The Weekly Newsletter of MIT Linguistics

Phonology Circle 10/21 - Anthony Brohan, Ezer Razin, and Sam Zukoff

Date/Time: Monday, Oct 21, 5:30pm
Location: 32-D831

Anthony Brohan: A case study in assimilation: The view from PBase

This talk will explore the hypothesis that the directionality of assimilation in a given language may be affected by typical stress location in a language. Data from PBase (Mielke 2008) and StressTyp (Goedmans et. al 1996) are used to develop a model of the characteristic behavior of features, which is then used to probe for directionality biases in languages based on stress systems. Second, a case study of lenition is presented, aiming to sharpen the findings of the stress/assimilation interaction. The “lattice” of leniting changes (Hock 1999) is empirically filled in with patterns from PBase and functional pressures of contrast preservation in lenitions (Gurevich 2004) are explored in this lattice.

Ezer Razin: An evaluation metric for Optimality Theory (joint work with Roni Katzir, Tel Aviv University)

Our goal is to develop an evaluation metric for OT, a criterion for comparing grammars given the data. Using this criterion, the child can try to search through the space of possible grammars, eliminating suboptimal grammars as it proceeds. Our empirical focus is the lexicon and the constraints, and our evaluation metric is based on the principle of Minimum Description Length (MDL). We wish to model aspects of knowledge such as the English-speaking child’s knowledge that the first segment in the word ‘cat’ involves aspiration, that [raiDer] is underlyingly /raiter/, and that [rai:Der] is underlyingly /raider/. We take it that any theory of phonology would require this knowledge to be learned rather than innate, making this a convenient place to start. The learner that we present succeeds in obtaining such knowledge, which, to our knowledge, makes it a first. The generality of the MDL-based evaluation metric allows us to learn additional parts of the grammar without changing our learner. We demonstrate this by learning not just the lexicon and the ranking of the constraints but also the content of the constraints (both markedness and faithfulness constraints) from general constraint schemata. The learner that we present succeeds in obtaining this knowledge, making it a first in this domain as well.

Sam Zukoff: On the Origins of Attic Reduplication

In Ancient Greek, the perfect tense is marked by reduplication. The default pattern of reduplication for consonant-initial roots is to have a CV reduplicant, and the default pattern for vowel-initial forms is to show lengthening of the initial vowel. However, for a subset of (synchronically) vowel-initial roots, there exists a different pattern, known as Attic Reduplication. Attic Reduplication forms have a reduplicant of the shape VC with concomitant lengthening of the root-initial vowel. For example, √ager- ‘gather together’ : perfect ἀγήγερμαι [agɛ̄germai], √eleuth- ‘go, come’: perfect ἐλήλουθα [elɛ̄loutha], √ol- ‘destroy’ : perfect ὄλωλα [olɔ̄la].

In this talk, I will argue that Attic Reduplication is a well-motivated outcome of the regular phonology of a Pre-Greek system that still contained laryngeals, rather than an analogical development or a stipulated alternative pattern.

The account that will be developed here uses independent evidence from the process of “vowel prothesis” and other alternative reduplication patterns, both in Greek and the other Indo-European daughter languages, to demonstrate that the normal CV reduplication pattern was blocked for laryngeal-initial roots due to markedness considerations. In avoiding these markedness violations, an alternative copying pattern emerges. This new pattern turns out to involve reduplicant-internal epenthesis and copying of both the root-initial laryngeal and the second root-consonant. The ranking which ultimately selects this repair is consistent with, and may even directly follow from, the intersection of the independent rankings necessary to generate vowel prothesis and the default reduplication pattern.