Olowkow
Philosopher
- Joined
- Oct 29, 2007
- Messages
- 8,230
This is just becoming intolerable.
I never said these existed in surface structure. These forms exist only in an underlying representation.To correct what you said:
Nueva Gramática ... 4.6f; and mainly 8.3 Verbos derivados en -ear y sus variantes (I). Sus bases léxicas, 8.4 Verbos derivados en -ear y sus variantes (II). Usos traslaticios, and 8.5 Verbos derivados en -ear y sus variantes (III). Alternancias verbales. (not much, just 15 pages, as the work is a synthesis).
There's no *clique, nor *chate in Spanish. Just clic and chat. Or pinche/pique y charla. "*Mope" doesn't exist either, except as a bad translation or Spanglish in the Spanish versions of sites mainly written in English and intended to serve the Hispanic market in the States. ..
If you prefer a class of verbs [+ear] with endings such as /+eo, eas, ea, eamos, ean/, in addition to the [+ar] class, fine. Go for it. The generative approach would prefer a deep structure rule that takes a noun "chat" ending in a consonant, inserts "e" for phonological purposes, and uses the common [+ar] infinitive suffix. >>>/chat+e+ar/ The nouns ending in a vowel remove that vowel, insert "e" and do the same: "mopa" /mop+e+ar/
If you say "el mop", then fine.
It takes a while to catch on. We're talking synchronic not historical derivations from a posited deep structure that hopefully represents what the native speaker "knows" or internalizes as linguistic knowledge.Generative phonology is a component of generative grammar that assigns the correct phonetic representations to utterances in such a way as to reflect a native speaker’s internalized grammar.
The challenge is to describe what a speaker needs to internalize in order to speak his language. One option would be to memorize all the well formed utterances possible, but that is an infinite set. So, some other principle must be involved, namely rule based derivations. If rules are the answer, then the fewer the better. Occam's razor.
Lots of literature on the subject. I assumed that someone entering into such a discussion would have been aware of the basics.Generative phonology posits two levels of phonological representation:
- An underlying representation is the most basic form of a word before any phonological rules have been applied to it. Underlying representations show what a native speaker knows about the abstract underlying phonology of the language.
- A phonetic representation is the form of a word that is spoken and heard.
-
Phonological rules
Phonological rules map underlying representations onto phonological representations. They delete, insert, or change segments, or change the features of segments.
http://www-01.sil.org/linguistics/glossaryoflinguisticterms/WhatIsAPhonologicalDerivation.htm
"impossible, irreverent, inept" all have different surface manifestations of the underlying /+in/ prefix, all predictable from phonological rules. Again, Occam's razor at work, and possibly, hopefully, a reflection of what a speaker needs to know.
The bottom line in generative linguistics is if your version of the rules of syntax, phonology or semantics needs less memory or fewer steps in a derivation than mine, then you win the game. It's not a matter of debate at all.
Last edited: