Mordva Finno-Ugric morph [Grk ‘form, shape’]

Theoretical basic element in structural language analysis, analogous to phoneme: the smallest meaningful element of language that, as a basic phonological and semantic element cannot be reduced into smaller elements, e.g. book, three, it, long. Morphemes are abstract (theoretical) units. They are represented phonetically and phonologically by morphs as the smallest meaningful, but unclassified, segments of meaning. If such morphs have the same meaning and a complement ary distribution or if they stand in free variation, then they are said to be allomorphs of the same morpheme, e.g. the allomorphs of the plural morpheme ‘s’ are /s/, /z/, /iz/ as in books, radios, and houses, though -s, -en, and -ø (as in doors, oxen, and sheep) constitute allomorphs of an abstract plural morpheme. Thus, only in particular cases do morphemes actually correspond to the grammatical category of word (word, we, soon); morphemes must be principally distinguished from the phonetic unit of syllable: syllables are concrete sound segments of a word on the level of parole, while morphemes are abstractions on the level of langue ( langue vs parole); any formal correspondence between morphemes and syllables is coincidental, cf. rent control, but: rent-al vs tor-rent. A syllable can consist of several morphemes: cf. thought, which can be analyzed as containing the four morphemes ‘think’

‘person,’ and ‘number,’ while today consists of two syllables but constitutes only one morpheme. Depending on the aspect of the study one can discern various typologies of classification and differentiation of morphemes. (a) Regarding the postulate of the unity of form and meaning a distinction must be drawn between (i) discontinuous morphemes, in which two or more morphs separated by other elements yield the morpheme’s meaning (as in Ger. ge+lieb +t, where ge-and -t together mark the participle) and (ii) the so-called portmanteau morphemes in which the smallest meaningful segments carry several meanings (cf. the analysis above of thought or Fr. au that is a blend of the morphs a and le). (b) Regarding their semantic function one distinguishes between (i) lexical morphemes ( lexeme), that denote objects, state of affairs, etc. of the extralinguistic world and whose relations are studied in semantics and lexicology and (ii) grammatical morphemes (also: inflectional morphemes) that express the grammatical relations in sentences and are studied in morphology (in the narrow sense) and syntax. (c) Regarding their occurrence or their independence one distinguishes between (i) free morphemes (also: roots or bases), which may have both a lexical (book, red, fast) as well as a grammatical function (out, and, it) and (ii) bound morphemes, in which it is a matter of either a lexical stem morpheme (e.g. typ-in type, typical) and inflectional morphemes (as in verb endings) or derivational morphemes of word formation (as un-, -able, -ness) ( affix). Also, so-called ‘cranberry morphemes’ (as cran-in cranberry) ( semimorpheme) are bound morphemes whose synchronic meaning is reduced to its distinctive function. This structuralist morpheme analysis, which is based primarily on distribution and operational processes of analysis, is limited by the changes in the forms that are not caused by relations of order, but rather are characterized by sound changes ( mutation), cf. the formation of the past tense in Eng. run: ran. See Matthews (1974) for a summary and critical view of these analyses. The relevance of the classical concept of morpheme to the description of word formation is doubted by Aronoff (1976), who eventually discards it. Accordingly, the lexicon does not consist of morphemes but rather of finished words of the language. According to Aronoff, outside the words in which they occur, morphemes have no independent existence: they are constituents of words. Word formation rules are interpreted as transformational operations within the lexicon that take a word as input and transform the same into a new word with phonologically, semantically, and syntactically determined characteristics. See Di Sciullo and Williams (1987) for criticisms of Aronoff s approach. In contrast to Aronoff, they posit combinatory word formation processes that combine morphemes into words.