Abstract
Studies of animal communication are often assumed to provide the ‘royal road’ to understanding the evolution of human language. After all, language is the pre-eminent system of human communication: doesn't it make sense to search for its precursors in animal communication systems? From this viewpoint, if some characteristic feature of human language is lacking in systems of animal communication, it represents a crucial gap in evolution, and evidence for an evolutionary discontinuity. Here I argue that we should reverse this logic: because a defining feature of human language is its ability to flexibly represent and recombine concepts, precursors for many important components of language should be sought in animal cognition rather than animal communication. Animal communication systems typically only permit expression of a small subset of the concepts that can be represented and manipulated by that species. Thus, if a particular concept is not expressed in a species' communication system this is not evidence that it lacks that concept. I conclude that if we focus exclusively on communicative signals, we sell the comparative analysis of language evolution short. Therefore, animal cognition provides a crucial (and often neglected) source of evidence regarding the biology and evolution of human language.This article is part of the theme issue ‘What can animal communication teach us about human language?’
Highlights
I have not, to my knowledge, spoken the word ‘octopus’ today or in the past week, but no one would conclude that I lack the concept OCTOPUS
I have spent many hours observing these creatures and read books about them but, like most of my mental concepts, OCTOPUS goes unexpressed in my speech most of the time
This is true of concepts captured by single words but for more complex cognitive constructs that I possess but have never spoken at all
Summary
I have not, to my knowledge, spoken the word ‘octopus’ today or in the past week, but no one would conclude that I lack the concept OCTOPUS (here I follow the philosopher’s convention, when necessary, of denoting conceptual representations in capital letters). Freeing ourselves from the shackles of this prescientific intuition is the first step to insightful scientific analysis Embracing this indirect, two-step nature of reference, I can state my argument more clearly: the first stage of reference—building representations that tie sensory input to conceptual representations—is built upon a chassis of cognitive processes (sensory processing, recognition, categorization, combination and inference) that has fundamental shared components between humans and other animals. Two-step nature of reference, I can state my argument more clearly: the first stage of reference—building representations that tie sensory input to conceptual representations—is built upon a chassis of cognitive processes (sensory processing, recognition, categorization, combination and inference) that has fundamental shared components between humans and other animals It was once common to take a link between concepts and language as definitional, such that a ‘true’ concept must be linked to a word [17,18], but this traditional notion seems unsustainable in the face of infant research, where infants can clearly represent and reason about things they have no words for [19,20,21,22]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
