Linguistic Flashmob 21 January - "Case" with Mark baker and Julie Legate. Moderator: Joseph Bayer

 

LIST OF QUESTIONS: 1) Are direct cases a by- product of Agree or do they derive from the algorithm of dependent case? 2) How do you account for ergative vs accusative alignment? 3) Morphological case vs. Syntactic case.

Here are some discussions in the Zoom Chat during the live streaming (you can trace it in the video from the reference time indicated here):

58:05

(to the question from Omer Preminger:)

From Hagit Borer: 

a bit as a continuation to Omer's point, why is case even a thing, if it is not grammatical?  There is no particular reason to mark whatever it marks, rather than myriads of other cognitively salient distinctions both semantic and syntactic, and which are not attested.  Even if there are languages purportedly without case, there are none that mark e.g. transparent vs. opaque, or pleasant vs. unpleasant, etc.

1:03:57

From Virginia Valian: 

My question is about languages like Chinese and, I gather, other East Asian languages, that do not have any external morphology suggesting case, but have phenomena that can only be explained (“ “) if one assumes underlying invisible case.

From Waltraud Paul: 

AS a follow-up on V. Valian's question: there ARE phenomena in Chinese in general explained in terms of Case, such as the fact that a transitive verb can only be followed by its object, but not at the same time by a duration/frequency XP such as '2times', '3 hours'/ which is allowed in postverbal position in the absence of an object:

Also cf. Audrey LI's 1985 thesis and 1990 book

From Ian Roberts: 

See also Sheehan and van der Wal (2018) in Journal of Linguistics.

1:09:37

From Adam Przepiórkowski: 

How about coordinations of NPs?  Wouldn't Dependent Case predict that the conjuncts should have different cases (given that the first conjunct c-commands the second on the most popular approach to coordination)?  And, conversely, can Agree handle unlike cases of the two conjuncts in some DOM languages (as discussed in the 2019 LI squib by Kalin and Weisser)?

From Omer Preminger: 

Re:Adam Przepiórkowski’s point about coordinations, there certainly are languages in which the higher conjunct receives the case marking you’d expect “from the outside”, but the lower conjunct receives a dedicated “conjunction case”. One could think about the latter as “dependent case when the domain is &P”….

(the “conjunction case” is usually doing double-duty as another case in the language - e.g. accusative - but that could be nothing more than morphological (meta-)syncretism)

Let us know your opinion!