This site © 2007 by the Faculty of English, University of Cambridge, CB3 9DP
Cynthia Lloyd
University of Leeds
Email: cynthia-l at tiscali dot co dot uk
I am studying the integration into English of the five nominal suffixes -ment, -ance (with its variant -ence), -ation, -age and -al, which entered Middle English via borrowings from French, and which now form abstract nouns in English by attaching themselves to various base categories, as in cord/cordage or adjust/adjustment. I wanted to consider the possibility that each of these five suffixes might individually affect the general semantic profile of nouns which it forms.
Booij (1986) and Plag (1999) have suggested a core meaning for groups of related suffixes, from which other meanings develop by extension rules. Earlier historical linguists, for example Marchand (1969) and Jespersen (1942) have seen the suffixes themselves as attracting a 'nexus' of thematic meanings such as Kastovsky’s (1986) agent, means or result of action, which are common to all of them and which cluster around a central semantic notion such as 'action/fact'. Marchand has further suggested that each suffix develops through time a unique combination of such meanings in a hierarchy of its own.
Semantic distinctions have been found between rival but related English suffixes such as de-adjectival -ness and -ity (Riddle, 1985), the adjective-forming suffixes -ish and -y (Malkiel, 1977) and the verbal suffixes -ate, -ize and -ify (Plag, 1999). I am investigating the possibility of comparable distinctions between the English latinate nominal suffixes.
By charting the meanings of each word in the sample in their first attributions, I hoped to arrive at a profile of the semantic preferences of each suffix as it first became established in the language.Then I compared the findings with the later use of words in these suffixes in selected texts by Shakespeare (that is, many of the same ME words plus later formations).
Even though all such nouns to a certain extent share a cluster of related meanings, my results suggest that these meanings were not shared indiscriminately, but that each suffix showed general preferences for certain semantic areas, both real-world and grammatical, which would be reflected in the real-world semantics of the bases chosen to form words, and in the grammatical contexts in which the words were used. Secondly, it seems to me that any deverbal noun may specialise in a distinct aspect of the central meaning 'action/fact', such as specific instance or quality. These aspects have been touched on by Marchand, but not considered systematically. Thirdly, such tendencies may be subject to change over time.