[Lingtyp] clause linker stacking/co-occurrence

Sun Feb 2 10:44:06 UTC 2025

Dear All,
thanks to Daniel Ross for this nice overview (I suppose, it‘s not exhaustive). When I first read the message by Joseph Brooks which was focused on switch reference (and languages far away from Europe), I was reluctant to send a reply, because the data which came to my mind is not related to switch reference, but to reported speech. But now Daniel pointed out more types of “stacking” of clausal connectives among which some are not alien to at least some European languages, namely to languages in Europe’s east.
               I have in mind cases in which a connective which is generally regarded as “declarative / default / standard / epistemically neutral complementizer” (like Engl. that) combines with a morpheme which marks directive or optative speech acts. This occurs in all Slavic languages (and probably in other languages of Europe’s eastern half). Compare the following examples (quoted ad libitum from corpora) in which the connectives are highlighted in bold; the first is the “complementizer”, the second the directive-optative marker (it almost always occurs with verbs in the indicative non-past):

Polish
(11a)     Stary odpowiedział, że niech nawet w więzieniu zgnije.
               ‘The old man replied that may he even rot in prison.’
               (PNC)
(11b)     Gdy złożyłam papiery do szkoły artystycznej, mama powiedziała jedynie, że niech się dzieje wola nieba i najlepiej poczekać na rezultaty.
‘When I submitted my papers to the art school, my mother only said that may heaven's will be done and it would be best to wait for the results.’
(PNC; Gazeta Poznańska, 2005)

Russian
(12a)     On govoril ej, čtoby ne vydumyvala, čto pust’ pol’zuetsja vsem, čem nado.
               ‘He told her not to invent, that she should use everything needed.’ (lit. ...that may she use...)
               (RNC)
(12b)     Babuška, dumaja, čto pust’ lučše lobotrjasy prixodjat k nim, čem Koka uxodit, každyj raz                                                       servirovala im stol s nenužnoj roskoš’ju.
               ‘Grandmother, thinking that it would be better for loafers to come to them than for Koka to leave,                         each time she served them a table with unnecessary luxury.’
(lit. ‘… thinking that may loafers come to them…’)
(RNC; M. Gigolašvili: Čertovo koleso. 2007)

Slovene
(13a)     Rekel mi je, da naj vas pričakam.
               ‘He told me to pick you up.’ (lit. …that may I…)
(http://opus.nlpl.eu)
(13b)     V enem dopolnilu denimo predlagajo, da naj reden dodiplomski študij ostane brezplačen.
               ‘In one supplement, for example, they propose that full-time undergraduate studies should remain                              free.’ (lit. ‘… that may … remain free.’)
(Gigafida; 2001)

The status of either element in such combinations is debatable. In particular, the second element has been variably treated as “mood marker” (auxiliary of 3rd person imperatives), as “particle” (just marking directive or optative illocutions), or it can even be treated as an emergent complementizer (whose illocutionary force is in concord with the lexical meaning of some possible complement-taking element in the preceding context, e.g. a verb denoting a directive speech act – see the examples above). Correspondingly, combinations like those above can be treated in different ways: (i) as complementizer + mood auxiliary, (ii) as double complementizer, (iii) as quotative marker (THAT) + mood / illocutionary marker. Importantly, the directive marker (= second element) also occurs in independent clauses and just marks directive/optative speech acts, and examples as those above are typical in that they show reports of directive (or optative) speech acts.
               The entire problem has been treated in a couple of papers:
https://doi.org/10.1515/slaw-2023-0012
https://doi.org/10.5817/LB2023-1-1
https://journals.ispan.edu.pl/index.php/sm/article/view/sm.3194

These cases can be subsumed under those which Daniel treats under 3(a) below, and I wonder whether they would count as “clause linker stacking”. Or should this expression be limited to cases with switch reference?
               On this occasion, I’d be happy to learn whether such phenomena like those above (alleged COMP + DIR/OPT-marker, with debatable syntactic status) occur in other regions and language groups of the world. I’d thus appreciate any information on that.

Best,
Björn Wiemer.

From: Lingtyp <lingtyp-bounces at listserv.linguistlist.org> On Behalf Of Daniel Ross via Lingtyp
Sent: Sunday, February 2, 2025 7:40 AM
To: Joseph Brooks <brooks.josephd at gmail.com>; Linguistic Typology <lingtyp at listserv.linguistlist.org>
Subject: Re: [Lingtyp] clause linker stacking/co-occurrence

Dear Joseph,

One term for (at least some of) what you have described would be para-hypotaxis, introduced to typology in this paper (following traditional usage for studies as especially a stylistic device in historical usage of European languages):
Bertinetto, Pier Marco & Luca Ciucci. 2012. Parataxis, Hypotaxis and Para-Hypotaxis in the Zamucoan Languages. Linguistic Discovery 10(1). 89–111. https://doi.org/10.1349/PS1.1537-0852.A.404

Also surveyed cross-linguistically in this presentation:
Ross, Daniel, Jesús Olguín Martínez & Luca Ciucci. 2018. Para-hypotaxis in the world’s languages: A cross-linguistic survey. Presented at Syntax of the World’s Languages (SWL) 8, INALCO, Paris, September 4, 2018. https://swl8.sciencesconf.org/data/pages/Ross_etal_PH_SWL_1.pdf

Note that Amele switch-reference plus conjunctions is mentioned on slide 25.

The typological survey is also discussed in chapter 3 of my dissertation:
Ross, Daniel. 2021. Pseudocoordination, Serial Verb Constructions and Multi-Verb Predicates: The relationship between form and structure. University of Illinois Urbana-Champaign Ph.D. dissertation. https://doi.org/10.5281/zenodo.5546425

Specifically, para-hypotaxis is when parataxis/coordination and hypotaxis/subordination are overtly mixed in a single clause-combining strategy. By this definition, hypotaxis could include switch-reference marking plus a coordinator for parataxis blended in the same construction.

Note that other possible combinations of conjunctions or other markers of clause-combining are not strictly included in the definition of para-hypotaxis, including:
1. Correlatives in general if we they aren't analyzed as "parataxis"+"hypotaxis".
2. Doubling of coordinators, e.g. "and so", "and yet" in English, or wa lakin 'but' (lit. 'and but') in Arabic, or similarly conjunction pairs like "not only..... but...".
3. Doubling of subordinators, e.g. preposition-related subordinators plus que 'that' in Spanish such as para que 'in order to' (lit. 'for that'). [Another analysis might be preferable here, but I just mentioned this for illustrative purposes. There are probably better examples.] Even more broadly, we could consider for example subordinate moods (e.g. subjunctives) plus complementizers to be some kind of "doubling", or is that some kind of agreement, or something else?
3b. There are also instances of subordinators combined with non-finite verbs (I think some of the examples summarized in Hannah's reply are probably of this type in the recent clause-chaining volume).
3c. For example doubly-suffixed non-finite verb forms such as a nominalized verb form further suffixed with a case-like suffix for adverbial functions.

On the diachrony of converbs related to (3c), and another mention of para-hypotaxis, see also:
Ross, Daniel. 2025. The life cycle of converbs: A diachronic typology. In Paola Cotticelli-Kurras, Eystein Dahl & Jelena Živojinović (eds.), Diachronic, Typological, and Areal Aspects of Converbs, 317–360. Berlin: De Gruyter Mouton. https://doi.org/10.1515/9783111335551-012

There are definitional questions that come up when we try to categorize doubling (stacking, co-occurrence, etc.), which I think often suggest a diachronic perspective for analysis. For example, para-hypotaxis can be introduced via language contact (e.g. in a language with clause-chaining plus borrowed conjunctions). Otherwise, whether it counts as "doubling" might depend on whether we consider form (two morphemes) or function (one clause-combining function, or more?).

Daniel Ross
UC Riverside

On Fri, Jan 31, 2025 at 2:22 AM Joseph Brooks via Lingtyp <lingtyp at listserv.linguistlist.org<mailto:lingtyp at listserv.linguistlist.org>> wrote:
Hi,

I'm wondering if anyone could point me to any sources or have that report on clause linker stacking/co-occurrence of clause linkers in a single clause. Esp looking for examples from languages with clause chaining and/or switch reference where linker stacking occurs in one or more constructions. (Unpublished examples also welcome).

Roberts' 1997 (137, 194) work on switch reference in languages of New Guinea claims that for languages with clause chaining, co-occurrence of markers is found exclusively in New Guinea, and even then only in a handful of languages. Wade 1997 is the only paper I know of that focuses on this topic explicitly, for the Apalɨ language.

Thanks,
Joseph
_______________________________________________
Lingtyp mailing list
Lingtyp at listserv.linguistlist.org<mailto:Lingtyp at listserv.linguistlist.org>
https://listserv.linguistlist.org/cgi-bin/mailman/listinfo/lingtyp
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20250202/1a3603d5/attachment.htm>