Re: Tense marked on nouns

----- Original Message -----
From: "Mark P. Line" <mark@...>


Sally uarly hdaro:
> >  What are the clause-level markers in which languages
> > that tend to be involved with valence assignment that could lead to a
> > development, in your mind, of tense marked on nouns?   Any in English?
> > Actually, you might answer this below in your example.
>
> My point was not that I think certain types of noun morphology might lead
> to clause-scope tense marked on nouns. My point was simply (and relatively
> trivially) that clause-scope morphology doesn't always have to be on the
> verb. A good example of this is *case*: case expresses relations at the
> level of the clause, it doesn't modify the meaning of the marked noun as
> such. I specifically had case in mind when I mentioned 'valence assignment
> and/or pragmatic functions' above: that's what case is for, it has a lot
> to do with the marked noun (i.e. its role(s) in the clause), and is
> usually marked on the noun.
Okay, gotcha.  Thanks.

> I was contrasting this with clause-level tense marking, which does *not*
> have a lot to do with any particular noun participant in the clause: tense
> is marked on the verb when it's marked morphologically. (At least, I
> haven't seen a counterexample yet.)
Only in some conlangs? :)  elry krespr, "past-I write"?  (I wrote)  Unless I
still misunderstand you here.  When the VERB is marked morphologically, you
say?  or when TENSE is marked morphologically?  Most Teonaht verbs have next
to no tense/aspect/person/number morphology.  Only moveable affixes.  They
are marked, however, for volitionality.

> > Mark responds [to Jim's snipped question]:
> >
> >> Glossed as you have them here, my answer would be 'no'. Tense is a way
> >> to
> >> distinguish temporal relations at the clause level, thus distinguishing
> >> the following:
> >>
> >> (a) John gave yesterday's paper to the former president.
> >> (b) John's giving yesterday's paper to the former president.
> >> (c) John's gonna give yesterday's paper to the former president.
> >>
> >> There are *other* temporal relations expressed in these examples (by
> >> modifying the nouns), but they're not tense because they're not
> >> expressing
> >> temporal relations of the event to which the clause is referring.
> >
> > Okay, this I understand.  Is this an example of what I was seeking from
> > you above?   I.e., clause level markers involved with valence assignment
> > and other pragmatic uses?
>
> No, these examples were meant to demonstrate the presence of two different
> kinds of temporal relations in the same clause -- because that's where a
> lot of people (including numerous linguists who've published on the
> subject) are getting confused.
Okay.

> The three clauses differ only in *tense*, which is a kind of temporal
> relation at the clause level. The entire event of John giving something to
> somebody -- that is, the referent of the entire clause -- occurs at
> different times with respect to the time of utterance in these three
> example.
>
> But there are other temporal relations expressed in these examples as
> well. It's *yesterday's* paper, and the *former* president. These nouns
> are not being marked for *tense*, which is a clause-level category,
> because "yesterday's" and "former" are just modifying the noun -- not
> expressing clause-level relations of any kind.
Right.

> (The literature on the subject is confusing because some linguists have
> chosen to refer to certain kinds of noun morphology as "nominal tense"
> because it has to do with time. I do not follow this practice, and
> discourage others from doing so on ontological grounds.)
Agreed.

> >> A language that truly marked *tense* on nouns instead of on the verb or
> >> periphrastically in the clause would have to something like this:
> >>
> >> (a) Djanden yestadepela pepa wantaimpela perezent gif.
> >> (b) Djanbi yestadepela pepa wantaimpela perezent gif.
> >> (c) Django yestadepela pepa wantaimpela perezent gif.
> >
> > "The then John"?  "The now John"?  "The future John"?   Yeah, that's
cool.
> > If I have that right.  But how does it differ from Jim's example?
>
> No, no. I intended *these* examples (a), (b) and (c) to have PRECISELY the
> same meaning as the earlier (English) examples (a), (b) and (c). Please
> compare them again.
I have.  And I wasn't saying that you didn't intend them to have the same
meaning as the earlier examples.  What confuses me is what it is that you
are counter-arguing; it seems to me that Jim gave almost precisely the same
example with "dog" (now I wish I hadn't snipped it above!), his only
omission being that he didn't put these words in a clause which would
illustrate their temporal relations within a sentence.  In other words, the
clause is needed to see how the tense-marked noun expresses valence?

> What I've done in the impromptu pidgin examples is to
> mark *tense* (REAL tense: an expression of temporal relations at the
> clause level) on the subject of the clause instead of any other way
> (morphologically on the verb, or periphrastically).
Yes, that I understood the first time around.  I'm trying to get a sense of
what is being argued.  I'm a linguistic autodidact, to a certain extent (my
formal training being in historical linguistics).  This listserv helps me
learn.  And this issue of tense marked on nouns is fascinating to me, since
I started marking tense on pronouns years ago for my one and only conlang.

> In the English examples, we have:
>
> (a) give+past = "gave"
> (b) give+present = "'s giving"
> (c) give+future = "'s gonna give"
>
> Clearly, tense (REAL tense, not "yesterday" or "former") is being marked
> on or around the verb here.
Yes, yes, yes.

> In the pseudo-pidgin examples, we have:
>
> (a) give+past = SUBJECT-den
> (b) give+present = SUBJECT-bi
> (c) give+future = SUBJECT-go
>
> The suffixes -den, -bi and -go precisely do *not* modify the subject
> 'Djan' to make it "the then John", "the now John" and "the future John".
> That's precisely the distinction that I'm trying to demonstrate!
Okay, bad choice of words on my part.  To my mind a language could
potentially say "the then-John give yesterday-paper to former-president" and
MEAN "John gave yesterday's paper to the former president."  And "then-"
would be an affix, and not a clitic.

> The
> suffixes -den, -bi and -go mark CLAUSE-LEVEL TENSE: the only way to mark
> tense (REAL tense) in this language is to put -den, -bi or -go on the
> subject.
>
> This, therefore, is an example of what a language would be like that does
> what the initiator of this thread was asking about: tense being marked on
> a noun instead of anywhere else (like on the verb or periphrastically).
Aha!!  But this I understood, or so I thought.

> >> And even here, we only have morphological tense marking on the noun
> >> 'Djan'
> >> if the attached morphemes '-den', '-bi' and '-go' really are *affixes*
> >> and
> >> not clitics.
> >
> > Why is that a problem?  What difference does it make if it's an affix or
a
> > clitic, which are very similar animals in my ignorant sense of these
> > things?
>
> We were being asked about nouns marked for tense. A noun to which a
> periphrastic tesne marker has been *cliticized* is not being marked for
> tense: it's simply joined phonologically with the next word, which
> happense to be the tense marker.
Yes, I remember.

> I made the distinction because there are *lots* of languages (including
> English) that *appear* to mark tense on nouns because there are
> tense-marking clitics that can attach to nouns. But as I pointed out in an
> earlier post in this thread, the sentence
>
>    (d) John is at work.
>
> no more marks tense on the noun than does
>
>    (e) John's at work.
>
> In (e), "'s" is a clitic, and marks tense (among other things). It is
> *not* a tense affix, and English is *not* an example of a language that
> marks tense on nouns (hence my distinction between clitics and affixes).
Yes, I remember your post, and remember agreeing with it, pace Peter.  :)

> > Now if each noun were internally changed, in the way some cases are, for
> > tense, would that solve the problem?  Djan, Djain, Djeno?  But then,
> > didn't cases start out with noun + affix?
>
> It's not a problem: I was merely stipulating that the tense morphemes of
> the pseudo-pidgin have to be affixes to qualify as an example of what we
> were looking for. Sure, you could use *any* morphological process instead:
> tone change, ablaut, infix, prefix, suprasegmental nasalization or
> devoicing, whatever.
I thought it was a problem because of your wording.   You said above

    "And EVEN here, we ONLY have morphological tense
    marking on the noun 'Djan'  if the attached morphemes
    '-den', '-bi' and '-go' really are *affixes* and not clitics."

My emphases.  "Even" and "only" seemed to express deficiencies; I'm sorry if
I misunderstood you.  I thought that for tense marked nouns there had to be
something other than affixes to complete the picture.  Now I understand you
to mean that morphological tense can only mark the noun if it is an affix
(we ONLY have), and not a clitic.  Re: prior argument that John's is a
tense-marked noun.

> > Thanks for bearing with..., Sally
>
> My pleasure. Hope this was helpful.
Yes, very!  This argument is super subtle, and just when I think I'm getting
used to it I get thrown.

yry nwetis firrimby,
Sally Caves
scaves@frontiernet.net
Al eskkoat ol ai sendran rohsan nuehra celyil takrem bomai nakuo.
My shadow follows me, putting strange new roses into the world.

Replies