Tomorrow I will be giving at a talk to the Iconicity Focus Group here at the MPI.
Here are the pertinent details:
Tuesday, December 6th, 2016
12:00 – 13:00
Recasting the magic bullet: Testing the limits of the effectiveness of iconicity for language learning
In this talk I will focus on outlining reasons to be more conservative in our expectations for the explanatory power of iconicity in langauge evolution and acquisition. I begin by discussing a pair of studies which demonstrate that some learning benefits previously attributed to iconicity are actually largely explainable in terms of a benefit for systematicity. Following from this, I explore the general idea that in addition to learnability benefits for iconic elements themselves, the acquisition of iconic tokens can bootstrap the acquisition of the rest of language. To do this, I explore a simple (and admittedly straw-man) version of the sound-symbolic bootstrapping hypothesis, making the argument that just because one thing (learning iconic tokens) happens before another (learning arbitrary tokens) does not mean that anything we can meaningfully call bootstrapping has taken place.
I’m really looking forward to giving this talk. It’s data from my PhD., so a few folks will already be familiar with the work, but I haven’t presented it before in this way, especially to a group of experts about iconicity, so I’m very much looking forward to hopefully getting lots of good feedback. For what it’s worth, my choice to present my data with this type of argument comes from discussions following Marcus Perlman’s talk to the LEvInSON group last month.
Slides and comments on the talk forthcoming later this week!