Introduction

What is the relation between the medium in which we think and the medium in which we talk? Opinions have varied with intellectual character and climate: some, from Wilhelm von Humboldt to Benjamin Whorf, from Lev Yygotsky to Daniel Dennett, have imagined language to be the great facilitator; others, like Bertrand Russell or Gilbert Ryle (or for that matter Alfred Korzybski), have conceived of language as a veil of confusion that hides the cognitive essentials. On top of this, some have assumed that languages will all have the same underlying semantics; others have assumed they will differ deeply at a conceptual level. These opinions have all the hallmarks of speculation without empirical constraint, and one could be forgiven for thinking that the subject should be left, like other phantoms, to the philosophers. But I believe that we can at least hedge in the fantasies by working in two directions. The first is to apply first principles to a few crucial facts about language. The second is to treat the problem as an empirical one: very well, let us roll up our sleeves and investigate whether a difference in linguistic conceptualization is or is not correlated with a difference in pattern of thinking. In this chapter, I sketch these two tacks in the last section, hoping to suggest that real progress could be made, while in the first section I try to sweep the ground clear of obvious misconceptions.