Skip to main content Accessibility help
  • Print publication year: 2010
  • Online publication date: May 2011





Proto-Indo-European clearly had suffixes, but is sometimes assumed (e.g., Meillet 1964) to have lacked prefixes, an hypothesis that fits the relatively late process in Indo-European of univerbation, by which particles came to form a unit with a verb (e.g., Lat. ob uos sacro > uos obsecro ‘I beseech you’). Latin's few prefixes were predominantly prepositions in origin and often had aspectual value, cf. edere vs. comedere ‘eat (up)’. In vulgar and late Latin the increase in prefixed forms was often matched by loss of semantic nuances and replacement of the simplex forms by originally prefixed ones, such as Sp. comer ‘eat’ (see Väänänen 1966:106–8). The original prefixes are often unrecognizable, cf. *com-initiare > Fr. commencer, It. cominciare, Sp. comenzar. Some underwent semantic shift: e.g., It. rileggere ‘read again’ vs. riposare ‘rest’. Meyer-Lübke (1894) reports seventeen prepositions and three adverbs used as prefixes in Latin verbal derivation surviving in Romance; many of these also occur in nominal derivation. Verbal derivatives are most widespread, and the use of two opposing prefixes with an otherwise non-existent verbal base is common; Fr. embarrasser ‘hinder’, débarrasser ‘clear’, but no **barrasser.

Productive prefixes tend to occur in learnèd formations. Originally prepositions (and adverbs) borrowed from Greek and Latin, they combine with nouns, adjectives and verbs, e.g., auto-, hyper-, super-, extra-, anti-, inter-. They primarily convey intensification, reversal, negation, iteration and – unlike suffixes – lack evaluative force. Prefixes may occasionally occur as independent elements, often with a specific meaning, e.g., Fr. ex-député ‘former representative’ and mon ex ‘my ex-(husband)’.

Related content

Powered by UNSILO