Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Vowel harmony one word at a time
Transcript of Vowel harmony one word at a time
Its only constraints work on surface structure. Underlying representations get a 'free ride' There are no Greek words which underlyingly have a geminate consonant (*ekstassi).
This is due to the same constraint which forces degemination (mas sosate > masosate). Suppose that several different inputs I1, I2 . . . , In, when
parsed by a grammar G lead to corresponding outputs O1 ,
O2, . . . , On, all of which are realized as the same phonetic
form F – these inputs are all phonetically equivalent with
respect to G. Now one of these outputs must be the most
harmonic, by virtue of incurring the least significant
violation marks: suppose this optimal one is labelled Ok .
Then the learner should choose, as the underlying form for
F, the input Ik . Soft constraints are too soft for OT Templatic requirement: The first two vowels of Asia Minor Greek are of the same (maximally high) sonority. Ito and Mester: The Stratified Lexicon Selective Lexicon Optimization Many words have adapted to this template. But there have always been exceptions The same is true for sAlthough it is difficult to quantify these things, it looks like especially highly frequent words and old words have adapted. Ito and Mester (1995, 2001, 2002) Native: Template >> Faith
Foreign: Faith >> Template Problems:
* It is not clear how a word can move from one lexical stratum to the next
* We don’t know how many strata there are, or how the child learns about theses trata
* or why words have the tendency to move towards the native stratum.
* Also, it is not clear why rerankings of faithfulness and markedness constraints,
as opposed to other divisions of the constraint set, are involved. variation: phonetic noise selection if a child has to learn /megariz/, it might sometimes misconstruct this as /magariz, megeriz, migariz .../ Diachrony proposes, synchrony disposes Selective Lexicon Optimisation:
In case of conflicting evidence, choose the underlying representation with the lowest violation profile * There is only one phonology.
◮* It is explained how a word can move from one lexical
stratum to the next,
◮* and why words have the tendency to move towards the native stratum
◮* Also, it is clear why faithfulness and markedness constraints, as opposed to other divisions of the constraint set, are involved, since this distinction is relevant for LO as well.
◮* Effects of frequency and ‘age’ can be understood without implementing them in the grammar examples putative
turkish source So, for instance, some varieties of Asia Minor Greek have acquired partial vowel harmony patterns through interference from Turkish (Dawkins 1916:68, and see examples in Thomason & Kaufman 1988:218); but since the relevant varieties of Asia Minor Greek have so many loanwords and even inflectional suxes borrowed from Turkish, the vowel harmony seems most likely to have arisen in Greek via abstraction from the harmonizing borrowed Turkish constructions. Thomason (2001): Will rules be borrowed? Stress-sensitive morphological structure southern greek
vowel assimilation Conclusions of Revithiadou et al. (2006) - There are two domains of 'harmony', one 'Southern Greek' at the left edge, one more 'Turkish' at the right edge - There are also a lot of exceptions and irregular patterns (efaksan, keremitsi). "We may see these forms either as lexical exceptions or as indications that other (diachronic) processes may have interfered."