Goddard, Cliff (1998). Universal semantic primes of space – A lost cause? LAUD Working Papers, Series A, General and Theoretical Papers, 434. PDF (open access)
Reissued in 2007 with divergent page numbering.
In recent years, a new wave of research on language and space has uncovered surprising variation in the linguistic coding of spatial relationships. It is now known that some languages, e.g. Tzeltal, exhibit remarkable lexico-grammatical elaboration of spatial relationships; that in many languages of Africa and Oceania apparently simple spatial relationships such as INSIDE and ABOVE are encoded by means of noun-like words, or by a combination of a preposition and a postposition, each of which may be independently meaningful. It has also been shown that children’s early acquisition of spatial terminology differs markedly between typologically different languages.
In almost all this recent work, the emphasis has been on cross-linguistic variation in spatial semantics. The question then arises whether there any semantic universals of space that are still viable in the light of the attested variation in formal realization and lexico-grammatical elaboration. In particular, what of the semantic primes of space proposed within the Natural Semantic Metalanguage (NSM) framework, namely: WHERE/PLACE, HERE, ABOVE, BELOW, INSIDE, ON (ONE) SIDE, NEAR, FAR?
After an introduction, the body of the paper has three sections. The first argues that three languages that exhibit markedly different spatial characteristics to English (Tzeltal, Longgu, Ewe) nevertheless still contain exponents of the NSM spatial primes. The second takes a fresh look at some of the new results on cross-linguistic variation in the acquisition of spatial semantics, with particular reference to Korean. The third surveys the grammaticalization of spatial meaning in a typological perspective, concluding that the items on the NSM inventory of spatial primes are all found as recurrent dimensions of grammaticalized meaning in a range of languages.
The overall conclusion is that NSM’s spatial primes are both viable and necessary for the description of spatial meanings within and across languages.
Research carried out by one or more experienced NSM practitioners
