An introduction to Montague semantics, also known as model-theoretic semantics or truth conditional semantics. The course covers some of the philosophical background as well as the intersection of semantics with syntax and pragmatics. Specific topics include: definitions of truth, predicate logic, quantification, and intensionality. The course assumes a bit of syntax but no mathematical or logical background. The text is Meaning and Grammar, 2nd edition (Chierchia and McConnell-Ginet, MIT Press). Students will also be encouraged to look at Logic, Language, and Meaning (L T F Gamut, Chicago Press) as well as other texts. The course is evaluated with a series of assignments during the semester (50%) and a final take-home assignment (50%)
Applications of speech and language processing are found everywhere today. Automated telephone systems, for example, incorporate voice recognition and synthesis. This seminar will explore how computers deal with natural language in such areas as speech recognition, speech generation, and machine translation. Intended as an introduction to the field, the course will survey a range of methodologies in speech and language processing and will cover the basic components of natural language systems, including the lexicon, syntax and parsing, semantic analysis and representation, discourse processing, and pragmatics.
This course, assuming no more than general familiarity with phonological concepts, offers an intensive introduction to the formal apparatus of modern generative phonology, with an emphasis on the development of fluency in analyzing phonological data. The presentation of material in class therefore assumes concurrent registration in the associated practicum (Ling 73600, Phonology I Practicum).
The basics of phonological description and theory -- inventories, distinctive features, natural classes, alternations, levels of representation, rule or constraint formulation -- are first introduced within the linear framework of classic generative phonology. With these basics in place, we motivate additions to the formalism -- feature geometry, autosegmental architecture, notions of underspecification, metrical representation -- in terms of their better capture of typologically common phonological phenomena. Finally, we review an altogether different analytic framework, Optimality Theory.
Taking the end-point of Syntax I as a starting-point (and as a prerequisite for registration), this course takes its participants from Chomsky’s (1981) original Government-Binding Theory all the way to the most recent incarnation of the principles-and-parameters approach to generative grammar: the minimalist program (Chomsky 1995). Along the way, it addresses, among other things: (i) Lasnik & Saito’s (1984) theory of ‘gamma-marking’ and intermediate trace deletion at LF, (ii) Chomsky’s (1986–Barriers) densely successive-cyclic derivations for A–movement via intermediate VP–adjunction, (iii) his unification of the theories of government and bounding, (iv) his analysis of A–movement and its dependency on head-chains, (v) the perspective on the general interdependency of A–movement and head-movement that the ‘equidistance’ based theory of locality in Chomsky (1993) gives rise to, (vi) the general premises of the minimalist program of Chomsky (1993) and Chomsky (1995), (vii) the reduction of the phrase-structure component of the theory (‘bare phrase structure’), and (viii) the connection between hierarchical relationships between constituents and their linear sequencing (‘antisymmetry’; Kayne 1994).
The course presupposes a solid command of the issues addressed in Syntax I (LING 72100), but no specific knowledge of syntactic theory beyond that point.
The Haegeman (1994) textbook that was used for Syntax I will continue to be used in the first weeks of classes; beyond that point, we will rely primarily on Chomsky (1995), The Minimalist Program (MIT Press) and materials that will be provided in class.
This is the second of a two- part course sequence to train students with a linguistics background in the core methodologies of computational linguistics. Successful completion of this two-course sequence will enable students to take graduate-level elective courses in computational linguistics; both courses are offered by the Graduate Center's Linguistics Program, as well as courses offered by the Computer Science Program. This course (Methods in Computational Linguistics II) will provide training in: the use of computational libraries built specifically for computational linguistics, the techniques used in performing computational analyses of electronic natural language corpora, and the foundational mathematics, probabilistic methods and statistics that are the backbone of modern computational linguistics. The course will go significantly beyond a survey of these topics. By completing the Methods in Computational Linguistics sequence, at the end of the first year, Computational Linguistics Master's students will have the skills they need to engage in further study of state-of-the-art topics in natural language processing.
The first 2 weeks will provide an overview of the field of language learnability, showing how it relates to other areas of computational, linguistic and psycholinguistic research. The goal of learnability theory is the development of models or hypotheses concerning the mental mechanisms of language acquisition, which reveal how it is possible for an infant, in 4 or 5 years, to attain command of the complex system that is a human language.
During the semester each student will choose a topic of interest to focus on, will consult with me for additional readings as needed, and at the end of the semester will give a short class presentation. Some example topics (in no particular order):
Formal proofs of (non)-learnability (from Gold’s theorem onward), and their potential implications for psychological models of language development.
Evidence for deterministic learning versus trial-and-erromodels (e.g. William Snyder’s acquisition data for preposition piping and stranding)
Child-directed speech: does it contain sufficient information for setting parameters? Does it determine the order in which parameters are set? (Charles Yang’s learning model)
Poverty of the stimulus issues (little or no negative evidence, incomplete positive evidence), seeking new examples to extend the scope of the longstanding debate.
What is a parameter value, in formal grammars of natural languages? Psychologically, can parameter setting plausibly be modeled as flipping switches?
The role of children’s sentence processing routines in acquisition of syntax; are they the only learning mechanism needed? (Sakas & Fodor, in press)
- Do learners apply the Subset Principle? If so, does it carry a heavy computational cost? (Fodor & Sakas 2005)
Parameter setting (or other learnability challenges) in the acquisition of phonology: Lisa Pearl’s extensions of classic work by Elan Dresher.
Learnability issues in semantics. How do children interpret quantifiers and scope, gradable adjectives, elided constituents?
Are statistical (connectionist, Bayesian) learning models able to acquire rich grammars without reliance on innate knowledge?
Do current learning models shed light on child bilingual acquisition?
This graduate level introductory course provides an overview of the field of sociolinguistics from its origins in dialectology and focus on the social stratification of linguistic variables across different social groups to more recent investigations of language variation based on aspects of identity such as ethnicity, gender, sexuality, and age. The course also examines contemporary research on language attitudes and ideologies, registers and styles, as well as languages in contact, globalization, World Englishes, and language use in New Media and Computer Mediated Communication.
This course will include supervision in data gathering and digital recording, linguistic databases and corpora, and data mapping. Enrollment is limited to graduate students already working on endangered language research projects.
The course will introduce Classical Chinese, also known as Literary Chinese, the language in which most of the pre-modern literature of China is written. Practical syntax and Chinese characters will be taught along with selections from Classical Chinese literature, including Confucian and Daoist texts, which will be introduced after the first few lessons. No previous knowledge of Chinese is required. Students who have had no Chinese before will learn the tonal system for reading Chinese properly, a skill applicable as well to speaking Modern Standard Chinese (popularly known as Mandarin).
Recent studies of language in its sociocultural context highlight the role of language ideologies and cultural conceptions of language in transforming social dynamics and power relations as well as language use and structure. In this seminar, we will explore theoretical frameworks and case studies from the Americas, Asia, Europe, Oceania, and Africa to study the relationship between language ideologies and social processes and their linguistic and social consequences. The topics considered include modern linguistics, colonialism, missionization, nationalism, globalization, identity formation, indigenous movements, dialect and language formation, language endangerment and revitalization.
This course presents basic knowledge about speech acoustics, production, and perception in a combined lecture/laboratory format. Laboratories are to be completed outside of class (approx 2 hrs/week). This is good preparation for the Speech Science First Exam or for courses in phonology. Students will write several short papers on various topics in speech science and acoustic phonetics: e.g., source-filter theory; myoelastic/aerodynamic theory of phonation, speech sound sources (how, where and by what physiological mechanisms they are produced), acoustic cues for vowels and for consonant manner, place and voicing; categorical perception. Two weeks will be split into two tracks, either practical implementation of speech science in the clinic, or issues of phonetic analysis of speech corpora.