the week of march 3

scalar implicatures

We looked at scalar implicatures and their derivation via Horn scales and a process of Gricean reasoning that crucially involves the maxim of Quantity. On this approach, the inference from "Lucie ate some of the cookies" to "Lucie ate some, but not all, of the cookies" involves the scale <all, some> and thus a comparison between "Lucie ate some of the cookies" and the stronger statement "Lucie ate all of the cookies". If, by uttering "Lucie ate some of the cookies", the speaker can be assumed to have made the strongest true and relevant statement that she was able to make, the hearer can in turn be expected to conclude that the speaker didn't have sufficient evidence for the stronger claim that Lucie ate all of the cookies. In contexts where it can moreover be assumed that the speaker knows of all alternatives under consideration whether or not they are true, the hearer can be expected to draw the stronger conclusion that Lucie didn't eat all of the cookies. The assumption that the speaker knows of all alternatives under consideration whether or not they are true is crucial for the derivation of the implicature. If the hearer couldn't make that additional assumption, she could only conclude that the speaker didn't have sufficient evidence for claiming that Lucie didn't eat all of the cookies. The additional assumption is plausible for a wide range of contexts where a speaker is justified to utter "Lucie ate some of the cookies" without violating Quality. It would be warranted, for example, if it could be presumed that the speaker had actually seen Lucie eat the cookies. Here is the relevant handout on generalized conversational implicatures, which includes the materials for our discussion of scalar implicatures and Horn scales.

another lab session on experimental design

Florian gave a summary of his comments on your work from the previous lab exercise. Then you did a new lab exercise that focused on critically reviewing experimental designs, as well as thinking about how they can be improved. Florian's comments on your work can be found here.