Jane has a nice sumup of OzPhon07 on her blog. Sounds like it was a great day!
One thing (well many things, but one thing to blog about) caught my eye:
Nasalisation was a good example of this range. Hywel Stoakes showed how Bininj Gun-Wok contrasts with English in having little anticipatory nasalisation on vowels preceding nasals (Nasal co-articulation in Bininj Gun-Wok: an aero-dynamic study). This is a low-level alternation that most of us would not be aware of. But it may go some way to explaining the presence of prestopping of nasals – which he suggested helps keep cues for place of articulation.
I have two comments on this: Firstly, Yan-nhaŋu has no anticipatory nasalisation in most contexts, but it does in the environment VN’C[-voice]: that is, the vowel before a nasal-glottal-stop cluster often has nasalisation. I don’t have aerodynamic data for this, just my ears, but I’m pretty sure about it. I’d want to see more data about the environments tested and about the number of speakers, since in my experience with Yan-nhaŋu, this is a variable.
Secondly, I’ve wondered about the mis-timing gesture and cue preservation stories about prenasalisation and how they relate to the prestopped laterals that are found in Karnic and Thura-Yura languages (to the south-east of Arandic). Clearly the velum-lowering mistiming story doesn’t apply in the case of laterals, but it’s not immediately clear to me that the cueing story works either. Often prestopping only affects apical laterals, for example. The only reason we’d want to look for a different explanation of prestopped laterals is if they are assumed to be ‘mimicry’ of the prestopped laterals. The prestopping in Karnic has a weird distribution, with different segments affected in different languages. I don’t see how a purely phonetic explanation would capture that.