3. Communicative Conception and Cognitive Conception (Introduction)

What is the place of natural language in human cognition and thought? People’s views on this matter have differed very markedly, dividing roughly into two opposing camps, as follows. On the one hand, the communicative conception of language: As the name suggests, the exclusive function and purpose of language is the communication of thought, where thought itself is largely independent of the means of its transmission from mind to mind. The communicative conception is now dominant in many areas of the cognitive sciences, that is broadly to include cognitive psychology, empirical-minded philosophy of mind, linguistics, artificial intelligence (AI), and cognitive neuroscience, for reasons which we shall soon begin to explore. On the other is the cognitive conception of language, which sees language as crucially implicated in human thinking. Roughly speaking, on this view, the natural language sentences are the vehicles of our thoughts. Those espousing the cognitive conception have not claimed that language is used exclusively for thought, of course; they have allowed that it is also used in communication. Carruthers sets up an opposition between the communicative conception and the cognitive conception, and then argues the cognitive conception. According to the communicative conception, “the function and purpose of natural language is to facilitate communication and not … to facilitate thinking” (Carruthers 1996a: 1). According to the cognitive conception, in contrast, “we often think in language, and the trains of reasoning which lead up to many of our decisions and actions will consist in sequence of naturallanguage sequences” (1996a: 2).

In discussing the cognitive conception of language, care needs to be taken to distinguish a requirement-thesis from a constitution-thesis. It is one thing to say that language is required for, or is a necessary condition of thought, or certain kinds of thought. Everyone should allow that some form of requirement-thesis has at least a degree of limited applicability, since it is plain that children rely upon language to acquire many of their beliefs and concepts. It seems obvious that a languageless person could never entertain thoughts about electrons or neutrinos, genes or cell-division, for example, since it is only through language that we can learn of these things. It is quite another, and more interesting, thing to say that language remains implicated in the beliefs so acquired -in such a way, for example, that a thought about electrons can only be actively entertained by activating a representation of the word ‘electron’.

In recent decades, according to Caruthers and Boucher’s observation, many of those working in the cognitive sciences have become convinced that the mind is more-or-less modular in structure, being made up of a number of distinct components; which are innately configured, and specialised for particular domains. Anyone influenced by this modular model is likely to regard the cognitive system which underpins language as just such a module. And then it seems almost inevitable that they should come to endorse the communicative conception of language, coming to regard language as but an input and output module for central cognition.

The modular model of the mind has been highly influential and productive. But it has tended to eclipse the cognitive conception of language, and to marginalise work which does not fit in easily with its dominant paradigm. To that extent, the relationship between language and thought has been relatively little discussed in recent decades; since many have thought the issue to be closed. The currently popular version of the communicative conception of language and its related emphasis on innateness and modularity are being questioned. Many of them explore the middle ground between the traditional extreme views of the relationship between language and thought, and some seek to rehabilitate a cognitive conception of language in modified form.

In the subsections which follow, I first give a brief overview of the traditional polarised views on the relationship between language and thought-beginning with the currently less-popular cognitive conception of language, and culminating with a brief description of the modular model of the mind, and the way in which it has generated a distinctive version of the communicative conception of language.