The question of whether or not all languages have a word for 'want' (as in 'I know what you want, I want the same') is far more important than many linguists appear to realize. Having studied and debated this question for many years, we welcome Olesya Khanina's (2008) paper "How universal is 'wanting'?", which, we believe, addresses a question of fundamental importance. Our own view - which we have sought to substantiate in a large number of publications, over many years (cf. Wierzbicka 1972, 1996; Goddard 1991, 2001; Goddard and Wierzbicka eds. 1994, 2002; Peeters ed. 2006) - is that WANT is a universal semantic prime, i.e. an indivisible unit of meaning with a lexical exponent in all languages. In the present article, we argue that although Khanina has produced valuable results about cross-linguistic patterns in the polysemy of exponents of WANT, she has failed to demonstrate her concluding point, namely, "that 'want' is not a universal semantic prime in the sense of the Natural Semantic Metalanguage 堛and] that the inclusion of WANT in this list [of semantic primes] is indeed false" (p. 848). Briefly, we will argue that Khanina's conclusion depends, first, on an a priori decision not to recognise the existence of polysemy; and second, on a misunderstanding of the NSM position on what it means to be a lexical exponent of a semantic prime. We will also argue that 'wanting' constitutes an indispensable conceptual building block in human communication and cognition, and in linguistic and psychological theorizing about communication and cognition.