This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Lecture notes: 9.1 July 21, 2009 1 Morphology 1.1 Allomorphy When we looked at English plurals, we saw some allomorphy in action. Allomor- phy: cases where a single morpheme has more than one different realization. Allomorphy is a very frequent phenomenon. One question that always arises is whether the allomorphy is something special about the morpheme in question, or whether were just seeing the general phonology of the language at work. Consider another case from English: the prefix un-: (1) Prefix un-: [n] untoward undignified unsurprising unthinking unanswerable [m] unperishable unbalanced unfavorable [ N ] uncaring uncharacteristic ungallant Here we have allomorphy affecting the final consonant of the prefix. In short, we get [m] before labials, [ N ] before velars, and [n] elsewhere. So we do have allomorphy here, but is it particular to this morpheme? No because the same thing happens with other morphemes: (2) indecent inhospitable 1 impossible inconceivable (3) non-default non-binding And more generally, we find that (almost across the board) English nasals have the same place of articulation as a following consonant: we find lots of words like i[n]dustry and i[ N ]carnate, but (almost) never find words like imdustry or i[ N ]teresting. That suggests that English has a general rule where nasals take on the place of a following consonant. That suggests that the allomorphy we observed with un- is really just a few examples from a more general phonological processes. One thing we want to do is build lexical entries for morphemes. We have made the (important) assumption already that morphemes, and not necessarily whole complex words, are stored in the minds of speakers. What exactly is stored the kind of encyclopedic knowledge about particular en- tries is referred to as a lexical entry. (The whole collection is called a lexicon in the case of one speaker, the mental lexicon of that speaker.) A lexical entry should contain a few kinds of information: the most basic pieces are the phonlogical shape of the morpheme, and the meaning. Well also see some other information we need, like what kind of morphemes or words the morpheme subcat- egorizes for. So the question we have about the lexical entry for this prefix: do we list all the allomorphs separately? Or do we just list one and let phonological rules take care of the rest? Logically, we could go either way. Empirically, its hard to test what people have stored in their brains. (Some efforts have been made at this, but its complicated.) Another important assumption we also make: List as little as possible; rely on rules that are as sweeping and broad as possible....
View Full Document
- Spring '09