24.962  24.962

Just to repeat a couple things that were announced in class this morning:  the final presentations will take place on Tues/Thurs next week (see previous announcement for assignment of dates)

• Please prepare a handout.  (We have generally needed about 12 copies)

• You will have 12 mins to present, and 8 mins for questions and discussion.  This is a very short amount of time to present, so you will need to be very clear and efficient about laying out the problem you are trying to solve, and sketching the analysis so far.  As Donca says, practice is the way to go in making sure that you can say what you hope to in the time allotted.

• You may enjoy the following guidelines for giving short conference presentations, developed by Bruce Fraser and Geoff Pullum for the LSA.  Although some of their rules are more crucial than others, one would not go wrong to follow them all.
  lsadc.org/info/lsa-res-guide.cfm

As always, if you have any questions, please just let us know!

The following snippet of R code was used to assign the final presentation slots in ultra-random fashion:  If you are on line 1, you are presenting on Tues May 12, and if you are on line 2 (starting with [5]), you are presenting on Thurs May 14.  More details about the organization will be announced in class this week.

> sample(c("Sam", "Youngah", "Natasha", "Marie", "Liuda", "Rafael", "Sasha", "Igor"))

[1] "Igor"    "Youngah" "Liuda"   "Sasha"  

[5] "Sam"    "Marie"   "Rafael"  "Natasha"

As you can see, not everybody's desires to present on Thursday could be accommodated-- but feel free to swap days, if you can find someone willing to trade a Thurs slot with you.  Also, if you have any questions about how to read which day you are presenting on, please let me know. 

Finally, be sure to consult with either or both of us as you encounter issues while developing your topic!

The in-class presentations of final projects is scheduled for the last two days of class, May 12 and 14.   If you have a preference about which of these days you would like to present, please let me know-- otherwise, I will assume that everybody's default preference is to present on Thurs (i.e., as late as possible), and I will use a randomized procedure to schedule presentations across the two days.

Yesterday in class we began our discussion of Schwabian by developing an Optimal Paradigms (OP) analysis in which correspondence was evaluated "democratically", and was overruled in the 2sg by a "Realize Morpheme"-type constraint demanding that forms containing phonologically distinct morphemes should be phonologically distinct on the surface (or some variant of this).  That constraint managed to do the trick of motivating a non-uniform paradigm (2sg different), but I agreed too hastily to the claim that this would actually solve the problem.  There was a candidate that we did not consider, in which the entire paradigm changes to [-anterior] `ʃ' except for the potentially homophonous 1sg form.  By my reckoning, this candidate actually wins under the constraint ranking that we discussed in class, reconfirming a point that we discussed concerning the predictions of the OP approach.  We'll discuss this at the start of class tomorrow, but I just wanted to call your attention to it if you are still writing up your discussion.