The Weekly Newsletter of MIT Linguistics

Issue of Monday, February 29th, 2016

Never tell me never

Very good news, not about linguistics but about life, from our colleague Jay.


MIT papers headed for GLOW

The GLOW Program is out for this year’s meeting in Göttingen, and several of our students will be presenting their discoveries there! “GLOW” stands for “Generative Linguistics in the Old World” and is the premiere international conference for generative linguistics in Europe (and, some might argue, internationally).


ESSL/LacqLab 2/29 - Zuzanna Fuchs

Speaker: Zuzanna Fuchs (Harvard)
Joint work with Gregory Scontras (Stanford) and Maria Polinsky (UMaryland)
Date and time: Monday, February 29, from 1:00 to 2:00 PM
Place: 32-D831

At some point in their childhood, heritage speakers shift from their first acquired language to their second language — the language of their community — that becomes their new dominant language. For heritage speaker, language comprehension and production in the non-dominant L1 is therefore difficult and costly, and thus the grammar of this L1 may be somewhat different than the baseline native grammar, shaped by principles of economy that make comprehension and production tasks easier. In this talk, we consider number and gender agreement in Heritage Spanish, in order to determine whether difficulties with agreement morphology observed in heritage speakers (Benmamoun et al. 2013, and references therein) reflect any underlying differences in how number and gender are represented in the grammar of Heritage Spanish. We put number and gender features into conflict with each other through agreement attraction, and observe how errors in agreement are perceived in a language comprehension task, replicating the methodology from Fuchs et al. (2015). The results indicate that number and gender in Heritage Spanish are bundled, unlike in the native grammar in which they are split. Thus, we identify an instance of divergence in grammar, which provides evidence that not all surface differences between native and heritage grammars can be ascribed to principles of economy of processing — representational economy also plays a role in shaping heritage grammars.

Phonology Circle 2/29 - Sam Zukoff

Speaker: Sam Zukoff (MIT)
Title: The Mirror Alignment Principle: Morpheme Ordering at the Morphosyntax-Phonology Interface (Part II: Arabic)
Date: Monday, February 29th
Time: 5-6:30
Place: 32D-831

The topic of Semitic nonconcatenative morphology is a vexed question in linguistic theory. Unlike most other languages, morphological derivation of complex forms in Semitic does not straightforwardly consist of sequential affixation to a fixed base of derivation. Individual morphemes can be segmented and identified with varying degrees of clarity and ease, but they are often interspersed within other morphemes, and their addition often significantly alters the segmental order and/or larger prosodic organization relative to the corresponding less derived morphological form. In this talk, I argue that the Mirror Alignment Principle approach to morpheme ordering (introduced last week at Ling Lunch; Zukoff 2016) provides the tools for deciphering this system, both from the phonological perspective and the morphosyntactic perspective.

The Mirror Alignment Principle (MAP) is an algorithm which maps c-command relations in the hierarchical morphosyntactic structure into ranking relations among Alignment constraints (McCarthy & Prince 1993) in the phonological component. By implement morpheme ordering in the phonological component using gradient, violable Alignment constraints, ordering preferences can interact with phonological constraints. The interaction between Alignment constraints, syllable well-formedness constraints, and a few morpheme-specific phonotactic constraints, will allow an analysis of the phonological aspects of the nonconcatenative system without any appeal to prosodic templates (McCarthy 1979, 1981).

Since the MAP directly relates the ranking of Alignment constraints to hierarchical morphosyntactic structure, the rankings determined through this phonological analysis inform morphosyntactic structure; the Map thus allows syntactic structure to be reverse engineered from the phonology. In considering the syntax deduced by this reasoning, we will observe larger regularities within the system. Based on these generalizations, I will suggest that certain apparent surface distinctions can be collapsed, such that the overall morphosyntactic verbal system looks generally unremarkable from a typological perspective. In fact, upon careful inspection, it even illustrates Mirror Principle effects (Baker 1985), such as mirror-image ordering that correlates with reversal in semantic interpretation. This approach thus shows that nonconcatenative morphological processes are fully compatible with the Mirror Principle, a result which Baker’s (1985) original proposal was unable to achieve.


LFRG 3/2 - Aurore Gonzalez & Sophie Moracchini

Speaker: Aurore Gonzalez (Harvard) & Sophie Moracchini (MIT)
Time: Wednesday, March 2, 1-2pm
Place: 32-D831
Title: Discussion of Guerzoni (2003)’s dissertation: Even across languages and the scope theory.

There’s been a slight change of plans for LFRG tomorrow. Sophie and Aurore will be presenting a chapter of Elena Guerzoni’s (2003) dissertation.


Ling Lunch 3/3 - Ray Jackendoff and Jenny Audring

Speaker(s): Ray Jackendoff (Tufts University) and Jenny Audring (Leiden University)
Title: Morphology in the Mental Lexicon
Date/Time: Thursday, March 3, 12:30pm-1:50pm
Location: 32-D461

We explore a theory of morphology grounded in the outlook of the Parallel Architecture (PA, Jackendoff 2002), drawing in large part on Construction Morphology (Booij 2010). The fundamental goal is to describe what a speaker stores and in what form, and to describe how this knowledge is put to use in constructing novel utterances. A basic tenet of PA is that linguistic structure is built out of independent phonological, syntactic, and semantic/conceptual structures, plus explicit interfaces that relate the three structures, often in many-to-many fashion.

Within this outlook, morphology emerges as the grammar of word-sized pieces of structure and their constituents, comprising morphosyntax and its interfaces to word phonology, lexical semantics, and phrasal syntax. Canonical morphology features a straightforward mapping among these components; irregular morphology is predominantly a matter of noncanonical mapping between constituents of morphosyntax and phonology.

As in Construction Grammar, PA encodes rules of grammar as schemas: pieces of linguistic structure that contain variables, but which are otherwise in the same format as words – in other words, the grammar is part of the lexicon. Novel utterances are constructed by instantiating variables in schemas through Unification. A compatible morphological theory must likewise state morphological patterns in terms of declarative schemas rather than procedural or realizational rules.

Non-productive morphological patterns can be described in terms of schemas that are formally parallel to those for productive patterns. They do not encode affordances for building new structures online; rather, they motivate relations among items stored in the lexicon. Productive schemas too can be used in this way, in addition to their standard use in building novel structures; hence they can be thought of as schemas that have “gone viral.” We conclude that morphological theory should be concerned with relations among lexical items, from productive to marginal, at least as much as on the online construction of novel forms.

This raises the question of how lexical relations are to be expressed. Beginning with the well-known mechanism of inheritance, we show that inheritance should be cashed out, not in terms of minimizing the number of symbols in the lexicon, but in terms of increased redundancy (or lower entropy). We propose a generalization of inheritance to include lexical relations that are nondirectional and symmetrical, and we develop a notation that pinpoints the regions of commonality between pairs of words, between words and schemas, and between pairs of schemas.

The upshot is a richly textured lexicon, one that invites comparison with other domains of human knowledge.


Colloquium 3/4 - Veneeta Dayal

Speaker: Veneeta Dayal (Rutgers)
Title: List Answers through Higher Order Questions
Date: Friday, March 4th
Time: 3:30-5:00 PM
Place: 32-141

Higher order questions have been invoked in the context of local as well as long-distance list answers. List answers to multiple wh questions and questions with quantifiers display functionality: domain cover and one-one/many-one pairings. One proposal for capturing this functionality involves iterating the question forming operation by projecting a double C structure. This results in a family of questions. Distributing an answerhood operator over the members of this set and intersecting the result yields the right type of list (Hagstrom 1998, Fox 2012). This talk explores the possibility of extending the proposal to scope marking constructions. Two problems are noted, one having to do with the absence of the truth requirement in these constructions, the other with restrictions on the type of predicate that can participate in scope marking. Higher order questions have also been used to explain list answers across wh islands (Dayal 1996). An embedded multiple wh question, interpreted as a family of questions, enters into a functional dependency with a wh expression in the matrix clause, resulting in a long-distance pair-list answer. This account is challenged by the phenomenon of trapped pair-lists (Ratiu 2005 and Cheng and Demirdache 2010), which seems to disallow a matrix wh from pairing up with one embedded wh to the exclusion of its clause-mate wh. Looking at this range of facts suggests that higher order questions do play a role in list answers, but not for all list answers.

Michel DeGraff’s letter to the New Yorker

Is Haitian a “French patois”?

Read Michel DeGraff’s answer to the New Yorker (Haitian version, original full-length version in English).