Over 40 years ago, Peter Marler proposed that animal signals were adaptive because they provided listeners with information (Marler, 1961, Journal of Theoretical Biology, 1, 295e317). But what was the nature of this information? How did it influence behaviour? And how might the information in animal signals compare with the information in human language? Here we review evidence that signals in a variety of social contexts are adaptive because they convey information. For recipients, meaning results from the integration of information from the signal and the social context. As a result, communication in animals e particularly in long-lived, social species where the same individuals interact repeatedly e constitutes a rich system of pragmatic inference in which the meaning of a communicative event depends on perception, memory and social knowledge. In the human lineage, pragmatics served as a precursor to the later evolution of semantics and syntax. Among primates, there is a striking difference in flexibility between constrained call production and more flexible perception and cognition. However, call production is more flexible in the wild, where it is affected by contextual cues, than in laboratory studies where contextual cues have been removed. Monkeys and apes may overcome the limits of constrained vocal production by producing composite signals in the same and different modalities.In this essay we focus on an issue that was central to Peter Marler's research on animal communication: the meaning of animal signals.Research on animal communication has made almost no progress in understanding the semantics of natural signaling behavior because it is an exceedingly difficult problem in both practical and theoretical terms. … There is an irresistible tendency to use language as a model, either for comparison or contrast. This would be more appropriate if we really understood human semantics and the processes by which language acquires meaning in the course of our own early development. In many respects our picture of how words acquire meaning in human infancy is hardly any clearer than our understanding of the meanings of signals for animals … In both human and animal studies, for example, there are assertions of the overwhelming importance of contextual cues in understanding meaning (Smith, 1977), but precisely how the context influences meaning in particular cases has hardly been explored.
(Marler, 1983)Marler's interests in the meaning of animal signals brought him squarely into the revolution in neurobiology, psychology, linguistics and cognitive science. When he began his research in the 1950s, questions about the 'minds' of animals had largely been ignored, because both psychologists like Skinner and ethologists like Tinbergen thought they were unanswerable and hence unsuited to scientific inquiry (Boakes, 1984;Burkhardt, 2005). Over the years, Marler developed a different view (Marler, 1961). He thought that, regardless of whether or not they were voluntary or intentional, animal signals provided listeners ...