Lexical diffusion

From Wikipedia, the free encyclopedia
Jump to: navigation, search

In historical linguistics, lexical diffusion is both a phenomenon and a theory. The phenomenon is that by which a phoneme is modified in a subset of the lexicon, and spreads gradually to other lexical items. For example, in English, /uː/ has changed to /ʊ/ in good and hood but not in food; some dialects have it in hoof and roof but others do not; in flood and blood it happened early enough that the words were affected by the change of /ʊ/ to /ʌ/, which is now no longer productive.

The related theory, proposed by William Wang in 1969, is that all sound changes originate in a single word or a small group of words and then spread to other words with a similar phonological make-up, but may not spread to all words in which they potentially could apply. The theory of lexical diffusion stands in contrast to the Neogrammarian hypothesis that a given sound change applies simultaneously to all words in which its context is found.

William Labov, in Principles of Linguistic Change, takes the position that there are two types of sound changes: regular sound change (respecting the Neogrammarian hypothesis) and lexical diffusion. Labov lists a typology, according to which certain phenomena are typically or exclusively regular (example, vowel quality changes), while others (example, metathesis, or vowel shortening) tend to follow a lexical diffusion pattern.

Paul Kiparsky, in the Handbook of Phonology (Goldsmith editor), argues that under a proper definition of analogy as optimization, lexical diffusion is not a type of sound change. Instead, Kiparsky claims it is similar to leveling, in that it is a non-proportional type of analogy.