Location
The workshop takes place in the Tourist hostel on the Szrenica Peak (1362 m), which is found in the Glowny Grzbiet Karkonozy (Main Ridge of the Karkonosze, west of Labski Szczyt). The hostel  has a very interesting history. The Jelenia Góra Valley inhibitants together with the Sudetes Germanes built it in 1922 to protest against the Czech taking away the license from the German hostel owner on the Czech side. In 1968 the hostel burnt down, but in 1982 it was successfully built anew, earning a more impressive appearance that better fits in with surrounding scenery.
    Around the peak there spread some areas ideal for skiing. In 1962 a two-section chair lift was built from Sklarska Poreba to Szrenica. From there also begin three ski trails. Close to the hostel one will have several tourist trails. The black trail leads to the north, in the direction of the Konskie Lby (1290 m). They are a formation of picturesque crags from where one gets a manificant view of the Jelenia Góra Valley, Sklarska Poreba and the Izerskie Heights. Along the Grzbiet Glówny (Main Ridge) the red trail goes east towards Trzy Swinski (1290 m), which consists of a few crags of most charming shapes. Next, the red trail leads one to the west, towards the Hostel in the Hala Szrenicka.

• opening  [mp3] [wma]

Abstracts

This paper provides an explanation for why intervention effects occur in questions (see e.g., Hornstein 1995, Beck 1996, Pesetsky 2000) by employing a new type of control operator and defining characteristics of a dynamic setup for semantics. The approach is applied to data from German, English, Japanese and Korean, and is shown to feed off Pesetsky's (2000) generalisation that the fewer opportunities a language has for phrasal wh-movement, the greater the number of intervention effects. Back

Hans Martin Gaertner
One Cheer for OT

Arguably the only genuine argument for an ot-application to syntax has been presented by vikner (1997). there it is suggested that the interpretation of indefinites in icelandic interacts with a purely morphosyntactic principle of verb positioning. Thus, if no verb remains inside vp, objects can undergo object-shift or remain inside vp. guided by something like the principle "avoid ambiguity", the object-shift position is associated with "strong" readings of the indefinite, the vp-internal position with "weak" readings.  as soon as
verb movement is blocked, object-shift cannot take place and indefinites inside vp acquire both weak and strong readings. unfortunately, subsequent tests with icelandic native speakers have not confirmed the original data. however, the same argument can be replicated in terms of a well-known phenomenon from
tagalog.  there so-called "non-trigger marked" theme-nps are standardly confined to indefinite readings. definite reference requires trigger-marking. however, in relative clauses, trigger-marking obligatorily affects the relativized np for purely morphosyntactic reasons.  this would predict that relativization of a non-theme can only take place if the theme is indefinite. yet, not quite surprisingly, this is not borne out. instead non-trigger marked theme nps in relative clauses are ambiguously definite or indefinite. again a morphosyntactic principle overrides "avoid ambiguity." (a similar phenomenon has actually been reported for cliticization in brabants . . .).  what is interesting about these arguments is the fact that other syntactic frameworks can be shown to either completely fail on these phenomena or to have to appeal to principles non-distinct from ot-principles.  Back

Henk Zeevat
Syntax? Prominence!

Judith Aissen introduces harmonic alignment in syntax to explain different object and subject marking, agreement and passivisation. The speaker wants to be talked out of the idea that syntax can be reduced to multidimensional prominence marking.  Back

Balder ten Cate
Information exchange as reduction towards normal form

Dynamic semantics suggests the following general view on information exchange: by making utterances, people make take some of their private information and make it common ground. Continuing this line of reasoning, we can argue that the process converges to a (hypothetical) situation in which all information is common ground and there is no private information left to communicate. Thus, information exchange can be seen as a process of reduction towards a normal form, where the objects of reduction are (tuples of) information states and the normal form is this hypothetical situation. Consequently, we can apply the theory of
Abstract Reduction Systems, which will allows us to reason about reduction strategies (in our case, strategies of information exchange). In the talk, I will show how this idea can be worked out formally and I will indicate how the approach can contribute to a general theory of pragmatics.   Back

Rosja Mastop
'But' as presupposing defaults and implying greater speaker-utility of the second conjunct

I think that 'but' sometimes presupposes that the conjuncts give rise to conflicting expectations (contra Umbach). It meets all the requirements and has all the properties to fit into the framework of Zeevat (2000). I will also try to work out Umbach's description of obligatory use of 'but' in the following terms: with 'but' the speaker changes the topic and if he/she is rational, utility optimizing, the new topic will be more useful to him/her. I will argue that the observation that the second conjunct in concessive but is somehow 'stronger' than the first one, is an implicature rather than a presupposition.  Back

John Duda and Darrin Hindsill
Bidirectional OT and constraints on functional and computable communication

We will present a number of puzzles(and possibly a solution or two) arising from the application of the framework of bidirectional OT to problems involving grammatical variation and change, rational cooperation in communication, and the notion of functionalism in grammar.    Back

Reinhard Blutner
Partitioning scales and relevance

Recently, C. Kennedy pointed out that adjectives associated with open scales tend to have context-sensitive standards (an interesting book) whereas adjectives associated with closed scales tend to have the endpoints as “standards” (an empty bottle). In order to fully explain the relation between scale structure and standards, the principles underlying the relation must be determined. Closed scale adjectives are perfectly compatible with variable, context sensitive standards; the fact that they make use of fixed endpoint standards instead therefore suggests the influence of some deeper principles of linguistic / cognitive organization. I want to suggest that van Rooy’s recent explication of relevance in information-theoretic can be applied, and the principle of optimal relevance is one of the explanatory principles wished-for.    Back

Jason Mattausch
Bidirectional OT for Discourse

We will present a generative strategy for NP selection and subject choice in discourse.  The constraint-based analysis relies jointly on the bidirectional Optimality Theory proposed in Blutner [Blutner 2000] and on the harmonic alignment of markedness scales proposed in Aissen [2000].   In the process, we will present a problem for the bidirectional Optimality Theory.  The problem we present arises in the generation of discourses when the proscriptions of each of these two procedures – one related to the economy of production, the other to interpretational accuracy – are in direct conflict with one another and where, as a result of this conflict, a compromise between the two is necessary for the aforementioned interdependence to be preserved.  We will attempt to give a description of that compromise, and to formulate a version of bidirectionality that reflects that description, arguing that the required compromise is not an equal one, and that bidirectionality ought to be modified to reflect an asymmetry in that interdependence as opposed to the symmetric, mutual reliance defended or assumed in contemporary definitions thereof.