Meaning and Universal Grammar: Theory and Empirical Findings (review)

This set of studies is an intrepid quest to find ‘semantic primes’ or ‘primitives’, that is, meanings shared by all languages and taken to constitute the core grammar of all human language. The aim is no less than ‘to lay the foundations for an integrated, semantically-based approach to universal grammar and linguistic typology’ (I, 3). The book consists of two volumes, each containing five chapters. The first two and last two chapters are general theoretical discussions written by either or both of the editors. The six chapters in between are attempts to find primes in individual languages, two written by the editors and four by other contributors.

In the six language-specific studies, each author attempts to identify categories of meaning in individual languages that can ostensibly be taken to be exponents of primes. Cliff Goddard claims to find 59 such meanings in colloquial Malay; Catherine Travis, 59 in Colombian Spanish; Hilary Chappell, 59 in Mandarin Chinese; Robert Bugenhagen, 56 in Mangaaba-Mbula a.k.a. Mangap-Mbula (an Austronesian language spoken by approximately 3,500 people on Umboi Island in Papua New Guinea); Anna Wierzbicka, 60 in Polish; and Nick Enfield, 60 in Lao. All six contributions are sensitive and insightful semantic studies which clearly illustrate the methodology of the natural semantic metalanguage framework (NSM). At the same time, this exercise calls into question the validity of the methodology and a number of the theoretical premises of the NSM framework.

The core theoretical issues and the methodology of NSM are candidly presented in the four general chapters by the editors. In his ‘Opening statement’, Goddard kicks off by saying ‘the prime if . . . is postulated to occur universally in a biclausal frame, so that one could express in any language the semantic equivalent of a sentence like “if you do this, something bad will happen” (notwithstanding that in some languages the exponent of if coincides with the exponent of when)’ (I, 2). There’s the catch. If we let ourselves get away with this, we sidestep the obvious fact that the reason a naïve native speaker of German makes mistakes with English when and if is precisely because the analogous categories of meaning in both languages happen not to be semantically equivalent. Each such instance of ‘notwithstanding’ obviates the crucial point that in different languages people actually say different things.

A central premise is that ‘any language can be adequately described within the resources of that language’ so that all meanings can be characterized by reductive paraphrasis in natural language. These algorithms of reductive paraphrasis characterize meanings in discrete, natural, intelligible, and noncircular propositional terms. The NSM approach involves the empirical investigation of many languages. Its proponents rightly claim that, because NSM integrates meaning and syntax from the very outset, their empirical approach is superior to the meaningless linguistics of the Chomskyite school. The anachronism in the ‘Opening statement’, whereby Leonard Bloomfield is retroactively faulted with Noam Chomsky’s antisemantic vices, is jarring, however.

In practice, the NSM school has hitherto used plain English as the ‘natural semantic metalanguage’. A novelty is that each of the six language-specific contributions includes two ‘native’ NSM texts, one satisfactorily characterizing the concept of the Good Samaritan, the other offering a debatable interpretation of the Chinese philosophy of the Middle Way. In fact, these metalanguage texts have been transposed from English originals into colloquial Malay, Colombian Spanish, Mandarin, Mangaaba-Mbula, Polish, and Lao.

Postulated semantic primes include i, you, someone, something, this, good, bad, happen, move, know, think, feel, want, say, live, die, where, when, not, maybe, like, kind of, and part of. Yet what does NSM make of ubiquitous inconveniences of life such as the fact that Nepali has no word or expression that means the same thing as English feel? Nepali châmnu ‘touch, grope, [End...