¹ W. H. Goodenough, Property, kin, and community on Truk (Yale University publications in anthropology, No. 46; 1951), hereafter PKC.

² Significatum and denotatum are used as defined by Charles Morris, Signs, language and behavior 17 (1946).

³ I wish to thank Henry M. Hoenigswald for his encouragement and John Cole for many fruitful discussions. Field work on Truk was undertaken in 1947 in connection with the Yale University expedition under George P. Murdock, part of the Coordinated Investigation of Micronesian Anthropology sponsored by the Pacific Science Board of the National Research Council, financed by the Office of Naval Research, the Wenner-Gren Foundation for Anthropological Research, and the Department of Anthropology of Yale University.

⁴ Leonard Bloomfield, Language 162 (1933). Just as a phone is a material (behavioral) manifestation of a phoneme, a denotatum is a manifestation of a significatum. For different applications of Bloomfield's terminology see Eugene A. Nida, Word 7.1–14 (1951), and Joseph H. Greenberg, Language in culture 3–19 (ed. Harry Hoijer; 1954). As will appear more plainly later, I use the term sememe in a somewhat different sense from Bloomfield.

⁵ I would carry the analysis further than Nida (8–9), who does not differentiate signification from connotation and metaphorical usage. By staying with linguistic forms and treating meanings as if they were self-evident, he fails to come to grips with the semantic problems. He turns his back on the ‘common denominator’ approach because he has no analytical method for going beyond a simple listing of allosemes. Any speaker of a language is able to use a given form in new contexts in ways perfectly intelligible to other speakers. Analysis must enable us to do the same.

⁶ Nida writes (6) : ‘A seme may be defined as (1) the meaning in a particular type of context of (a) a morpheme or (b) a formal part of a morpheme, or (2) a meaning implicit in the forms of a paradigmatic series. Semes of type 1 are overtly symbolized and those of type 2 are covertly indicated.‘ We shall see that his type-1 semes are combinations of his type-2 semes. The former are what I call sememes, the latter what I regard as the basic components of signification.

⁷ By kinship I mean a series of Trukese social distinctions, and the terms signifying them, which more closely fit the cross-cultural concept of kinship than any other series of distinctions known in Truk.

⁸ W. H. R. Rivers, The Todas (1906).

⁹ All kinship expressions are cited in the first-person singular form. Some morphemes which signify kinship with possessives do not do so without them; other kinship morphemes occur only as bound forms in possessive formations. The orthography is explained in PKC 26–8.

¹⁰ This is true, of course, only with respect to descriptions of significata. My uncle's uncle, for example, may or may not be a kinsman of mine. The connotation, however, is that he is not; for the expression rarely occurs where the expression my great-uncle can be used.

¹¹ For further comment on the fact that expressions on similar morphological levels may belong to different semantic levels, see Lg. 31.244–5 (1955).

¹² Though I seem to use the term differently, my use of lexeme in fact corresponds quite closely to that of Eleanor H. Jorden, The syntax of modern colloquial Japanese 8–26 (Lg. dissertation No. 52, 1955). The main difference is that I have explicitly added significational criteria to the definition.

¹³ All these Trukese matrilineal groups are described in PKC.

¹⁴ If a + b denote the class of all things that are either in a or in b but not in both, and if a + b = 1 (where 1 is any class of which a and b are subclasses), then a is the complément of 6.

¹⁵ When α = b in the sense that any denotatum of a is also a denotatum of b, and conversely, then a and b are synonyms.

¹⁶ These three clusters are known respectively as pwiipwiicëk ‘just pwii’ pwiipwi winisam ‘pwii with semej in common’, and pwiipwi winipwyny ‘pwii with pwynywej in common’.

¹⁷ Language 162; see also Floyd G. Lounsbury, Oneida verb morphology 11 (Yale Univ. publ, in anthr., No. 48, 1953), and Nida, op. cit.

¹⁸ The asterisk indicates a morpheme. Strictly speaking all these are allomorphs of a morpheme *fina (as in fine-n nöömw and fin acaw) ; but to introduce the other allomorphs would complicate the presentation without affecting the point to be illustrated.

¹⁹ Journal of the Royal Anthropological Institute 34.77–84 (1909).

²⁰ Social structure 96–7 (1949).

²¹ For recent theories and a review of earlier ones, see Murdock, Social structure 113–83.

²² This was, in fact, the view published in PKC 107–8.

²³ Murdock 106–12.

²⁴ I am indebted to John Cole for first calling this to my attention.

^24a See, for example, Harold C. Conklin's analysis of Hanunoo color categories, Southwestern journal of anthropology 11.339–44 (1955).

²⁵ See my comments in Lg. 31.243, and my discussion of kinship behavior as related to kinship terms in PKC 115–8.

²⁶ For examples of such analyses, see PKC 111–9.

²⁷ Alfred L. Kroeber, The nature of culture 3 (1952).