Scheer – Why modularity predicts that the relationship between phonetics and phonology is arbitrary

Download the handout for this course.

This course exposes the take of the standard theory of Cognitive Science, Modularity, on computational systems and the way they communicate. This general setting is then applied to computational systems that are specialized in treating language, i.e. the items of the inverted T that make the standard generative architecture of grammar since Aspects: morphology/syntax (where concatenation of pieces extracted from long term memory is operated), semantics (where a meaning of the output of concatenation is calculated) and phonology (where its pronunciation is determined).

The course focuses on the comparison between the upper interface of phonology (with morpho-syntax) and its lower interface (with phonetics). The workings of Modularity are the same for all computational systems and interfaces. Hence having two substantially distinct mechanisms running the upper and the lower interface is not an option. Since we know more about the upper interface, its properties and workings will be applied to the lower interface. One central, though perfectly trivial and uncontroversial aspect of the upper interface is the arbitrariness between the item that is spelled out and its phonological exponent: there is no reason why for example the morpho-syntactic structure (subset of the tree) that represents “past tense” in English is represented by ‑ed rather than, say, by ‑et, ‑p, ‑a etc. Any exponent would do, provided that there is one.

If both interfaces have the same workings, the lower one must also be arbitrary in kind, i.e. relate the item that is spelled out (say, a set of melodic primes A, B, C that is linked to a coda) to a phonetic exponent that may be anything and its reverse and is not predictable in any way from A, B, C and their syllabic affiliation.

The detail of the spell-out mechanism will be discussed, as well as different options of its implementation within the realm of Modularity (e.g. lexical vs. computational translation of one vocabulary set into another).

Finally, consequences of the arbitrariness enforced by Modularity are evaluated for the setup of phonology and the traditional understanding of the relationship between phonological primes and their phonetic realization. Diachronic considerations will be prominent here, based on so-called crazy rules (ones that make no sense phonetically) and Bach & Hrams’ analysis thereof: rules are not born crazy but become crazy when they age: the items that they manipulate (A → B / C)  may be subject to independent (typically across-the-board) changes (e.g. tƒs > s in the history of English, which made velar softening k → tƒs / __i strange/crazy). This approach is shown to be able to provide an answer to the following obvious question: if the interface setup is arbitrary in kind, why is the overwhelming majority of phonological patterns found in natural language well-behaved (they make phonetic sense), rather than crazy? This is a major difference with the upper interface, where all spell-out relationships are “crazy” by definition.



1. Modularity in general

Segal, Gabriel 1996. The modularity of theory of mind. Theories of Theories of Mind, edited by P. Carruthers & P. Smith, 141-157. Cambridge: CUP.

Gerrans, Philip 2002. Modularity reconsidered. Language and Communication 22: 259-268.

Scheer, Tobias 2012. Direct Interface and One-Channel Translation. A Non-Diacritic Theory of the Morphosyntax-Phonology Interface. Vol.2 of A Lateral Theory of phonology. Berlin: de Gruyter.
==> general workings of Modularity: chapter “Modularity and its consequence, translation” §§29ff.
==> lexical vs. computational translation: chapter “Just one Channel…”, §§160ff.
2. Lower interface with spell-out

in general:
Scheer, Tobias 2014. Spell-Out, Post-Phonological. Crossing Phonetics-Phonology Lines, edited by Eugeniusz Cyran & Jolanta Szpyra-Kozlowska, 255-275. Newcastle upon Tyne: Cambridge Scholars.

applied to laryngeal phonology:
Cyran, Eugeniusz 2011. Laryngeal realism and laryngeal relativism: two voicing systems in Polish? Studies in Polish Linguistics 6: 45-80.
==> sections 1 to 3.
3. Crazy rules:

Bach, Emmon & R. T. Harms 1972. How do languages get crazy rules? Linguistic change and generative theory, edited by Robert Stockwell & Ronald Macaulay, 1-21. Bloomington: Indiana University Press.

Scheer, Tobias 2015. How diachronic is synchronic grammar? Crazy rules, regularity and naturalness. The Handbook of Historical Phonology, edited by Patrick Honeybone & Joseph C. Salmons, 313-336. Oxford: OUP.