Retour vers liste

Détail de la contribution

Auteur: Philippe SCHLENKER

Co-Auteur(s): Emmanuel Chemla, Kate Arnold, Alban Lemasson, Karim Ouattara, Sumir Keenan, Claudia Stephan, Robin Ryder, Klaus Zuberbühler

Titre:
Dialectal variation in the meanings of Campbell's monkey alarm calls


Abstract/Résumé: Ouattara et al. (2009a,b) show that male Campbell's monkey alarm calls have a root-suffix morphological structure and a partly rule-governed syntax. Working with experimental data from the Tai forest (Ivory Coast) and Tiwai island (Sierra Leone), we argue that the same alarm calls are interpreted differently in the two communities. We conclude that either (a) part of the meanings is learned, or (b) the meanings are entirely innate, but environment-dependent. Either way, the semantic sophistication of alarm calls has been underestimated. [1] Main finding: In Tai, krak is often used in leopard situations, and hok in eagle situations. On Tiwai, leopards have not been seen for at least 30 years, but krak and hok are both used (as are other alarm calls). If meanings (i) were entirely innate, and (ii) solely pertained to the situation of utterance, vocalizations should be roughly the same in eagle situations in Tai and on Tiwai. This prediction is refuted: the proportions of hok and krak is roughly the same in Tai and on Tiwai in researcher-induced leopard situations; but in eagle situations, they are entirely different. Krak, which is primarily used in leopard situations in Tai, has the distribution of a general alarm call on Tiwai. [2] Theory I: One possible analysis, is that krak has (i) an innate meaning of general alert, and (ii) in Tai only, an acquired meaning pertaining to leopards. Only (i) is used on Tiwai; both (i) and (ii) are used in Tai, and the meaning of krak-oo is derived from (i). [3] Theory II: An alternative is that krak has an innate meaning of general alert and no acquired meaning, while krak-oo has an attenuative meaning ('a weak krak-type disturbance'). Crucially, we posit that the meaning of krak is strengthened by a process akin to scalar implicatures: krak-oo and hok are both more specific, hence from krak one infers the negation of krak-oo (hence: non-weak disturbance) and the negation of hok (hence: non-aerial predator). In Tai, strengthening applies maximally and krak acquires a pragmatic meaning of 'dangerous non-aerial predator'. On Tiwai, the same strengthening gives rise to a near-contradiction, since there are no ground predators; and thus the unstrengthened meaning is used instead.