[Lingtyp] Innovation of functional categories

Bohnemeyer, Juergen jb77 at buffalo.edu
Wed Jun 17 16:51:35 UTC 2020


Dear Adam — Thanks for your questions! My sense is that some languages have a productive open class of adverbs and others do not. 

In the latter kind of language (e.g., Mayan languages), expressions that one would traditionally treat as adverbs because they combine directly with the verb or some verbal projection are functional expressions of varying subtypes. 

In the former - and here I would include all Indo-European languages as far as I know - the vast majority of adverbs are not functional expressions. 

The problem is that since we traditionally treat all non-phrasal adverbial modifiers as adverbs, the open adverb class will in such languages swallow up a lot of adverbial or adsentential particles that are semantically clearly functional elements - e.g., focus particles, adverbial quantifiers, and all kinds of interjections. 

So the (or one?) inherent weakness of my definition is that I would like to exclude those particles from getting sucked into the adverb category, but unfortunately I do not know how. In my thinking, this is an instance of the general “leakiness” (to borrow Sally Rice’ phrase) of grammars. 

(It seems that some Generative Grammarians would do away altogether with the adverb as a lexical category, but that strikes me too broad a brush.)

As to closed classes of adjectives, I would still want to treat those as lexical categories, so I was wrong to include _major_ (in “major lexical category”) in my definition. 

The definition of ‘lexical category’ carries a lot of weight in my definition of ‘functional expression’ - really the bulk of it. So the question is can we independently define ‘lexical category’ in order to avoid circularity?

Let me try: a ‘lexical category’ is a class of expressions defined in terms of shared morphosyntactic properties. The members of lexical categories are inherently suitable for expressing ‘at-issue’ content and are backgrounded only when they appear in certain syntactic positions (e.g., attributive as opposed to predicative adjectives). Their combinatorial types (e.g., in a Categorial Grammar framework) are relatively simple. They express concepts of basic ontological categories. 

(On this approach, lexical categories share the property of expressing primarily at-issue content with functional expressions that are not pragmatically redundant, such as modals, negation, demonstratives, etc.. They differ from those functional expressions in their combinatorial properties.)

It might be best to treat this list of properties as defining the notion of ‘lexical category’ as a cluster/radio/prototype concept.   

As to classifiers. Mayan languages have two flavors of numeral classifiers, first distinguished by Berlin (1968) in Tseltal. He calls them ‘inherent’ vs. ‘temporary’ classifiers. Inherent classifiers are redundant and thus fall into the class of functional expressions I call ‘restrictors’ - the class that also contains tense, viewpoint aspect, gender, noun class, and definiteness. Yucatec has only three of these - one for humans and animals, one for plants and hair, and one for inanimates. OTOH the so-called temporary classifiers form a very large set and are in Yucatec actually primarily used in nonverbal predication. They are non-redundant and I consider them to belong to the same class of functional expressions as negation, modals, and do on.

Best — Juergen

> On Jun 17, 2020, at 3:22 AM, Adam James Ross Tallman <ajrtallman at utexas.edu> wrote:
> 
> Dear Juergen,
> 
> Just a clarifying question (I'm interested because I've attempted to develop a method to quantify the degree to which some set of morphemes is morphologized and I have struggled with defining "functional" in a consistent fashion, and actually I have just given up)
> 
> Wouldn't your definition imply that anything that was not an open lexical class would be "functional"? 
> 
> There's plenty of languages that have a closed class of adjectives - shouldn't these be "functional" in your sense?
> 
> Maybe adjectives could be added to your class of morphemes that tend to become functional regardless of contact [?]... but just in case they are not a lexical class.  But do adjectives express redundant information or not?
> I'm also skeptical that an easy decision can be made regarding the lexical vs. functional status of classifiers, but this is perhaps outside the scope of your research question.
> (I would take a close look at Krasnoukhova's dissertation on the Noun Phrase in South American languages for both of these issues)
> 
> best,
> 
> Adam
> 
>   
> 
> On Wed, Jun 17, 2020 at 8:25 AM Östen Dahl <oesten at ling.su.se> wrote:
> Thanks, Frans, for the link to this paper, which I had not seen. (I did read Fehling’s paper, however, quite long ago.) For the record, though: although Peust claims (reasonably, it seems) that Egyptian is the ultimate source, he doesn’t say that Greek got it straight from there. Instead, he says that it is remarkable that the definite article shows up in Greek in the same time period as the Greeks took over the Phoenician script, thus suggesting Phoenician, a Semitic language, as the proximate source for the Greek definite article.
> 
>  
> 
> In light of Peust’s claims, it is maybe Egyptian that is most relevant for Jürgen’s project. Although who knows if they didn’t get the article from somebody else?
> 
>  
> 
> 	• Östen
>  
> 
>  
> 
> Från: Uni KN <frans.plank at uni-konstanz.de> 
> Skickat: den 17 juni 2020 00:04
> Till: Östen Dahl <oesten at ling.su.se>
> Kopia: LINGTYP <lingtyp at listserv.linguistlist.org>
> Ämne: Re: [Lingtyp] Innovation of functional categories
> 
>  
> 
> Close, Östen:  they got it from Egyptian.  Or so argues Carsten Peust, in Göttinger Beiträge zur Sprachwissenschaft 2, 1999, S. 99-120
> 
>  
> 
> Fälle von strukturellem Einfluss des Ägyptischen auf europäische Sprachen
> 
> (1) Die Herausbildung des definiten Artikels, (2) Die Entwicklung des grammatischen femininen Genus, (3) Die inklusive Zählweise von Zeitintervallen
> 
>  
> 
> https://archiv.ub.uni-heidelberg.de/propylaeumdok/2274/1/Peust_Faelle_von_strukturellen_Einfluessen_1999.pdf
> 
>  
> 
> Similarly
> 
> LEVIN, Saul 1992: Studies in comparative grammar: I. The definite article, an Egyptian/Semitic/Indo­European etymology, in General Linguistics 32:1­-15.
> 
> FEHLING, Detlev 1980: The origins of European syntax, in Folia Linguistica Historica 1:353-387.
> 
>  
> 
> Frans
> 
>  
> 
> 
> 
> 
> On 16. Jun 2020, at 18:25, Östen Dahl <oesten at ling.su.se> wrote:
> 
>  
> 
> This topic happened to come up in my recent conversation with Martin Haspelmath on his blog (https://dlc.hypotheses.org/2361). There are also some references there to earlier literature.
> 
> I would not bet on the definite article in Ancient Greek as an independent development. After all, definite articles were around in the neighbouring Semitic languages. If the Greeks got their alphabet from the Semitic-speaking peoples, they could also get the article from them, I think.
> 
> - Östen 
> 
> 
> -----Ursprungligt meddelande-----
> Från: Lingtyp <lingtyp-bounces at listserv.linguistlist.org> För Bohnemeyer, Juergen
> Skickat: den 16 juni 2020 15:44
> Till: LINGTYP <lingtyp at listserv.linguistlist.org>
> Ämne: Re: [Lingtyp] Innovation of functional categories
> 
> Dear Christian — Thank you very much for your response! I'll have much more to say about your suggestions, but for now, I’d just like to try a clarification:
> 
> 
> 
> On Jun 16, 2020, at 6:41 AM, Christian Lehmann <christian.lehmann at uni-erfurt.de> wrote:
> 
> 
> To the extent that the contribution made by such expressions to the sentence meaning is indeed redundant, it would mean that the respective information is already contained in the context, and to this extent there would be no need for the hearer to employ inferencing.
> 
> 
> 
> I’m assuming a view of communication on which it is largely inference-based. The question on this view is not whether but how much inferencing the hearer has to do. 
> 
> Consider the information added by gender markers to pronouns and agreement morphology. In the vast majority of cases, this information is not needed for identifying the referent. But having it by my hypothesis still facilitates processing  by further boosting the predictability of the referent. As long as the added effort for speaker and hearer in processing the gender information is minimal (that’s where grammaticalization comes in), this may confer a minuscule processing advantage. 
> 
> Same story with tense or definiteness: in the vast majority of uses, tense markers and articles are not terribly informative (witness all the speech communities that get by happily without them), so that can’t be the reason why we grammaticalize them (that’s my thinking, anyway).
> 
> (As to Givón, yes, absolutely, I’m well aware that I’m merely trying to retell a story functionalists have been telling since the dawn of functionalism :-))
> 
> Best — Juergen
> 
> --
> Juergen Bohnemeyer (He/Him)
> 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 will be held by Skype, WebEx, or phone until further notice. Email me to schedule a call at any time. I will in addition hold Tu 12:30-1:30 and Th 2:30-3:20 open specifically for remote office hours.
> 
> There’s A Crack In Everything - That’s How The Light Gets In (Leonard Cohen)  
> 
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
> 
>  
> 
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
> 
> 
> -- 
> Adam J.R. Tallman
> PhD, University of Texas at Austin
> Investigador del Museo de Etnografía y Folklore, la Paz
> ELDP -- Postdoctorante
> CNRS -- Dynamique Du Langage (UMR 5596)
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp


-- 
Juergen Bohnemeyer (He/Him)
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 will be held by Skype, WebEx, or phone until further notice. Email me to schedule a call at any time. I will in addition hold Tu 12:30-1:30 and Th 2:30-3:20 open specifically for remote office hours.

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



More information about the Lingtyp mailing list