in a bilingual culture will not use a single
stylistics for the two languages that could be attributed to his culture
or his upbringing, i.e. a person in, say Alsace-Lorraine who is bilingual
French-German will have no trouble following the rules for the individual
langauges properly that are mentioned above.  the word order possibilities
are a part of the language itself.  of course what variant a person
produces at a particular moment will depend on external factors, but the
possibilities that are available to him are language-specific.

and for this argument its actually not crucial whether a big semantic or
pragmatic difference is tied to the choice between two variants like 'une
maison tres grande' and 'une tres grande maison'.  whats important is that
while these two variants exist in French, a whole bunch of others
don't.  and in most cases where multiple variants exist, there is an
identifiable difference, even if its hard to find.  usually it's not a
difference in the truth-conditional semantics, but there's perhaps a
difference in what is implied, or in the conditions where you would say it.