Faust – What’s phonological in phonologically-conditioned allomorphy: an introduction to the phonology-morphology interface

In Russian, the nominative singular form of the word ‘city’ is [górət], but the nominative plural form is [gəradá]. The same information, namely ‘city’, is thus realized in two different ways, stressed [górət] and unstressed [gərad-] ([-a] denotes plurality). This state-of-affairs is called “allomorphy”. It is encountered in most if not all languages and comes in many flavors. In this course, we will investigate the phenomenon of phonologically-conditioned allomorphy.  This investigation will take us from the basic generative assumptions about storage and production, through the different takes on the workings of the different types of allomorphy in Generative Phonology and outside it, and to a set of cutting-edge questions yet to have been resolved. One question will accompany us throughout the course: is phonological allomorph-selection performed by the same module that is responsible for phonological computation?
The 1st class will set the stage by stating and motivating the basic assumptions: underlying vs. surface representations, autosegmental phonology, phonological optimization, the preference for uniformity of phonological representation and the place of phonology and morphology in the grammar. In light of all this, the limits of allomorphy will be set. The 2nd and 3rd classes will discuss competing takes on phonologically-conditioned allomorphy. We will ask whether phonology does indeed choose between underlying representations, or maybe there is only one “abstract” underlying representation that is realized as it can be. By stripping the proposals from their specific formalizations, we will try to clarify what the real differences are between them, and explore their predictions. The fourth class will explore the limits that a modular view places on allomorphy. We will compare the predictions of a view that states that selection precedes phonology to one that claims that it is part of phonology. The fifth class will discuss the notions of weak and strong suppletion, which have been claimed to be cognitively identical, and present positive evidence to the contrary. This class will also summarize and conclude the course.


Nevins, Andrew. 2011. “Phonologically conditioned allmorph selection”. In Marc van Oostendorp et al. The Blackwell companion to phonology 4. Oxford: Blackwell. pp. 2357-2382.
Paster, Mary. 2015. Phonologically conditioned suppletive allomorphy: Cross-linguistic results and theoretical consequences. In Eulàlia Bonet, Maria-Rosa Lloret, and Joan Mascaró, (eds.) Understanding Allomorphy: Perspectives from OT. Advances in Optimality Theory series. London: Equinox.
Scheer, Tobias. 2016. Melody-free syntax and phonologically conditioned allomorphy. To appear in Morphology (available online).