Courses

This course is a continuation of LING 311 ("Syntax I"). We will examine various syntactic patterns found in natural language (some found in English, some also found in other languages, and some only found in other languages), and develop a model of the speaker's grammatical knowledge to account for these patterns.

As in LING 311, the goal of learning the details of a particular theory of syntax is important, but secondary; the primary goal is to master the fundamentals of how to theorize in syntax: how to formulate precise syntactic hypotheses; how to use those hypothesis to guide your search for interesting data; how to separate relevant data from "noise" that your data may contain; and how to develop (and critique) syntactic argumentation.

Therefore, you will find that it is often less important to find the "right" answer (on occasion, there might not even be a "right" answer yet), and more important to explain, carefully and in detail, the argument or arguments that led you to the answer you provided.

A provisional list of topics:

• Review and (re-)orientation
  • A "refresher course" on the fundamentals of theoretical linguistics
  • Syntax and its place in the mental grammar
  • Binding Theory: review of the three Binding Conditions (from LING 311), plus introduction of "Condition Q"
  • Raising & Control (four constructions + three and a half diagnostics)
• Movement
  • Basics
    • Motivating its existence
    • Properties of wh-movement (landing sites, do-support/subj-aux inversion)
  • The "zoo" of syntactic islands
  • Using islands as a "movement detector"
  • The A-vs.-A' distinction
    • Three diagnostics
    • Interactions between movement & binding
  • Subjacency
• Ellipsis
  • Basics
  • The Identity Condition
  • Antecedent-Contained Deletion and the argument for covert Quantifier Raising

This course is a continuation of LING 610: Syntactic Theory. The course will have two main (and interrelated) prongs. First, we will examine the analytical approaches, techniques, and new emphases that emerged in the wake of Chomsky's "Minimalist Program" (Chomsky 1993, 1995). Second, we will examine how crosslinguistic variation simultaneously informs and restricts the theory of syntax, and to what extent (if at all) there is a tension between such variation and the aims of minimalism.

As we survey these more recent proposals, we will be revisiting some of their predecessors (broadly speaking, those that fall within the Government & Binding (GB) framework), and consider what gains have been made.

A provisional list of topics:

• the transition from GB to minimalism
• the Copy(/Multidominance) Theory of Movement
• probe-goal
• phases
• unaccusativity
• the Split-VP Hypothesis (="little v")
• the nature of ergativity
• head movement
• case, agreement, and the Person Case Constraint

What are the very smallest units of syntactic composition, and what sorts of constraints do they put on semantics and phonology? In this seminar, we will pursue these questions in several directions, including their implications for psycho- and neurolinguistics.

This course is an introduction to the foundational findings and concepts of the generative approach to the syntax of natural languages.

Some of the issues we will touch on include:

• the hierarchical organization of expressions in language
• the syntax of disjoint reference and covariation (a.k.a. Binding Theory)
• movement (i.e., the displacement of syntactic units) and its subtypes

For each of these, we will first present a sampling of the kind of data that (hopefully) convinces us that there is a phenomenon to be explained, in the first place. Then, we will present (and critically evaluate) theoretical accounts of these phenomena. In doing so, we will keep our sights set on the following overarching factors:

• cross-linguistic differences, and cross-linguistic invariants (a.k.a. "universals"), and how to account for each
• the relationship between the linguistic system that humans end up acquiring and the data/input available to them in the course of this acquisition

Traditional approaches to grammar appeal to a "lexicon" as the list of all things idiosyncratic / all things that cannot be computed automatically / all things that need to be learned on an item-by-item basis.

But what does this "lexicon" contain? Words? Morphemes?

What about idioms? Where are they listed?

And what about complex words whose form & meaning is fully predictable from their parts? Do they still need to be listed?

In this course, we will carefully investigate what does and doesn't need to be listed in a speaker's mental grammar. We will see that the traditional "lexicon" is inadequate in a myriad of ways. It lists both too much and too little.

Our careful investigation will lead to some surprising conclusions: the fundamental building-blocks of syntactic structure have, rather inescapably, neither form nor meaning. Furthermore, structured linguistic utterances ("sentences") contain no Saussurean signs, i.e., no individual mappings from form to meaning, whatsoever (!). Finally, we will explore the consequences of these findings for the theory of formal features and where they come from.

A graduate-level introduction to syntactic theory.

For several years, Howard has taught this course on his own. This year, however, we have decided to do a little experiment and co-teach the class. Since this is an experiment/work-in-progress, we're not quite sure how exactly the course will shape up. But we're betting it'll be interesting!

This is a graduate syntax seminar on the topic of Agreement, with an emphasis on generative work from the past two decades or so.Possible topics include:

• Is there a coherent pre-theoretical definition of "agreement"?
• The (ignominious) history of pre-minimalist treatments of agreement in generative linguistics
• The nature & source of the obligatoriness of agreement
• The interaction of agreement & case
• The Person Case Constraint
• Agreement and coordination
• The modular locus of agreement (PF vs. syntax. vs. both)
• Other things that may or may not be "agreement"
• The "great" directionality wars

My part in this course concerns the Definiteness Effect (Milsark 1974, 1977, Barwise & Cooper 1981, i.a.), and why – contrary to frequent claims in the literature – attempts to account for the phenomenon semantically are fundamentally hopeless. Here is a handout.

This course will focus on the syntax of Person and Number, and will also be of immediate relevance to those interested in the syntax-morphology and syntax-semantics mappings.

We will explore several case studies concerning the syntax of Person and Number and how it relates to the way these substantive categories are treated in other modules of the grammar.

In some of the cases we examine, the mapping between syntax and other modules will turn out to be more transparent than is usually assumed. In particular, we will see evidence that there is generally no such thing as phonologically null agreement (i.e., valuation of Person/Number features in syntax which then receives no morphological exponence). This result contradicts frequent claims in the minimalist canon (cf. Chomsky 2000, 2001, i.m.a.), while at the same time significantly tightening the mapping between syntax and morphology.

In other domains, however, we will see that the mapping is much less transparent than is usually assumed. In particular, we will see that the "marked" member of Person and Number oppositions need not be the same in morphology or in semantics as it is in syntax. This, in turn, suggests that it is time to abandon the (outdated) notion that there are grammatical primitives like "plural" or "singular" – or, for that matter, their counterparts in the domain of Person – that cross-cut different grammatical modules. This should be welcome news, as it brings Person and Number back in line with most other grammatical primitives, where no such cross-modularly consistent primitives exist, either. (Compare: "verb" vs. "open-class predicate of events" vs. "potential bearer of inflectional morphology.")

Utterances in natural language consist of many kinds of units. Some units are very big, like clauses. Some are slightly smaller, like phrases. Some are very small indeed, like morphemes, or even phonological segments. In between morphemes and phrases, traditional grammarians assume a mid-sized unit called the "word." But what is this unit? If you grew up speaking – and, more importantly, reading & writing – a language like English, you might think it's obvious that there are "words." After all, how else would we know where to put the spaces?! But is there anything else that defines a word as a linguistic unit (and setting aside phonological wordhood, which is a decidedly different notion), besides "the thing that goes between two spaces (in some languages)"? Think about a 4-year old child who does not yet read or write, but speaks their native language fairly well. Does this child know "words"? If so, then the notion of word must be definable in terms that do not refer to reading or writing. What defines it, then?

In this course, we will investigate this question in detail. We will ask: what are the rules for combining morphemes together, and do they differ from the rules for putting together larger units (e.g. phrases)? In more concrete terms: does the relation between dog and s in "dogs" differ from the relation between a and dog in "a dog"? And if so, how? Are there special phonological, syntactic, and/or semantic rules for one kind of combination that don't apply to the other?

We will investigate these questions along a couple of dimensions. One is cross-linguistic: we will compare English with other languages, both related and unrelated. Another is diachronic: we will consider evidence from older stages of the English language, and see how the differences and changes bear on the questions we are interested in.

The website for the course can be accessed through ELMS.

The idea of this seminar is to take a single linguistic phenomenon (or a well-defined set of them), and explore different ways of looking at it. How does a theoretical syntactician think about it? A computational linguist? A psycho-/neuro-linguist? Students of all persuasions will get familiar with how one approaches research questions from all three perspectives, an endeavor which – before long – will have us rooting around in foundational assumptions and concepts that each perspective takes for granted. By the time we're done, you will know how to think about scientific questions about language like a syntactician, a computational linguist, an experimentalist. Gather up your machetes: we plan to cut ruthlessly through jargon and other barriers that divide us.

This course is a graduate-level introduction to morphology. We will ponder the question of what morphology is: Is it merely an 'interface', the point where syntax ends and phonology begins? Or is it a computational module unto itself, with its own proprietary grammatical primitives and computations? And if it is the latter, what distinguishes morphological computations from syntactic and/or phonological ones?

Next, we will critically examine the notions of 'word' (in anything but the phonological sense), and of a unitary 'lexicon' – and explore the reasons why many morphologists practicing today eschew both of these notions.

We will then acquaint ourselves with two major approaches to morphology in contemporary generative linguistics: Distributed Morphology (Halle & Marantz 1993, Noyer 1992, 1997, Embick 1997, i.a.) and Spanning (Starke 2009, Svenonius 2012, 2016, i.a.). We will examine what architectural assumptions each framework rests on, how they differ from one another, what each approach does well (and what it does poorly).

Finally, I hope to take a step back and examine the consequences of these (very powerful) morphological frameworks for the mapping from syntactic structure to phoneme strings, with an eye towards the fact that the looser this mapping, the harder the job is for both the linguist and the language learner.

Course materials and information will be distributed by email and/or via the "LING625" folder in the PDF locker.

In this course, we will explore the two core empirical domains that Chomsky's (2000, 2001) Agree system was originally intended to account for: (i) syntactic relations involving valuation of PERSON, NUMBER, and/or GENDER/NOUN-CLASS features; and (ii) the assignment of (abstract) case to nominals. We will see a variety of arguments that the Agree system, as formulated, is not an adequate model for either of these two domains.

Next, we will explore what modifications are necessary in order to turn this system into an empirically adequate one. We will see that the crucial change required is that Agree (or whatever operation we replace it with) must be allowed to fail, and that this failure cannot be cause for ungrammaticality (or a "crash").

We will then explore several important consequences and extensions of these results. The first is the outright falsification of Chomsky's (2001) "Strong Minimalist Thesis." The second is the existence of privative features values in syntax. Specifically, the fact that traditional categories like '3rd person', 'singular', and 'nominative' are not feature values unto themselves, but represent the outright absence of feature values of the relevant kind. Finally, we will see how this tolerated-failure logic extends to empirical domains beyond case and agreement.

The arguments presented will be based on primary data from Kaqchikel (Mayan); Zulu (Bantu); Sakha (Turkic); Icelandic (North Germanic); Tsez (NE Caucasian); and Basque.

How does one discover the properties of a language about which very little is known beforehand? This course is a practical introduction to the methods used to accomplish this task. A native speaker of a language that is relatively unknown (to the students) will be working with us throughout the semester; this semester, we will be working with a speaker of Georgian (Kartvelian). After a few introductory sessions, and every week thereafter, a different student will be in charge of running an in-class elicitation session with the speaker, supervised by the course instructors, to uncover facts about the grammar of this language.

The website for the course can be accessed here. (Note: access is currently restricted to course participants only.)

This two day seminar will be a fast-paced introduction to agreement in generative grammar: key data, theoretical foundations, and current issues.

You can find the handouts for the course here.

Work on syntactic movement has recognized at least two major movement types: phrasal movement (movement of an entire syntactic phrase) and head movement (movement of a single syntactic term, stranding the rest of its phrase). The two differ in a number of important respects:

1. head movement forms atomic syntactic units, phrasal movement does not
2. head movement is more local than phrasal movement
3. head movement feeds affixation; phrasal movement does not
4. head movement seems to have no semantic effects, but phrasal movement does

In this course, we will investigate the syntax of head movement, arguably the less well-understood of the two movement types. We will see that the distinction between head movement and phrasal movement is more muddled than (1-4) would suggest. Long Head Movement (e.g. Borsley, Rivero & Stephens 1996, Embick & Izvorski 1995, Harizanov 2016, Lema & Rivero 1990) calls (2) into question. The existence of phrasal affixes (e.g. the Saxon genitive marker 's; Giorgi & Longobardi 1991) presents a challenge to (3). Finally, a growing body of research argues that contra (4), head movement does have semantic consequences (e.g. Hartman 2011, Lechner 2006).

More generally, it remains an outstanding question why the grammar should permit two distinct types of movement in the first place. This concern, among other considerations, has led to proposals that exclude head movement from the narrow syntax, and relegate it to PF (e.g. Adger, Harbour & Watkins 2011, Chomsky 2000, Harley 2004), re‑analyzing it as remnant movement (e.g. Koopman & Szabolcsi 2000, Mahajan 2000), or treating it as the interleaving of syntactic phrasal movement and morphological constituent-formation (e.g. Matushansky 2006).

These issues are only exacerbated in light of the existence of cross-categorial head movement processes – namely, noun incorporation (e.g. Baker 1988 et seq.), and clitic doubling (Anagnostopoulou 2003, 2006, and references therein). These phenomena are often ignored in proposals concerning the correct treatment of head movement proper, and a theory of noun incorporation and/or clitic doubling is left for a time when canonical head movement has been fully understood. In this course, we will try turning this logic on its head: we will employ insights from noun incorporation and clitic doubling to better inform the treatment of even the more canonical instances of head movement.

This course is an in-depth investigation of the structure of languages of the Mayan family. The family comprises over forty languages, and the goal is to build the foundation for independent analytical work on these languages; such future work could include an investigation of a particular language, a micro-comparative study within Mayan, a comparison of Mayan with unrelated but structurally similar languages, a processing study, a computational modeling study, or work on acquisition of Mayan languages.

We will address the basics of language documentation and work on understudied languages, and then cover a range of topics in the phonology, morphology, and syntax of Mayan languages, as well as a sampling of experimental and acquisition work on these languages.

The course is a good "springboard" for students wishing to participate in a trip to the new field station in Guatemala. The first such trip will take place quite soon after the semester's end, in May/June 2016.

The website for the course can be accessed here. (Note: access is currently restricted to course participants only; if you would like access, please get in touch with me by email.)

This course is an introduction to linguistics and linguistic theory. The central theme of the course is linguistic diversity; the challenge it poses for a rational, universalist view of linguistic competence; and how that challenge is met. We will consider this issue from the perspectives of phonology, morphology, and syntax. The students will attain basic familiarity with these linguistic sub-fields, and with issues of theory building and theory development more generally.

The website for the course can be accessed through Blackboard.

This course explores advanced topics in contemporary syntactic theory. Topics include: the A- vs. A‑bar distinction in syntactic movement, and the interaction of syntactic movement with binding; syntactic islands and the locality of movement; nominal licensing and case; and argument structure and valence-changing verb alternations. One of the primary goals of this class is to have the students practice looking at linguistic data from a theoretically informed perspective – i.e., using syntactic hypotheses to guide their search for interesting data, and separating relevant data and generalizations from "noise" that the data may contain – as well as honing their skills at developing (and critiquing) syntactic argumentation.

The website for the course can be accessed through Blackboard.

This course is about the words of the English language: their sound patterns, their structure, their meanings, and their history. These issues are approached from a formal linguistic perspective.

Throughout the course, the study of words serves as both a means and an end. On the one hand, words serve as a prism through which the students will encounter and familiarize themselves with the fundamental concepts and methodologies used in modern linguistics. On the other hand, by applying these concepts and skills to the words of English, the students will gain a much deeper understanding of the words that make up the vocabulary of English, and of what exactly it means to know the words of a language, in the first place.

No previous knowledge of linguistics is required or assumed; as such, the course can serve as an introduction to modern linguistics for students that may or may not take another linguistics course in the future. The students will be introduced, as needed, to tools from different sub-disciplines of modern linguistics: phonology, morphology, syntax, semantics, language acquisition, psycholinguistics, and historical linguistics.

Finally, please note that the goal of this course is *not* to teach you a lot of English words; there are dictionaries for that. You might end up learning some new terms in the course of the semester, but many of them will probably be specific to linguistics.

The website for the course can be accessed through Blackboard.

Does the syntax of phi-features (person, number, gender/noun-class) differ in significant ways from the syntax of other features? What is the proper representation of phi-features in syntax, and does it differ from their representation in morphology and/or in semantics? Does phi-feature agreement give rise to case-marking on nominals, or do structural cases (nominative, absolutive) arise even in the absence of agreement relations? Is there evidence for PF playing a role in agreement relations, and if so, what is the nature of this evidence? These questions are addressed in this one-week seminar.

• class notes can be downloaded here

In this course, we explore issues in the theory and typology of case-marking and agreement systems. Some of the issues discussed include: head-marking vs. dependent-marking; ergativity and split ergativity; quirky case, and the relation between morphologically overt case marking and so-called "abstract case"; the adequacy or inadequacy of recruiting case as the licensor of noun phrases in linguistic theory; and the oft-misconstrued relationship between case and agreement.

The website for the course can be accessed through Blackboard.

In the spring of 2012, I taught the undergraduate "Introduction to Linguistics" course at MIT. In this course, we tried to provide some answers to basic questions about the nature of human language. Throughout the course, we examined a number of ways in which human language proves to be a complex but law-governed mental system.

In the first two thirds of the class, we studied some core aspects of this system in detail. In the final part of the class, we used what we had learned to address a variety of other questions – including how language is acquired, how dialects arise, how languages change over time, and others.

The website for the course can be accessed here. (Note: most of the site is currently accessible to MIT students only.)

In the fall of 2011, I taught a graduate course at MIT called "Syntactic Models." In this course, we examined competing syntactic frameworks, both synchronically (GB/minimalism vs. HPSG vs. LFG) and historically (Generative Semantics vs. the "Conditions" framework vs. GB vs. minimalism), with the twin goals of (i) achieving literacy in the formalisms of these various frameworks; and (ii) understanding where the differences between these frameworks amount to actual differences in expressive power and empirical coverage, and where they do not.

• you can download the syllabus here
• the website for the course can be accessed here

In the spring of 2011, I taught a 3-day seminar at LUCL (Leiden), called "Agreement and case: Patterns, interactions, and implications."

• the handout for this course can be downloaded here

In the summer of 2010, I taught two courses at the EGG summer-school in Constanţa, Romania. I co-taught "Intro to Syntax" with Michal Starke, and taught a topics class on "Recent developments in (the theory of) ergativity."

• handouts/slide-sheets from the Ergativity course can be downloaded here

In the summer of 2009, I taught two courses at the EGG summer-school in Poznań, Poland. One was "Intro to Syntax"; the other was a course called "Agreement and its failures"  – which incorporated, among other things, some of my own research on Basque and on Hebrew.

• handouts/slide-sheets from both courses can be downloaded here