[Lingtyp] instant resumption

[Christian Lehmann]

Thank you everybody for your helpful contributions to my problem. Let me 
start by asking those of you who mentioned relevant publications to 
concede me a PDF if they have it. (An emeritus no longer has that easy 
access to library services.)

Far from wanting to terminate the discussion at this point, let me take 
up some issues that have been raised and we might wish to pursue. Since 
some of the interventions appear to make slightly different conceptual 
presuppositions, I ask for your indulgence for defining some concepts 
just to make sure that we are not talking past each other.

Disregarding first and second person pronouns, the distinction between a 
pronoun and a determiner is essentially not a semantic one, i.e. it does 
not (necessarily) involve the morphosemantic features (like 
demonstrative, interrogative etc.) marked on these formatives, but is 
just merely a structural one: a pronoun can represent an NP, a 
determiner is a constituent of an NP. (Both of these definitory 
stipulations need some refinement, which we can forego here.) Many 
proforms, including importantly demonstratives in many languages, have 
both uses. This is also the case in Cabecar.

Some demonstrative pronouns may be used as an anaphor. Where there is a 
deictic paradigm of demonstratives, usually not just any demonstrative 
can serve as an anaphor. Commonly it is an unmarked member of the 
paradigm, typically the one that some of us call medial. This is, again, 
the case in Cabecar. Sometimes (not in Cabecar), both ‘this’ and ‘that’ 
can take up referents of the preceding context.

Now if such a demonstrative appears at the margin of an NP, the problem 
arises whether it is a pronoun or a determiner. If determiners have a 
fixed position other than at the end of the NP, a demonstrative 
appearing in that position cannot be a determiner. That is, again, the 
case in Cabecar.

Besides demonstratives, third person pronouns can serve in intraclausal 
anaphora. In my previous post, I silently presupposed that they, too, 
can be used as instant resumptives (though not in Cabecar). The 
conditions under which one or the other is used are a topic for an 
entire workshop.


@ Stef, Edoardo, Yury:

There are different processes of intraclausal anaphora. Only instant 
resumption leads to an appositional construction. Another possibility is 
for an anaphoric or personal pronoun to attach to the dependency 
controller and then develop into a pronominal index or referentiality 
marker. Such personal pronouns are typically clitic. The Cabecar 
demonstrative is definitely not clitic. On the contrary, it serves as a 
prop for clitics.

The Koine personal pronoun in the relative clause is superficially 
another case of intraclausal anaphora. However, I think it first of all 
represents the domain nominal inside the relative clause, thus 
introducing a new relativization strategy which, in Modern Greek, is to 
supersede the relative pronoun strategy.


@ Randy:

The two Old Chinese examples do look similar to the Cabecar case. The 
Cabecar resumptive has focus in some, but not in all cases. It also 
follows second mentions (instead of replacing them). Interestingly, 
Cabecar /jé/does not develop into a copula. Instead, it takes an 
existent copula-like formative as enclitic. I will probably soon launch 
another post on this formative, which also functions as an 
information-structure articulator.


@ Marianne:

I have too few recordings to tell for sure. The following seems to be 
the case: The resumptive can always combine with the antecedent into one 
phonological word. With left-dislocation, however, it may optionally 
follow a hanging intonation plus pause, as in many other languages.


@ Zygmunt. Matthew, James:

Some of you mention demonstratives following different kinds of nominal 
expressions. Assume the language in question has postnominal 
determiners. One criterion for distinguishing resumptive pronouns from 
determiners here is the completeness of the paradigm: If it comprises 
all of the deictic values present in demonstratives, the formative in 
question is probably a determiner. If the formative that can appear in 
that position reduces to the anaphoric member of the paradigm, it is 
probably a resumptive pronoun.

This regards, in particular, proforms following relative clauses. In the 
spirit of the discussion with Jürgen [next], it seems quite possible to 
me that a “recapitulating” pronoun serving the following context as a 
point of reference for the preceding relative clause may be especially 
necessary for such constructions whose nominal nature is not 
sufficiently explicit.


@ Matthew:

In the example about /those people who have not …/, /those/is clearly a 
determiner. I would guess it emphasizes the formation of a subset (or in 
the singular, the singling out of an individual).


@ Jürgen [sent directly to me]:

Some formatives which combine with nominal expressions may have 
distributional restrictions in the sense that they do not combine 
directly with nominalized clauses. In English, e.g., most prepositions 
do not govern a /that/ clause. Cabecar postpositions do not have such a 
problem. They simply follow what happens to be the last constituent of a 
nominalized clause. Still, it may be true that instant resumption is the 
more preferred the more complex a nominal expression is and the less its 
nominal nature is transparent. The neutral demonstrative may be some 
kind of NP /par excellence/, something that stands for the pure category 
‘NP’. It represents its antecedent qua nominal expression for the 
context, and that is what it does as an instant resumptive. (Discourse 
is full of grammatical formatives which do nothing but redundantly 
signal the grammatical category of something which belongs to that 
category, anyway.)


@ Don:

These data do look similar to the Cabecar case. However, assume there is 
such a thing as instant resumption as characterized before. Then the 
question arises whether it is a possible cradle for the development of 
determiners, as some had assumed formerly for Cabecar. You then need to 
answer at least two questions:

  *

    What happens to the semantic determination (or referentiality) of
    the NP once the demonstrative becomes its determiner? Instant
    resumption – at least in Cabecar – is possible with semantically
    indefinite antecedents, as in “fruits, cereals, cabbage, that we
    grow” [real example]. The resumptive is necessarily definite. Once
    it is converted into a determiner, the NP would then be definite.

  *

    How does the entire deictic or referential paradigm of postnominal
    determiners arise? The resumptive is definite and deictically
    neutral. How do indefinite proforms and demonstratives with
    different deictic features join the paradigm of determiners?

My inability to answer these questions has led me to despair about the 
hypothesis that the prenominal determiners of Cabecar are being replaced 
by postnominal ones or, more generally, that postnominal determination 
may arise through the grammaticalization of intraclausal anaphors.

Now on your questions:

  *

    Instant resumption in Cabecar is indifferent to the referentiality
    of the antecedent. It is also indifferent to its topic/focus
    distinction, with one exception: it does not follow the third person
    pronoun, which presupposes givenness of its referent.

  *

    The resumptive, not the antecedent, carries the plural suffix and
    the postposition, if any, that would attach to the antecedent if
    there were no resumptive. However, such contextual circumstances are
    neither a necessary nor a sufficient condition for the resumptive to
    occur.

  *

    The phonological conditions remain to be investigated.

  *

    The resumptive does follow first and second person pronouns,
    amazingly. This is one of the circumstances which make me diagnose
    an advanced degree of grammaticalization.


@ Julie:

Although a preposed demonstrative cannot originate by instant 
resumption, the variation of prenominal vs. postnominal determiner does 
raise the question of how these constructions and the sequential order 
come about. The syntactic function of the determiner, i.e. the nature of 
its syntactic relation to the lexical nominal expression, is possibly 
unique in grammar. One symptom of it is the fact the noun phrase has 
been renamed determiner phrase in certain quarters. It seems quite 
possible to me that it starts out as an apposition of a demonstrative 
pronoun and a lexical NP. Next thing we would want to know is what 
determines which of the two comes first …


-- 

Prof. em. Dr. Christian Lehmann
Rudolfstr. 4
99092 Erfurt
Deutschland

Tel.: 	+49/361/2113417
E-Post: 	christianw_lehmann at arcor.de
Web: 	https://www.christianlehmann.eu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20210920/388c0dbc/attachment.htm>