Re: Branching typologies [was: Re: "easiest" languages, SE Asian word-order typologies]

:::delurking:::

In a message dated 9/26/01 3:42:47 PM, trwier@MIDWAY.UCHICAGO.EDU writes:
<snip>
>I'm curious: what is the most common branching typology for our
>conlangs? Phaleran is very left-branching: SOV word order, relative
>clauses before the noun they modify, adjectival particles before
>the noun they modify.  Degaspregos was/is more right branching:
>although it has mainly SOV word order, relative clauses and
>adjectives usually come after the modified noun.
Eloshtan is somewhere in the middle, then. It is SOV, and mostly
left-branching but just right-branching enough to make drawing tree diagrams
a pain (but still lots of fun). Adjectives come before the noun, but relative
clauses come after. It uses postpositions. Only a few types of pp phrases can
occur with a noun (pretty much only accusative, dative, and possessive ones)
and these must come before the noun. Any other pp phrases must be turned into
relative clauses (which then come after the noun).

The same set of pp phrases that can occur before nouns also accur before
verbs (hence the SOV word order) (oh and there are also pp phases with "le",
which function for verbs somewhat as adjectives do for nouns, though I
wouldn't call them adverbs), but all other pp phrases come after the verb.
And there's one other thing - there is a particle "pe" that marks proper
nouns which would probably come after the noun in a truly left-branching
language, but it comes before nouns (and adjectives too).

It seems to me like Eloshtan in the future would develop an OVS structure,
and indeed one of its sisters did just that. I believe I read somewhere that
VSO and SOV were the most natural/ideal word orders, but then I was thinking,
if you analyze them at just that top level, SOV is actually right-branching,
isn't it? The S is the head on the left, and then the right goes to a verb
phrase, made up of O and V. But then the rest of the grammar is suppsoed to
be left-branching. With OVS, on the other hand, it is left-branching from the
beginning. And then, for similar reasons, SVO would seem more truly
right-branching that VSO. Does that make any sense? I suppose there are other
factors that I'm not taking into account.

>You know what -- we should compile a brief synopsis of all types
>of conlinguistic typologies.  It would shed some light not so much
>on language, but rather more likely, on the conlanger population.
>
>==============================
>Thomas Wier <trwier@...>
>
>"Aspidi men Saiôn tis agalletai, hên para thamnôi
>  entos amômêton kallipon ouk ethelôn;
>autos d' exephugon thanatou telos: aspis ekeinê
>  erretô; exautês ktêsomai ou kakiô" - Arkhilokhos
Josh Roth
members.aol.com/fuscian/eloshtan.html