On Sat, Feb 12, 2011 at 5:54 AM, 1 2 <[log in to unmask]> wrote:
> My questions are:
> - What are other possible strategies for non-subject relativization?
> - The examples I gave above are for nominative-accusative languages. How
> would an ergative-absolutive language, for example, render this situation?

Japanese dispenses with relative clauses completely. It just doesn't
have relative pronouns at all. Instead, it uses participial phrases
for just about everything. This is slightly obscured by terminology:
Japanese participles are usually not referred to as such. Instead,
participle inflections are referred to as the "plain form", since they
can double as finite verbs in casual speech (and the "plain" nonpast
affirmative is the citation form for verbs).

Basically, you just stick a clause in front of the noun you're
modifying, with the verb of the clause in the plain form, and the
modified noun acts as one of the otherwise unspecified participants in
that clause. The way it's been explained to me, it can be any
participant. In practice, it really seems to be limited to the core
arguments and any participant that is considered particularly
important (re: commonly specified) for the verb, which is usually
recipient or location. So for "iku", to go, it could be the subject or
the destination; for "ageru", to give, it could be the subject,
object, or recipient.

I adopted this approach, more or less, for my Ilion project.
Originally I was using relative pronouns, because I didn't really know
there was another way, along with resumptive pronouns, because I'd
read about it in the LCK (or maybe Pablo David Flores' tutorial?) and
the flexibility appealed to me (along with the fact that Ilion grammar
generally doesn't play well with gaps). I changed my mind later,
though I kept the resumptive pronouns, and participial phrases follow
instead of precede the modified noun.

On Sat, Feb 12, 2011 at 1:47 PM, Roger Mills <[log in to unmask]> wrote:
> --- On Sat, 2/12/11, And Rosta <[log in to unmask]> wrote:
>> From: And Rosta <[log in to unmask]>
>> Subject: Re: Non-subject relativization strategies
>> To: [log in to unmask]
>> Date: Saturday, February 12, 2011, 1:24 PM
>> Logan Kearsley, On 12/02/2011 16:01:
>> > On Sat, Feb 12, 2011 at 7:21 AM, Philip Newton<[log in to unmask]>
>> wrote:
>> >> just as English can't relativise things such
>> >> as "This is the man that my wife is the daughter
>> of Sandy and ___").
>> >
>> > Personally, I don't find any problem with that
>> sentence.
>> I suspect your judgements about which sentences you find
>> some problem with is not a useful indicator of whether the
>> sentence is English (in any idiolect thereof). Do you use
>> such a construction, or does anybody else? If yes, show some
>> attested examples, which will be of great service to
>> grammarians. If no, then why not? The reason why not is that
>> there is some sort of problem with the construction. (The
>> construction is plainly useful and easy to understand, so
>> the problem must be syntactic.)
> Note that it's perfectly OK if you substitute "and" for "that" (and fill in "him" in the blank at the end). It's also semi-acceptable (but see below) if you leave off the "Sandy and ___" and just stop with "of".

This native speaker (Northern California, no yiddish influence AFAICT
beyond calling some people "schmucks") finds "This is the man that my
wife is the daughter of." perfectly acceptable in colloquial speech.
It's only when the gap is part of a coordinate construction that it
falls apart. "This is the man that my wife is the daughter of Sandy
and." leaves me waiting for the other shoe to drop.