<div dir="ltr"><div><div><div><div><div><div>Dear Emily,<br><br>thank you for your reaction to my report.<br><br></div>In principle, I consider the scenario possible (but see below) but I think it does not affect the data just collected through your list.<br>
<br></div>As far as I can tell, the discussion of this subtle aspect of the superiority effect has been confined to GB/OT circles, and there are only VERY few discussions of it at all. Given that the frameworks usually ignore each other's work, I doubt that many of the HPSG list members have read through Lasnik & Saito (just like most GB people are not familiar with the recent HPSG literature). So, ideally at least, the judgments have come from experienced syntacticians who have NOT considered the contrast before. Judgments are currently also being collected from phonologists and semanticists, who are even less likely to have read the recent GB/OT/minimalist stuff.<br>
<br></div>I think the judgments of linguists (as a group) are different because they are people who are used to processing complex sentences, who have an affection for language, who are able to ignore side effects, etc. etc., properties that we will probably also find among laypersons, but with much less frequency. <br>
<br></div>In general, I think we know that repeated exposure has an effect on judgments, but the really clear cases seem to all involve the transfer of a pattern from one dialect to the other. <br><br></div>Thanks,<br></div>
Gisbert <br><div><div><div><div><br><br>
<br><br><div><div><div><br><br><br><br><div class="gmail_extra"><br><br><div class="gmail_quote">On Tue, May 28, 2013 at 4:33 PM, Emily M. Bender <span dir="ltr"><<a href="mailto:ebender@uw.edu" target="_blank">ebender@uw.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear Gisbert,<br>
<br>
If I understand you correctly, you are asserting that the claim about<br>
superiority<br>
violation mitigation is a "linguistic fact", except that it cannot be<br>
shown using<br>
judgments from non-linguists only.<br>
<br>
The more likely explanation (it seems to me) is that the claim is in<br>
fact spurious,<br>
but one that linguists working in syntax have been trained to see in the data<br>
through exposure to the relevant type of sentences with *s on them.<br>
<br>
Emily<br>
<div><div><br>
On Tue, May 28, 2013 at 7:12 AM, Gisbert Fanselow<br>
<<a href="mailto:gisbert.fanselow@gmail.com" target="_blank">gisbert.fanselow@gmail.com</a>> wrote:<br>
> Dear HPSG-List members,<br>
><br>
> thanks for the numerous replies to my data query.<br>
><br>
> The claim has been made in the GB/minimailst literature that the superiority<br>
> violation in the lower clause of<br>
><br>
> who wonders what who bought<br>
><br>
> is licensed (mitigated) only with an interpretation of the question in which<br>
> the lower _who_ takes matrix scope, i.e. it is licensed only as a question<br>
> that would be answered as<br>
><br>
> John wonders what Jane bought, Mary wonders what Bill bought ...<br>
><br>
> In two experiments with non-linguists, we could not confirm this claim.<br>
><br>
> However, among the members of the HPSG list who responded (some 20), there<br>
> were three or four native speakers who showed a shift of preferences in the<br>
> direction predicted by the claim cited about, i.e. they preferred answer a)<br>
><br>
> "John does"<br>
><br>
> for the non-superiority violating question<br>
><br>
> who wonders who bought what<br>
><br>
> but the more complex answer to the question involving crossing movement in<br>
> the lower clause. The reverse preference shift does not occur.<br>
><br>
> I will have to sort out various non-native replies, so that I can see<br>
> whether there are enough native judgments left for drawing a firm<br>
> conclusion, but I am convinced our small survey shows that<br>
><br>
> a) a subtle claim concerning readings of multiple wh's made in community A<br>
> can be confrmed by judgments from community B<br>
><br>
> and<br>
><br>
> b) there are linguistic facts which at least the standard method cannot<br>
> establish by consulting non-linguists.<br>
><br>
> Currently, I am checking if the GB/minimalist intuitions are also shared by<br>
> phonologists and semanticists.<br>
><br>
> Thanks A LOT for your help!<br>
><br>
> Gisbert Fanselow<br>
><br>
><br>
><br>
><br>
><br>
><br>
><br>
> --<br>
> Gisbert Fanselow<br>
> Linguistics, University of Potsdam<br>
> Karl-Liebknecht-Strasse 23-24<br>
> 14476 Potsdam<br>
> x331-977 2446<br>
><br>
><br>
><br>
<br>
<br>
<br>
</div></div><span><font color="#888888">--<br>
Emily M. Bender<br>
Associate Professor<br>
Department of Linguistics<br>
Check out CLMS on facebook! <a href="http://www.facebook.com/uwclma" target="_blank">http://www.facebook.com/uwclma</a><br>
</font></span></blockquote></div><br><br clear="all"><br>-- <br>Gisbert Fanselow<br>Linguistics, University of Potsdam<br>Karl-Liebknecht-Strasse 23-24<br>14476 Potsdam<br>x331-977 2446 <br>
</div></div></div></div></div></div></div></div></div>