I. FORMAL SEMANTICS AND PRAGMATICS

Pre-requisites

Competent semanticists are expected to have taken a fairly standard set of courses that include the following (or some equivalent) at the graduate level (see here for some of my old lecture notes on semantics):

Semantics I: extensional semantics (e.g. as in Heim and Kratzer's Semantics in Generative Grammar, or Gamut)

Semantics II: intensional semantics and advanced topics in extensional semantics (e.g. as in von Fintel and Heim's lectures notes on Intensional Semantics; see also here);

Formal Pragmatics: including in particular contemporary theories of implicatures, presuppositions, and supplements.

Some experience with formal syntax as well as the syntax/semantics interface is strongly recommended: any semantics is the semantics of a syntax, and one ignores the latter at one's peril.

A solid background in philosophical/mathematical logic and in the philosophy of language can be extremely helpful.

In addition:

– semanticists who wish to be considered for teaching positions in linguistics departments are expected to be able to teach general introductions to linguistics with at least some phonology and some syntax in addition to semantics/pragmatics. Thus a general education in linguistics is strongly advised as well. 

– work in any branch of linguistics involves experimental methods; today's young semanticists are increasingly conversant in psycholinguistics.

Education in semantics (and more generally in linguistics) can be obtained in several excellent graduate programs, both in Europe and in France, including at ENS.

Further expectations

Students who consider writing a dissertation in semantics should strive to work (i) on several topics, and if possible (ii) in interaction with diverse researchers before they pick a dissertation topic. This is a standard requirement in PhD programs in the US, where the thesis itself is only a part of the program (1 or 2 years), preceded by (a) advanced introductions to the various subfields of linguistics, and (b) 2 research-level papers ('qualifying papers') in different subfields. 

A large part of a linguist's training consists in analyzing complex sets of data and thus 'hands on' experience with linguistic analysis is crucial.

Job situation

The academic job market in linguistics is not easy at the international level: there are many good graduate programs that produce excellent semanticists, and competition for a small number of positions is fierce. It is possible to find semantics positions at French universities, but this requires some knowledge of French, as well as a specific procedure (involving 'qualification' by the Conseil National des Universités). 

Topics

I have conducted research on the following topics:

Modal constructions and  attitude reports (with special reference to context shift)

Indexicals

Anaphora

Presuppositions

Supplements: expressives and appositives relative clauses

Each of these topics can be investigated in parallel in spoken and signed languages - see below.

II. SIGN LANGUAGE SEMANTICS

Sign language semantics is a recognized but emerging subfield of linguistics. The following pertains an ideal education in this area.

Pre-requisites

Ideally, specialists of sign laguage semantics should have the general training of formal semanticists as described above. But they should also be competent sign language linguists. The following would be particularly helpful:

1. Take as early as possible courses in the sign language that you will be working on. Conducting elicitation work with deaf consultants requires a detailed knowledge of the language - the more detailed, the better. As with any language, practice makes perfect, and it can be acquired by interacting with deaf colleagues and friends, and by watching sign language videos. 

(It should go without saying that Deaf students and native signers are particularly welcome in linguistics in general and in sign language linguistics in particular.)

2. ASL is, de facto, one of the international sign languages in science (although International Sign  is also used at conferences). In addition, there is an important body of work on ASL. Students working on sign language might consider spending several months early in their career at Gallaudet University in order to (i) learn ASL (there are intensive summer programs), and (ii) if possible, take classes on linguistics or other topics taught in ASL. This would be a huge help for a student's longterm career. (It should be noted that Gallaudet University is in Washington DC, not very far from Maryland University, which has an excellent Linguistics Department - particularly noted for the interaction between linguistic theory and psycholinguistics. It is apparently possible to cross-register from Maryland to Gallaudet and vice versa).

3. Specific courses on various aspects of sign language linguistics would of course be helpful.

4. Hands-on experience with elicitation work with deaf consultants is crucial to sign language research – and it is time-consuming. It is wise to think early on about optimal elicitation methods, recording methods (to ensure that no data are ever lost), and dissemination of raw data (to ensure maximum transparency). 

Job situation

The job situation in sign language linguistics is generally difficult, as is the case in linguistics in general. Postdoc possibilities might be a bit more common than in other areas of linguistics, however.

Topics

I have conducted on the following topics within sign language semantics:

Anaphora

Indexicality

Focus

Plurals 

Logic and iconicity in sign language

Sign/gesture comparison

Formal pragmatics in sign language, with special reference to presuppositions

Note: one general issue in gesture and sign language semantics pertains to the construction of an explicit semantics with iconicity. It is for instance essential in order to understand cases in which discourse referents are made available by purely iconic means. An 'iconic semantics' has been developed for pictures by philosopher Gabe Greenberg, but too little has been done so far on the linguistic side

III. GESTURE SEMANTICS AND PRAGMATICS 

While there is a long tradition of linguistic and psycholinguistic work on gestures, it is only recently that this has become a research topic in formal semantics. One should keep in mind that a pure specialization in gesture semantics might be considered by some to be overly narrow. Thus it is wise to acquire a general education in semantics/pragmatics (as detailed above), and to do gesture work in addition to work on other topics. 

Pre-requisites

For reasons stated above, pre-requisites should be a version of those that are standard for semantics/pragmatics. Due to the importance of experimental methods in emerging work on gestures, a good background in psycholinguistics might prove very helpful as well.

It should be kept in mind that the comparison between some aspects of signs and gestures is a topic of increasing importance. Thus gesture specialists should strive to learn a lot about sign languages (and possibly even learn a sign language).

Job situation

A pure specialization in gesture semantics is probably not optimal for the job market. But there is increasing interest in gesture semantics as an emerging new topic. Because the data are mostly new relative to more traditional topics, it is particularly important to think in detail about elicitation methods and possibly experimental extensions.

Topics

I have conducted research on the following topics (experimental work is typically conducted in collaboration with Emmanuel Chemla and Lyn Tieu):

Co-speech gestures (which co-occur with spoken words they modify)

Post-speech gestures (which follow spoken words they modify)

Pro-speech gestures (which fully replace spoken words)

Inferential typology (presuppositions, implicatures, supplements, etc.) obtained with pro-speech estures

Grammar of pro-speech gestures

Gestures in sign language (with special reference to non-grammatical facial expressions)

IV. PRIMATE LINGUISTICS

In recent years, the LINGUAE group, collaborating with Klaus Zuberbühler's primatology group, has published a series of studies on 'primate linguistics', the investigation of primate communication by way of formal models (which use the general methodology but rarely the specific use of linguistics). A summary can be found here. It is important to keep in mind, however, that as things stand there is no academic field of primate or animal linguistics. Students who are interested in doing work in that area could consider two general directions:

1. becoming ethologists (with a full education in that area), while acquiring relevant formal and analytical methods in linguistics and allied fields [relevant ethology groups working on animal communication include those of Klaus Zuberbühler, Alban Lemasson, and Simon Townsend].

2. becoming linguists, and keeping an interest in animal communication as one research topic among several others (most recent pieces in primate linguistics have been published in linguistics journals).

Yet another possibility would be to study animal linguistics within philosophy.

Pre-requisites

They will entirely depend on the chosen path: an education in linguistics is entirely different from an education in primatology. 

Job situation

The job situation in linguistics is, as mentioned above, difficult. The job situation in animal communication within ethology might be even more difficult. There are some postdoc possibilities in that research area, however. 

Topics

My research has been conducted in collaboration with Emmanuel Chemla and Klaus Zuberbühler, among others. New projects would typically take the following form:

(i) identification of a dataset (on any animal) that offers the potential for clear and challenging generalizations about meaning;

(ii) contacting owners of the dataset to ask whether they might be potentially interested in a collaboration;

(iii) analysis of raw transcriptions and construction of new theories;

(iv) possibly, definition (in tandem with ethologists) of experiments that would help decide among competing theories.

V. MUSIC SEMANTICS

While there is a long tradition of investigations of musical syntax  (see Lerdahl and Jackendoff), formal studies of musical meaning are very recent; in fact, the very idea that music has a semantics is still controversial (for good reason). There is definitely no academic field of 'music semantics', and thus students who are interested in this topic should either investigate it 'on the side' (e.g. as linguists, or as philosophers), or acquire a very solid education in music cognition. Two excellent music cognition groups with clear interests in music semantics are:

Jean-Julien Aucouturier's group at IRCAM in Paris

Martin Rohrmeier's group at EPFL in Lausanne

Pre-requisites

They will entirely depend on the chosen academic path. For linguistics, see above. Education in music cognition requires considerable knowledge of musicology, of experimental psychology, and if possible of formal methods. The main challenge is to combine:

(i) a deep understanding of one or (preferably!) several musical traditions;

(ii) a good knowledge of traditional and scientific theories, including those that are stated with formal methods;

(iii) a good understanding of psycholinguistic tools that make it possible to test these theories.

Since few people have all these skills together, quite a bit of work is routinely of a collaborative nature.

Topics

My research on music semantics has so far consisted in the definition and illustration of a general framework (described in some lectures and papers available here). The next steps could be of two types:

(i) analysis of snippets of real music, by way of:

a. constructions of minimal pairs (by modifying the score or the interpretation of the snippet);

b. assessment and explanation of the ways in which these modifications affect the semantic effects obtained;

c. if possible, experimental verification of the fact that subjects genuinely obtain those associations.

This work requires knowledge of composition, or collaboration with composers or very competent professional musicians.

(ii) experimental test of various claims made about musical and non-musical stimuli (as well as their correlations) made in initial work on music semantics.