[Lingtyp] Verbs of success with dative subject

Sat Jan 5 17:27:50 UTC 2019

I think the beauty and utility of "subject" is precisely its
looseness.  If typologists can use it in passing without making a
precise claim, we can all continue to understand the gist of what is
meant and work out the precise details later in dedicated work on that
topic.  For new, precisely defined technical concepts let's have new
terms instead of repurposed existing terms.

I use "dative subject" like that:  a dative with subject properties,
identifiable as a distinct grammatical entity before an exhaustive
list of subject properties has been drawn up and surveyed across the
relevant verbal lexicon, and before we all agree on a precise
definition of "experiencer" that will let me distinguish dative
subjects from other dative experiencers.  And meanwhile I think
everybody understands what is meant.

Johanna

On Fri, Jan 4, 2019 at 11:13 PM Martin Haspelmath <haspelmath at shh.mpg.de> wrote:
>
> I think that Ilja Seržant is right: A term like "dative experiencer" would be better for constructions like:
>
> à Sasha tout lui réussissait (French, cited by Jocelyne Fernandez-Vest)
> to Sasha all to-her succeeded
>
> Calling the dative experiencer (à Sasha) a "non-canonical (dative) subject" here (maybe on the basis of its pre-verbal position) is confusing, because there are no limits to what could be called a "non-canonical subject" – one might propose, for example, to call inanimate objects in Spanish non-canonical subjects because they are subject-like in that they lack an object-marking preposition.
>
> Spike says that the question “what is a non-canonical subject in theory?” is ultimately necessary, but I don't think so. An explanatory theory might not make use of the notion "subject" at all (and instead rely on more fine-grained parameters), let alone the notion of "non-canonical subject".
>
> What we do need is a general definition of the *term* subject", because we use it all the time anyway and it would be best if we used it uniformly. Nobody disputes that the transitive A-argument and the intransitive S-argument are subjects, so I think it's best to say that "subject" is "A or S" (as in Dixon 1994). Since A, S and P are defined in terms of their coding properties (Haspelmath 2011), this means that a dative-marked argument is never a subject (since A/S are by definition nominative/absolutive- or ergative-marked).
>
> It is interesting, of course, that some non-A/S arguments share some behavioural properties with A/S arguments, but behavioural properties are extremely diverse and are not a good basis for terminology. Pre-verbal order is extremely common when an argument is the only animate argument of a predicate, so dative experiencers will very often be subject-like in this regard – but do we want to call this a "subject property" in view of the fact that inanimate S-arguments often occur in a later position? It seems that word order is more driven by animacy and specificity than by semantic/syntactic role. It seems best to define subject/S/A by argument coding (flagging and indexing), not by behavioural properties.
>
> Martin
>
> On 04.01.19 17:22, Spike Gildea wrote:
>
> First, I thank everyone for sharing  examples of dative subjects with predicates of success. Alongside the expected examples in Indo-European languages of the Slavic, Romance, Germanic, and Indic families, examples were proposed from Causasian languages in general (with Akhvakh as an example), North Saami and Finnish (Uralic), Hebrew (Semitic), and Japhug (Tibeto-Burman) — while there are at least examples outside of IE, this is not a particularly robust cross-linguistic attestation of the phenomenon. I originally posted the query because I am aware of no examples in the non-canonical case-marking languages of South America, and it is interesting that nobody has mentioned examples from the language families of North America or Austronesia
> that are known for semantic alignment.
>
> Second, with regard to Ilja’s query, there is a long tradition of disputing the use of the term “subject” for apparent primary arguments that do not bear the canonical case-marking of subjects in a given language, in particular for analyses of "dative subjects". Much of Jóhanna’s own work (particularly Eythorsson & Barðdal 2005, Barðdal & Eyth̩órsson 2012) participates in this dispute, in that she has consistently used a range of syntactic tests to distinguish dative subjects from non-subject dative experiencers, such as order, raising, reflexivization (both long-distance and clause-bound), control infinitives, and conjunction reduction. The disputes arise from the fact that these syntactic tests do not give consistent results, even in closely related Germanic languages like Icelandic, where all such tests show that the only distinction between nominative subjects and non-canonical subjects is case-marking and verb agreement, and German (which is more akin to the range of other European languages), where only a subset of the tests syntactically align potential dative subjects with nominative subjects. It is true that different theoretical perspectives interpret this phenomenon differently, and in particular, some prefer to privilege the term “subject” as a theoretical label that should not be assigned on the basis of some (non-specific) subset of “subject tests”.
>
> In this query, I was hoping to finesse the (ultimately necessary) question of “what is a non-canonical subject in theory?” and its operational correlate “which criteria should count most in identifying them?” That is, I hoped just to use the term “dative subject” as a shorthand by which colleagues might recognize constructions in individual languages that show a combination of properties that would then constitute potentially interesting cases for follow-up. I could re-formulate the query in more precise terms as follows: we are looking for indications of languages for which (i) predicates of success mark the “succeeder” as a dative (or other non-canonical case that could be used to mark recipients or benefactives), and (ii) the syntactic properties associated with this dative “succeeder” are distinct from clear “indirect object” dative arguments in that they share one or more syntactic properties with canonical subjects.
>
> Best,
> Spike
>
> References
> Barðdal, Jóhanna & Thórhallur Eythórsson. 2012. ‘Hungering and lusting for women and fleshly delicacies’: Reconstructing grammatical relations for Proto-Germanic. Transactions of the Philological Society 110(3): 363–393.
> Eythórsson, Thórhallur & Jóhanna Barðdal. 2005. Oblique Subjects: A Common Germanic Inheritance. Language 81(4): 824–881.
>
>
> On Jan 3, 2019, at 11:34 PM, Ilja Seržant <ilja.serzants at uni-leipzig.de> wrote:
>
> Dear all,
>
> I apologize for a side remark. But do we call any kind of argumental and non-argumental animate (experiencer) dative NP a non-canonical subject? :-) Does it really make sense to use the notion of subject that way? Woudn't be a term like "dative experiencer" or "dative/recipient-like experiencer" be more adequate for a cross-linguistic comparison?
>
> Best,
>
> Ilja
>
> Am 21.12.2018 um 17:00 schrieb Spike Gildea:
>
> Dear colleagues,
>
> I forward a query from my colleague, Jóhanna Barðdal, who is looking for examples of predicates of "success” with non canonical subject marking, in particular those that take a dative subject.
>
> We are working on Indo-European verbs/predicates with the meanings 'succeed', 'be successful', 'make progress', 'turn out well', 'go well'. The last one in the sense "he is doing well in his dance class" or even "he is doing well in life”.
>
>
> Thank you in advance for indications of other places in the world where we might find such predicates taking a dative subject!
>
> Best,
> Spike
>
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
>
> --
> Ilja A. Seržant, postdoc
> Project "Grammatical Universals"
> Universität Leipzig (IPF 141199)
> Nikolaistraße 6-10
> 04109 Leipzig
>
> URL: http://home.uni-leipzig.de/serzant/
>
> Tel.: + 49 341 97 37713
> Room 5.22
>
>
>
>
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
>
>
> --
> 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