No Proto-Celtic?

[Gabor Sandi]

[ moderator edited ]

I agree that the expression "free word order" is something of a misnomer. In
heavily flexional languages (e.g. classical Latin, Greek and Sanskrit, as well
as my native Hungarian) you can mix up elements more than you can in, say,
English and Chinese, but every specific order other than the default one brings
some additional semantic or grammatical information.

E.g. in Hungarian, you can say "La'tom a kutya't" (I see the dog), and this is
the default (unmarked) order. If you say "A kutya't la'tom" (the dog I see),
there is a contextual inference: "I see the dog, but not something else that
has just been mentioned (e.g. the cat), or that I am thinking about". There is
definitely semantic information added by the changed word order. I would guess
that the difference between Latin "canem video" and "video canem" similarly
brings in some kind of inferential distinction, but I leave it to people
knowing more Latin than I do (this wouldn't be hard...) to tell us what this
difference is.

Hungarian leaves off the copula in the third person present (sing. or plural)
in sentences of the type "the house is big" and "he is the president", although
not in sentences like "he is in London". As a result, word order acquires
grammatical significance in the following pair of constructions:

"A ha'z piros" (the house is red)

"A piros ha'z" (the red house)

There is a problem, however, when we try to extrapolate from evidence in
flexional languages to PIE, and specifically to the issue of the origin of IE
personal endings in verbs. If, for example, the singular endings -m, -s, -t of
athematic verbs are to be derived from postposed personal pronouns, we are
looking at a stage in the language which may have been as isolating as modern
Chinese for all we know, with a fixed word order. They might have said:

*Wodon i es me (I am/was in the water)

*Wodon i es se (you are/were in the water)

*Wodon i es te (he/she/it is/was in the water)

To us it may look strange that the pronoun would have come after the verb and
not before, but this may happen, and it is standard in modern Irish, isn't it?
E.g.: Scri'obhann SE' litir (he writes a letter).

As for postposed particles for nouns (as the proposed *i for "in" above), there
are plenty of parallels in modern languages, from Hungarian "A hajo' a hi'd
ALATT van" (the boat is UNDER the bridge), to Japanese "Fune ga hashi NO SHITA
arimasu" (idem).

The coagulation of postposed particles and pronouns with preceding nominal and
verbal elements is for me one of the most fascinating aspects of pre-PIE
linguistics. I wonder what other participants in this group think about it?

With my best wishes to all,
Gabor Sandi