[Lingtyp] Topic and focus markers with other functions

Sat Aug 3 15:31:08 UTC 2019

Dear all — There are two important distinctions that bear on this discussion: 

* Semantic vs. pragmatic meaning (in particular, conversational implicature)

* Polysemy vs. homophony.

For the following, I’m assuming two expressions (morphemes, phrases, constructions) X and Y, which cooccur (not necessarily adjacently), and when they do, Y is associated with meaning B, while X is associated with meaning A across all of X's uses. Schematically:

X … Y
 |       |
A      B

Szenario 1: When Y occurs in the context of X, the interpretation of Y frequently, but not in all cases, takes on interpretation B. In this case, the co-occurrence of X and Y may conversationally implicate B, but does not semantically express B (in other words, contribute a B entailment or B presupposition to utterances that feature it in suitable contexts). 

This is the case, for example, of telicity in Russian, which is expressed overtly via verbal prefixes (simplifying slightly here for the sake of argument) and statistically associated with semantic definiteness, which is not overtly expressed. The presence of a telic prefix on a transitive verb is a cue to the definiteness of the direct object. But the telic prefixes don’t express object definiteness because their use isn’t restricted to VPs with telic objects. 

In Eva’s example, if ergative case is only statistically associated with focus, then its use is a focus cue, but it isn’t a focus marker.

Scenario 2: Like Szenario 1, but now Y *always* takes on interpretation B whenever it occurs in the presence of X. In this case, we can consider the *use* of X in the context of Y an expression of B. But whether we should consider X itself an expression of B depends on X’s other uses. Obviously, if X is always associated with B in all its uses, then X is for all intents and purposes an expression of B. But if X is only associated with B when it occurs in the context of Y, then we have to consider additional evidence. 

Take Martin’s example of Spanish _a_. Its use as marking direct objects (in other words, as expressing accusative case) is restricted to animate objects. But “the same preposition” has spatial uses which do not bear this restriction. The question whether _a_ can be said to express animacy is in first approximation pragmatically the same as the question whether _a_ expresses accusative (“pragmatically the same” because it must be decided on the same grounds): do we consider the spatial (allative) and syntactic (accusative) uses of this preposition to constitute polysemy or homophony? If we decide to go with polysemy, then animacy is at most part of one polysemous sense of _a_. But if we go with homophony, then we wind up with two homophonous prepositions, one of which can be said to express accusative case (if we want to postulate nominal case marking in Spanish) and animacy. 

Wise men and women have looked at the distinction between polysemy and homophony and decided to give it wide berth :-) There are semantic diagnostics, but they’re in many instances inapplicable or yield unreliable judgments. Outside such tests, psycholinguists, descriptive linguists, historical linguists, and computational linguists all have their different grounds and motivations for preferring polysemous or homophonous analyses of a given phenomenon, which are well-motivated within their own disciplines. 

To bring matters back to the domain of focus (and Fritz’ query), consider focus particles such as _only_: should we consider _only_ an expression of focus? Obviously, the overwhelming majority of foci in English do not occur with _only_, so _only_ can’t be an obligatory expression of focus. But the question is really whether _only_ is always associated with focus across all its uses. If so, then for all intents and purposes, _only_ expresses focus (and so do all other focus particles that always co-occur with focus). I’m not sure what the correct answer is (I’ve never worked on focus particles), but my sense is that it is possible to construct perhaps slightly tortured examples in which the association is broken:

(1) [Q: At the party last night, everybody was imbibing liberally except for one person who had one beer and then left. Who was that? - A:] It was Floyd who had only one beer. 

(One might argue that _only_ is associated with a secondary focus in (1), but the important point for me is that the property of having consumed one beer is already built into (and backgrounded by) the question under discussion, so based on discourse-pragmatic notions of focus that rely on this criterion, the use of _only_ in (1) is not focus-related.)

Finally, Peter Austin brought up Michael Silverstein’s use of ‘function’. The key question is indeed whether meaning B is a function, in the algebraic sense, of the combination X … Y. This use of the term ‘function’ aligns with Montague’s, who famously proposed a formulation of the Principle of Compositionality that relies on this notion (whereas Frege’s original formulation makes use of the metaphor of semantic and syntactic composition being mirror images of one another).

That said, I don’t think it would be particularly helpful to say that 'B is a function of X’ instead of ‘X expresses/marks B’. But, I’m not sure that Peter (or Michael) meant to suggest that.

Best — Juergen

> On Aug 3, 2019, at 1:44 AM, Martin Haspelmath <haspelmath at shh.mpg.de> wrote:
> 
> Eva's post raises the interesting general question whether markers that occur under two necessary conditions should be said to mark both of these simultaneously:
> 
> – Do Quechua markers like -mi or -si "mark" only evidentiality, or do they also "mark" focus (because they always occur on the focused constituent and thus serve to identify it)?
> 
> – Do ("optional") ergative markers in Jaminjung "mark" the nominal only for ergative role, or do they also "mark" focus (because they are generally restricted to focused nominals and thus serve to identify the nominals as fucused)?
> 
> – Do case flags in "split" systems, like the preposition a+ in Spanish (which is generally restricted to animate nominals), "mark" only the syntactic role, or do they also "mark" animacy?
> 
> – Do plural markers that occur only when the nominal is definite or animate only mark "plural", or do they also "mark" definiteness or animacy? (This often happens in creole languages, see https://apics-online.info/parameters/22.chapter.html)
> 
> Linguists often say that these markers have one "function", and the other factor that plays a role in their occurrence is a "condition", but I often find these difficult to distinguish. Couldn't one say (conversely) that the Jaminjung nominal focus marker is restricted in its distribution in that it only occurs when the nominal is a (transitive) agent?
> 
> (In other words, couldn't one say that in cases of this sort, the marker has two functions simultaneously? – Sorry, this is leading away from Fritz Newmeyer's original post.)
> 
> Martin
> 
> On 02.08.19 22:10, Eva Schultze-Berndt wrote:
>> 
>> Quite apart from the problems of defining focus (on which I'm less sceptical than the sources Eitan cites), in some of the categories that have been cited the literature as focus markers the question arises whether they really *mark* focus, or are rather (i) attracted to a focused constituent, or (ii) focus plays a role in their distribution.
>> 
>> Regarding (i), I'm not a Quechuanist, but have had discussions with colleagues who are, and it is by no means clear that everyone considers evidentials in these languages as also marking focus, rather than as attaching to the focused constituent.
>> 
>> Regarding (ii), I have worked on "optional" ergativity, and would not consider ergative marking associated with focus as "marking" focus. In any of the types of "classical" split ergative system, with a split, say, between humans and non-humans, or non-past and past, we would not consider the ergative case as a marker of the categories of "non-human" or "past" – either because there is a clear segmental marking already (past), or because there is no marking of the category at all outside agents (non-human). So if ergative case only occurs on focused agents (at least probabilistically – such systems often don't seem to be entirely categorical), recognisable as focused through prosody and context, why would we consider the ergative as a focus marker?
>> 
>> Eva
>> 
>> 
>> -------------------------------------------------------
>> Eva Schultze-Berndt
>> Professor of Linguistics
>> Linguistics and English Language
>> School of Arts, Languages and Cultures
>> The University of Manchester
>> Oxford Road
>> M13 9PL
>> Manchester, UK
>> E-mail: eva.schultze-berndt at manchester.ac.uk
>> 
> 
> 
> -- 
> Martin Haspelmath (
> haspelmath at shh.mpg.de
> )
> Max Planck Institute for the Science of Human History
> Kahlaische Strasse 10	
> D-07745 Jena  
> &
> Leipzig University
> Institut fuer Anglistik 
> IPF 141199
> D-04081 Leipzig  
> 
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp

Juergen Bohnemeyer, Professor and Director of Graduate Studies 
Department of Linguistics and Center for Cognitive Science 
University at Buffalo 

Office: 642 Baldy Hall, UB North Campus * Mailing address: 609 Baldy Hall, Buffalo, NY 14260 
Phone: (716) 645 0127 
Fax: (716) 645 3825 * Email: jb77 at buffalo.edu * Web: http://www.acsu.buffalo.edu/~jb77/ 

Office hours M 12:30 – 1:30pm / W 1:00 – 1:50 / F 12:30 – 1:50pm

There’s A Crack In Everything - That’s How The Light Gets In (Leonard Cohen)