Layman's Summary

Linguists distinguish between "phonetics" and "phonology". "Phonetics" is the study of the physical, real-world aspects of speech and speech perception; for example, it is a fact of phonetics that the sound d is harder to say (in most contexts) than the sound t. Phonology is the study of the patterns of speech sounds in various languages; for example, it is a fact of phonology that there are languages that do not have the sound d. Linguists also generally believe that phonetic facts affect phonological patterns: that some languages lack d because of the phonetic fact that d is more difficult to pronounce than t.

If a language has both an "easy" sound and its "harder" counterpart, the easy sound will occur more often in that language than the hard one; for example, there are more t's in English than d's. Logically, this difference could come from either phonetics, or phonology, or both: maybe t is more frequent than d because it's easier to pronounce, or maybe because some languages lack d but very few lack t, or maybe both. The question I ask is which of these possibilities is actually true: are these imbalances related to phonetics, or to phonology?

I examined two cases where phonetics and phonology behave differently, and found that in both cases, these frequency imbalances matched the predictions of phonology but not of phonetics. My conclusion is that frequency imbalances are related to phonology, not phonetics.

Summary

Phonological markedness is known to be reflected in (lexical) frequency: a marked sound occurs less often in a given language's lexicon than its unmarked counterpart. Phonological markedness is also known to be heavily influenced by phonetic pressures. I investigate two cases of "underphonologization", in which phonological and phonetic patterns diverge, and compare their predictions to actual lexical frequency data from several languages. In both cases, lexical frequency reflects the phonological patterns, not the phonetic ones; thus, I conclude that lexical frequency is sensitive to phonological pressures rather than phonetic ones.

Downloadable Stuff

(See CV for citation information. Listed from newest to oldest.)

Symposium on Phonologization Handout
2008 Lexical sensitivity to phonetic and phonological pressures
Updated presentation; adds data on Korean.

Qualifying Paper
2008 Markedness and phonetic grounding in nasal-stop clusters
See section 3.2 for a prose writeup of the material presented at the OCP.

OCP Handout
2008 What the lexicon knows about phonology
Handout from a presentation given at the Old World Conference in Phonology 5. Includes data on English, German, Dutch, and French.