Forschungs- & Entwicklungsinformationsdienst der Gemeinschaft - CORDIS


FRONTSEM Report Summary

Project reference: 324115
Funded under: FP7-IDEAS-ERC

Mid-Term Report Summary - FRONTSEM (New Frontiers of Formal Semantics)

Formal semantics, the investigation of meaning with tools from formal logic, has been a success story of contemporary linguistics. The project 'New Frontiers of Formal Semantics' has sought to expand its domain of application in two new directions.
First, the project contributed to the development of a formal semantics of sign language, with two main results. One is that sign language sometimes reveals the 'Logical Form' of sentences in cases in which it is covert in spoken language. In such cases, sign language has a more transparent and explicit logical structure than its spoken language counterpart. A case in point pertains to logical variables, which are explicitly realized as positions in signing space called 'loci' in sign language, but which must be inferred rather indirectly in spoken language. The second result is that along some dimensions sign language is more expressive than spoken language because it makes use of rich iconic resources, including at its logical core. For instance, 'loci' may simultaneously function as logical variables and simplified pictures of what they denote – hence an interesting challenge for formal approaches to meaning. This result suggests that in some respects spoken languages can be seen as a 'simplified' version (mostly devoid of iconic means) of a richer logical system whose full extent can only be studied in sign languages.
A second part of the project was devoted to an entirely different extension of the program of formal semantics, namely to objects that are less rather than more sophisticated than the traditional ones. The project pioneered the study of 'primate semantics', which is the investigation of the meaning of primate calls with the general tools and methods of formal semantics. Case studies have been developed for 5 species of Old and New World monkeys, with two main findings. First, as Ouattara et al. 2009 had argued, some calls (those of Campbell's monkeys) arguably have a non-trivial root/suffix structure, whose semantics was investigated in detail. Second, although the project emphatically did not argue that non-trivial similarities can be found between 'monkey languages' and human language, it did find that one key concept of contemporary semantics proves illuminating in the analysis of primate systems. In a nutshell, when a call m is strictly less informative than a call m', an utterance of m suggests that m' could not have been used truly – with the result that m comes to often mean m and not m'. This mechanism has gained wide acceptance in linguistics under the name 'implicature', and it is for instance used to explain why p or q comes to mean p or q but not bot because it competes with p and q, which is more informative. If we are right, there are numerous 'monkey implicatures' in the wild.


AVIGDOR, Louis (Responsable du Service Partenariat et Valorisation)
Tel.: +33 1 42349417
Record Number: 183399 / Last updated on: 2016-05-27
Information source: SESAM