The 29th GLOW Colloquium was held April 6-9 in Barcelona, preceded by a day of workshops on April 5. The following is a brief report on Workshop 2: Approaches to phonological opacity — and I hope someone who attended all of Workshop 3: Prosodic phrasing, or the one day of phonology talks at GLOW on April 6, will follow suit.
(Unfortunately, I don’t have time to do more than briefly summarize and provide links here, so sorry, no commentary.)
The organizers of this workshop were Eulàlia Bonet (UAB) and Maria-Rosa Lloret (U. of Barcelona), who also organized a great “phonologists’ dinner” for the evening of April 6. The description of the workshop begins like this:
Over the years, opacity has been a major theme in phonological research, but, despite the attention received, the issue is far from being settled.
(Note: in some browsers, such as Safari for the Mac, some parts of the GLOW website don’t display correctly. The workshop description just cited is one of those parts, as is the program for the colloquium.)
As the description later clarifies — and something that became abundantly clear over the course of the workshop — there are actually several “issues” surrounding opacity that are far from being settled, not just that “strictly parallel OT has a problem with (some of) it”. In bulleted question form, these are the issues identified in the workshop description:
- What is the theoretical cost of various devices introduced to account for opacity in parallel approaches? (To which I might add: how are they distinguished empirically?)
- What are the implications of opacity for learnability?
- What are the implications of opacity for language typology?
- Do opaque interactions exist between processes operating within a single morphophonological level (as these are defined in Lexical Phonology approaches)?
- Do opaque interactions exist in the synchronic phonology of languages at all?
Another question raised by Marc van Oostendorp in the discussion period after the talks is: what is the function of opacity? (In his talk, Marc presented a hypothesis about at least one of the functions of opacity.)
There were eight presenters in the workshop.
My talk was first, and I spoke about a novel kind of opaque generalization that I call counterfactual derivation (the abstract is here, the handout is here; it was based on this paper, which was itself based on this phonoloblog post).
The second presentation was by Gabriel Poliquin, who spoke about opaque interactions in Canadian French vowel harmony (abstract here; the work available on his website revolves around the same topic).
The third presentation was by Marc van Oostendorp, who spoke about examples in which opaque phonological generalizations can be attributed to a drive to keep morphological generalizations (in the form of constituent structure) transparent (abstract here, .pdf of presentation slides here).
The sixth presentation was by Andrew Nevins, who presented joint work with Maximiliano Guimaraes about experiments they conducted with Brazilian Portuguese speakers using language games, which they claim to reveal much about the reality, opacity, and analysis of nasal harmony in BP (abstract here, manuscript here).
Finally, the eighth presentation was by invited commentator Ricardo Bermúdez-Otero, who spoke on ‘Phonological domains and opacity effects: a new look at continuancy and voicing in Catalan’ (no abstract, but the handout is here, based in part on this earlier paper).
There was an hour or so of open discussion afterwards, moderated by Ricardo Bermúdez-Otero. The following day, the first invited speaker of GLOW was John McCarthy, whose talk was also very relevant to the topic of opacity. (A handout from another version of John’s talk was recently posted on ROA.)