BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Phonotactics with[awt] rules: the learnability of a simple\, unnat
 ural pattern in English - John Harris\, UCL (in collaboration with Nick Ne
 asom (UCL) and Kevin Tang (Yale))
DTSTART:20160519T153000Z
DTEND:20160519T173000Z
UID:TALK60919@talks.cam.ac.uk
CONTACT:Jamie Douglas
DESCRIPTION:There is a well-established collection of speaker-independent 
 methods for discovering\nphonotactic patterns in languages\, e.g. comparat
 ive reconstruction\, phonological analysis and\ncomputational learning. Th
 ere is also an increasingly varied collection of experimental\nmethods for
  ascertaining how much of this patterning is actually internalised by\nspe
 aker-hearers. In seeking to determine what makes a phonotactic pattern lea
 rnable or not\,\nresearchers have focused on a variety of factors\, includ
 ing phonological regularity\,\nproductivity\, naturalness\, and formal sim
 plicity. Experimental studies have investigated\nvarious permutations of t
 hese factors\, with results that are more or less surprising. For\nexample
 \, speakers have been shown to have internalised and to be able to product
 ively apply\n(a) patterns that are regular\, simple and natural (e.g. wug 
 tests of English *-s*) but also (b)\npatterns that are irregular\, relativ
 ely complex and not synchronically natural\, such as English\nvelar soften
 ing (e.g. Pierrehumbert 2006).\n\nIn this paper\, we examine the English p
 honotactic pattern where consonants following\n/aw/ are restricted to coro
 nals\; hence *tout*\, but not /tawk/\, or /tawp/ (e.g. Halle & Clements\n1
 983). The pattern (‘awT’) is pretty regular\, more so than velar softe
 ning. It is general\, in that\nit affects a large swath of the lexicon. It
  is formally quite simple\, more so than the *-s* pattern.\nAnd it is not 
 natural. It is the synchronically accidental outcome of a series of largel
 y\nunrelated sound changes\; each of the changes might be natural\, but th
 eir cumulative effect is\nnot.\n\nWe report the results of two non-word ju
 dgement experiments designed to test the extent\nto which native speakers 
 of English have tacit knowledge of the awT pattern. In both\nexperiments\,
  listeners were presented with C1-3VC non-word stimuli containing the\ndip
 hthongs /aw/\, /ow/\, /ij/\, followed by a range of consonants\, and were 
 asked to judge how\nEnglish-like they sounded. The non-words were controll
 ed for lexical neighbourhood density\,\nweighted by frequency. In the firs
 t test\, listeners made forced choices between paired words\ndistinguished
  solely by whether the vowel was followed by a coronal versus a non-corona
 l\nconsonant. In the second\, listeners rated individually presented stimu
 li on a scale of\nEnglishness\, drawn from a sample of ~1200 nonwords.\n\n
 The question of whether speakers have implicit knowledge of a given phonot
 actic pattern\ncan be approached in two stages: (a) do they have any tacit
  awareness of the pattern at all and\,\nif so\, (b) is the awareness comme
 nsurate with the pattern being stored as a grammatical rule?\nBroadly spea
 king\, the results of both experiments show weak evidence of an awareness 
 of\nawT but little or no evidence that this reflects grammaticalised knowl
 edge. That is\, to the\nextent that speakers have any tacit inkling of the
  pattern at all\, it is probably not encapsulated\nin anything like a phon
 ologist’s rule or constraint. Where a coronal preference is observable\,
 \nit does not generalise across different manners of articulation\, as wou
 ld be expected if there\nwere a rule-driven bias towards formal simplicity
 . Also\, the preference is influenced by onset\nsize and lexical neighbour
 hood factors\, which suggests subjects were making on-the-fly\njudgements 
 of how much the non-words resemble real words.\n\nWe conclude that awT is 
 a case where phonologists know more about a phonotactic\npattern than spea
 kers know. In the light of our results\, we consider whether this should b
 e\nattributed to the fact that awT is not natural (cf. Hayes & White 2013)
  or to other factors\,\nsuch as that it is not involved in alternations.
LOCATION:GR04\, English Faculty\, 9 West Road (Sidgwick Site)
END:VEVENT
END:VCALENDAR