<html><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"><meta http-equiv="Content-Type" content="text/html; charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">Im Rahmen meines Proseminars aus Grammatiktheorie und kognitiver Sprachwissenschaft findet am 9.12.2019 ein Gastvortrag von Thomas McFadden (ZAS) & Sandhya Sundaresan (Leipzig) statt, zu dem Interessierte herzlich eingeladen sind.<div class=""><br class=""></div><div class=""><div class="">Zeit: 15:15 - 16:45</div><div class="">Ort: Seminarraum 2, Sensengasse 3a</div><div class=""><br class=""></div><div class=""><b class="">Da dieser Raum eventuell nicht genug Platz bietet, ersuche ich um kurze unverbindliche Rückmeldung derjenigen, die zum Vortrag kommen wollen (Änderungen sind natürlich möglich).</b></div><div class=""><br class=""></div><div class=""><div class=""><br class=""><i class="">Deriving Selective Opacity for Adjuncts vs. Complements via Path-based Locality <br class=""></i>Thomas McFadden (ZAS) & Sandhya Sundaresan (Leipzig), based on joint work with Hedde Zeijlstra (Göttingen)<br class=""><br class="">Adjuncts have long been considered syntactic islands (Ross 1967). Prior approaches to adjunct/argument asymmetries have involved according adjuncts some special primitive status: e.g. claiming that they can sideward A-move (Drummond & Hornstein, 2014) or are late-merged (Lebeaux, 1991; Fox, 2002; Abe, 2018) or are phasal (Chomsky, 2008; Mueller, 2010). <br class=""><br class="">In this talk, we start with the core empirical insight that selective opacity effects involving adjuncts and complements involve a systematic directionality restriction: complements are transparent to syntactic dependencies both into and out of them; in contrast, adjuncts are (selectively) opaque only to dependencies from above (e.g. long-distance phi-agreement on a host verb reflecting the phi-features of an argument inside the adjunct) but transparently allow syntactic dependencies out of them (e.g. adjunct control). In Minimalist terms, probing out of an adjunct is allowed, while probing into it is disallowed; in contrast, probing out of, and into, complements is allowed. <br class=""><br class="">To accommodate such asymmetries, we will extend a particular approach to structure building (Merge) and labeling which unifies insights from Minimalism and Categorial Grammar. On the strength of this, we redefine Agree, the core Minimalist operation driving syntactic dependencies, in terms of sisterhood + path-based locality (a notion that hasn't found much currency in current Minimalism but has parallels in other frameworks like HPSG/LFG, CCG and TAG, as well as in early versions of GB). This is the idea that two elements X and Y (in this case, a Probe and Goal) are syntactically visible to each other iff they are connected by an uninterrupted sequence of steps, each of which satisfies the same (syntactic) condition. We will show that this theoretical machinery can accurately account for selective opacity patterns with adjuncts, including certain island effects. A potential concern, which we will briefly discuss, is that the model might be too powerful as it stands, undergenerating in particular the so-called Truswell effects --- licit cases of movement out of adjuncts under particular semantic conditions. A further non-trivial advantage of this model is that it allows us to significantly reduce the need for domain-based locality units such as the phase. <br class="">~~~~~ </div></div><div class=""><br class=""></div><div class=""><i class="">Erinnerung: Gastvortrag von Zheng Shen morgen, 25.11.2019, 15:15, Seminarraum 2!</i></div><div class=""><br class=""></div><div class=""><div class="">Ich freue mich auf Euer Erscheinen!</div><div class=""><br class=""></div></div><div class="">—Susanne Wurmbrand. </div></div><br class="">
<br class=""></body></html>