Handouts

Handouts

2009

Course given in Vienna, January 2009

Subordinate Tense: The SOT-Parameter (with Atle Grønn)

Talk given in Oslo on September 18th. We say that a language is an SOT language if and only if its verbal quantifiers transmit temporal features under binding. We investigate tense in relative clause, in attitude complements and in temporal adjuncts (before/after-clauses) in English, Russian, and Japanese.

Phrasal before/after (with Doris Penka)

Talk given in Oslo on September 23th. ("John came before/after Mary"). The construction has interesting consequences for the LF-interface. We need a parasytic scope device for the movement of the the corelate. In addition the befor/after-adjunct has to be inserted counter-cyclically.

Temporal interpretation of counterfactual conditionals (with Atle Grønn)

We argue that would conditionals are semantically Present. would-have conditionals are ambiguous between semantic Present (with Perfect in the consequent) and semantic Present + Perfect out-scoping WOULD. We investigate English, German, Norwegian, Ancient Greek, Latin, Italian, French. We use the theory that transmits temporal features under binding. The feature "subjunctive2" plays an important role: when uPast and uSubj2 are transmitted to the auxiliary "have", it is not interpreted.

The (Non-)Interpetation of Subordinate Tense (with Atle Grønn)

Talk given at the University of Göttingen on November 9th. Contains material of the Oslo talk on subordinate tense. New: attitudes in German and the role of the subjunctive; semantic analyse of perception verbs and the explanation of simultaneous readings in Past und Past constructions in Russian.

2008

4 Lectures in Degree Semantics

Paris, École Normale Superieure, September 2008

Tense, Modals and Attitudes as Verbal Quantifiers

ESSLLI, Hamburg, August 2008

Tense and Presuppositons in Counterfactual Conditionals.

Joint work with Atle Grønn. Handout of the talke given at NYU, April 18 2008.

We are interested in the temporal structure and presupposition of counterfactual conditionals.
1. Tense-less counterfactuals have the following structure: WOULD [if-clause] [prejacent].
2. WOULD in itself tense-less. It can be embedded under semantic PRES or PAST.
3. WOULD is a modal of temporal control. It follows that the past morphology in the antecedent is semantically void.
4. The presuppositions of counterfactuals are evaluated in the counterfactual worlds. They carry over to the real world at the branching point by Lewis’ principle of “no gratuitous change of the past”.
5. We deal with Ippolito’s challenge concerning presupposition projection.
6. So-called “two-past” counterfactuals are ambiguous between PRES WOULD [PERF A] [C] and PAST WOULD [A] [C].
7. The temporal properties of counterfactuals are explained by the method of i/u-features used for the analysis of Sequence of Tense Phenomena and Negative concord.
9. We analyse counterfactuals in a number of languages (English, German, Norwegian, French, Russian).
10. We don’t discuss previous work in this talk.

2007

Syntactic and Lexical Causativization: CAUSE and BECOME again

Handout of the talk given at CONSOL XV in Brussels.

• There is decomposition in the syntax for some inchoative and causative verbs like aufwachen/aufwecken 'wake up' itr/tr, öffnen / sich öffnen 'open'

• There is no decomposition in the syntax in resultative constructions involving small clauses, goal PPs. These are treated by means of a lexical operation CAUSE*, called Principle R in (von Stechow, 1995) . It depends on the formulation of that principle whether control is involved or not.

• The reason for decomposition comes from different readings generated by the adverb wieder 'again'. In my approach this adverb expresses pure repetition and the different readings are accounted alone in terms of scope.

• A recent important challenge comes from Jäger & Blutner; cf. (Jäger and Blutner, 2000, Jäger and Blutner, 2003) . They argue that decomposition in the syntax is not possible in causative constructions that involve subject or object control. Otherwise, the so-called unspecific restitutive reading (UR) cannot be generated. J & B conclude that there is no decomposition in the syntax at all.

• If face the criticism. J&B's counter example can easily be explained. More serious problems arise with construction involving secondary predication. Appropriate formulations of Principle R can deal with most of these.

• The talk presents a theory of decomposition triggering. Decomposition in the syntax is triggered by a stative verb that has the uninterpretable feature [u-CAUSE], which must checked against the interpretable feature [i-CAUSE]. Result phrases have the u-feature as well. They check against a lexically causative verb for type reasons. I am relying on the theory of features in (Zeijlstra, 2004) . Secondary predicates have the feature [u-CAUSE] as well and trigger lexical causativization.

• The discussion is based on an explicit compositional semantics for all examples discussed.

Reflexive Constructions

Talk given in Berlin (Semantic Network) in June 07. I treat ordinary reflexives (1), inherent reflexive (2), anticausative reflexives (3), reflexive middles (4) and reflexive passives (5):

(1) Rossi rasiert sich : r Lx V(x, sich-x)

(2) Rossi freut sich : r Lx V(sich-x)

(3) Die Türe öffnet sich t Lx BEC (O(sich-x)

(4) Das Buch verkauft sich : b Lx V(MAN, sich-x)

(5) La casa si vende : c Lx V(MAN, sich-x)

2006

Spatial Prepositions in Interval semantics

Handout of the talk given on September 20 at the Semantics Network Workshop in Barcelona. The account is not yet finished.

In this talk I develop the semantics for spatial prepositions for a great deal of their uses. The main difference to Cresswell is the extensive use of events and their paths in the semantics of the verb. Cresswell insists that we can do a great deal semantics without having events. I believe that we need events for many purposes and we should therefore formulate the semantics for spatial prepositions in terms of events. The approach assumes a number of small operations (here localisers), which accommodate spatial prepositions to different uses. This strategy is similar to the one found in (Beaver and Condoravdi, 2004) . At least this aspect of my analysis is novel.

Types of iF/uF-agreement

This is the handout of my talk at the ESSLI workshop "Concord Phenomena at the Syntax-Semantics Interface", Malaga 7 - 11th of August

The basic idea is always that a morphological feature is uninterpreted (an uF) and point at a semantic operator (an iF). This relation is called agreement or concord. This approach provides the key for a compositional semantics for a wide range of constructions that have been puzzling for many years. I discuss:

phi-features: person, gender, number; negative concord, simple and complex tenses, SOT, Aspect, Comparative, CAUSE-decomposition, Subjunctive in Spanish, Modal Concord in Counterfactuals

Eine Semantik für minimalistische Phasen-Modelle (15. Juli 2006)

Diese Handout enthält eine Ausarbeitung des Appendix zu meinen Handout des GGS-Vortrags vom Mai 2005 zur semantischen Interpretation eines Phasenmodells im Sinne Chomskys oder G.Müllers. Ich habe damals behauptet, dass eine strikte phasenweise Interpretation von Phasen auf LF und PF nicht möglich ist wegen langer Bindung von Pronomina bzw. morphologischen Kongruenz auf PF. W. Sternefeld hat damals eingewandt, dass das doch möglich ist, wenn man die Methode Bennett/Sternefeld zur Variablenbindung benutzt. Ich habe damals gesagt, dass so ein System kompliziert wird und dass man sich bei jeder Phase die bereits eingeführten freien Variablen merken muss. Das alte Handout hat die einschlägigen Methoden nur wenig ausgearbeitet. Ich habe das Handout nun ergänzt und die Kompositionsprinzipien präzisiert ("hybride Kompositionsprinzipien"). Das neue Ergebnis steht also im Widerspruch zu den anfänglichen Behauptungen, die ich aber im Handout gelassen habe. Die Revision ist mein augenblicklicher Reflexionsstand.

Das Ergebnis ist dieses:

Das System wird formal nicht wesentlich komplizierter
Man muss sich bei jeder Phase die neue eingeführten Variblen merken + die Bedeutung der Phase
Jede Abstraktion wirft eine Variable aus dem Speicher
Ansonsten ist die Semantik strikt lokal

Zeiten als Grade: Die Semantik von früh(er)/spät(er)

Times as Degres: the semantics of earli(er)/late(r)

Vortrag in Leipzig am 14. Juli 2006

Abstrakt

Es gibt eine reiche Literatur über before/after aber fast nichts zu früher/später. Die Adjektive haben eine Reihe von interessanten Eigenschaften, die neues Licht auf die Semantik von Gradadjektiven werfen. Wie zu erwarten, lizensieren sie NPIs im Komplement.

1 Alla kam früher/später zur Arbeit als jemals zuvor.

Die Semantik ist offensichtlich verbunden mit der Bedeutung der Phasenadverbien schon/noch und schon/erst . Da ist zunächst das

Typ-1 Puzzle : Warum besteht der folgende Kontrast?

2 a. Es ist schon spät

2 b. *Es ist schon früh

Dann gibt es das Typ-2 Puzzle : Wie erklärt sich der folgende Umschlag der Implikatur?

3 a. Alla kam erst um 6 Uhr. (spät!)

3 b. Es ist erst 6 Uhr. (früh!)

4. a. Alla kam schon um 6 Uhr. (early!)

4 b. Es war schon 6 Uhr. (late!)

Ich entwickle eine Semantik für früh/spät welche die Probleme löst. Die Semantik wird Gebrauch machen von einer kontextuellen Restriktion der Zeitskala T auf ein kontextuell relevantes Intervall I, für das es einen neutralen Bereich N(I) gibt. Der Positiv wird als Universalquantor gedeutet werden, der durch N(I) beschränkt ist. Als Nebenprodukt ergibt sich eine Semantik für die Phasenpartikel schon/noch und schon/erst welche bisherige Vorschläge in der Literatur Hegelsch aufhebt.

LFs für Ereignissemantik (Handout zu Beck-von Stechow "Events in Semantics")

2005

Temporal comparatives: früher 'earlier'/später 'later'

Talk presented at the workshops 'Tense and Mood', Stuttgart, December 3 2005 and "QP structure, Nominalizations , and the role of DP", Saarbrücken, December 16/17 2005.

There is a rich literatur about before/after, but I am not aware of a comparable literature about earlier/later . The talk develops a degree semantics for the antonyms früh 'early' and spät 'late' and derives the meanings of the comparative forms for these adjectives. I defend Heim's negation theory of antonymy. I define a new positive operator Pos that applies to the positive and the negative pole of an antonym pair without any ad hoc stipulation. Each context determines a delineation interval L that separates the negative from the positive pole of an antonymy pair. Pos applied to a set of degrees D says that D holds of every degree in the interval L. A number of non trivial constructions are analysed. The presuppositions of the particles erst/schon can be described in terms of 'earlier'/'later'.

Sigrid Beck & Arnim von Stechow: Pluriactional adverbials

Talk presented at Sinn und Bedeutung 10, Berlin, October 13 2005

±interpretable features, tense ±interpreted

Talk given in Stuttgart, June 9 2005

• The features person, number, gender, negation, tense, aspect and possibly others have interpretable and uninterpretable counterparts (i-F, u-F).
• Stem: A u-feature is licensed by an interpretable feature via agreement (SS = S0)
LF: Features are deleted.
PF: i-features and u-features of overt material are pronounced.
• In this talk I will focus on tense. Tense inflection is semantically vacuous and licensed by agreement. Separate the morphology from the semantic operator.
• SOT-phenomena exhibit multiple agreement (MA). Other temporal constructions, especially conditionals, require MA as well.
• Complication:Certain environments license the insertion of covert tense, FUT and PERF, which have no morphological reflex.
• Agreement is constrained syntactically. There are different ways of implementing that.

Semantic Licensing of Some Uninterpretable Features

Talk given in at the University of Tromsø, May 26 2005

Complements of Emotive Factives

Talk given at Semantic Network, Berlin May 2005

LF in einem Phasenmodell

Vortrag auf der GGS in Tübingen, Mai 2005

Temporal Orientation of Modals and Attitudes (and covert temporal operators)

Talk to be delivered at Cornell University, April 4, 2005

1. The temporal orientation of modals and attitudes are different: lexically, attitudes may have a “present” or a “future” orientation. Modals only have a “present” orientation. Attitudes embed two-place temporal properties. Modals embed one-place temporal properties.
2. An observed future orientation of a modal is completely determined by the Aktionsart of its complement. It should be analysed by the interpolation of a covert relative future (FUT).
3. Temporal and modal subordination is best analysed by assuming a feature theory in the style of the Minimalist Program (MP), (Chomsky, 1995), that distinguishes between interpretable features [i-F] and corresponding uninterpretable feature[u-F]. Semantic tenses and aspects will have an i-feature, the morphology will encode a u-feature. u-features are checked by i-features under c-command. If an i-feature checks more than one u-feature, we speak of Multiple Agree (MA) (Hiraiwa, 2005). The difference between Sequence of Tense (SOT) and Non-SOT languages is described by the MA-parameter: SOT-languages have MA, Non-SOT-languages don’t.
The first part of the talk develops the theory. The second part says something about the literature ((Kratzer, 1998), (Abusch, 2004), (Ogihara, 1989), (Condoravdi, 2002).

Anankastic conditionals (talk delivered at Cornell, on April 5)

Anankastic conditionals are exhibited by the following pattern (Sæbø, 1986). (Bech, 1955/57)

(1) a. You have to take the A train if you want to go to Harlem.
b. If you don’t take the A train you can’t go to Harlem.
c. To go to Harlem you have to take the A train.
To solve the problem discussed in the literature we make two steps: First: the construction is elliptical. The if-clause provides the material for a to-cause belonging to the consequent. The if clause in itself delivers an appropriateness condition for the use of the consequent.

(2) a. If you want to go to Harlem, you have to take the A train to do this.
b. If you want to go to Harlem, you have to take the A train to go to Harlem.

The consequent has the following truth condition:

(3) The nearest worlds where you go to Harlem, are contained in the worlds where you take the A train.

The proposal solves the problems that have been discussed in the literature.

2004

Modalitaet Wien (pdf 756KB)

Handout des Kurses, den ich im Mai an der Uni Wien gehalten habe. Neu gegenüber dem Material in der Semantikvorlesung sind die Überlegungen zum Modalverb genügen.

Arnim von Stechow, Sveta Krasikova & Doris Penka: The Meaning of German um zu: Necessary Condition and enough/too (pdf. 324KB)

The talk was given at the Tübingen workshop on modal verbs and modality, March 25 2004. The first part investigates Sæbø's theory of anankastc modals as in Müller muss nach Hamburg fahren, um mit Schmidt zu verhandeln 'To negotiate with Schmidt, Müller has to travel to Hamburg'. Contra Sæbø we argue that the proposition expressed by the purpose clause is not added to the ordering source but rather to the modal base. We discuss alternative proposals by von Fintel & Iatridou, Spector and Ceplova. The second part discusses the theory of too/enough-constructions by Cecile Meier. We defend the view that “enough” is the equative with a “must”-complement, and “too” is the comparative with a “can”-complement.

Covert Future?

Handout of the talk given in Paris 8, October 9 2004. I treat the possible future orientation of infinitves under modals and counterfactuals. Can they be described by an covert relative Future operator under modal? Furthermore, I explore the consequences of Graham Katz's treatment of present and future oriented verbs of attitutudes. There are comments on Condoravdi's treatment of the temporal behaviour of modals.

2002

2002: Binding by Verbs: Tense, Person and Mood under Attitudes: Abstract(pdf 44KB)

This is the abstract of my NELS talk that will be given on November 10th in Boston. The article upon which the abstract rests will be avaible as a download assoon as the English will have been corrected by a native speaker.

2001

2001: The Janus Face of Aspect (pdf.188KB)
This is the talk given in Milano on October 5 of 2001. I distinguish between morphological tense/semantic TENSE, morphological aspect, ASPECTUALRELATIONS and ASPECTUAL CLASSES. Much of the confusion in found in aspectology comes from the fact that these notions are not kept apart. For instance, the Russian Perfective selects a quantized VP, the aspectual class, but it licenses two aspectual relations, viz. INCLUSION ("PERFECTIVE") of the event time in the reference time ore ANTERIORITY of the event time with respect to the reference time. The Imperfective, a privative opposition, is compatible with every aspectual class and every aspectual relation, at least in principle. I present a rought aspectual typology for several languages. A special section of the talk is devoted to the representation of resultant state. I argue that "lexically accessible" resultant states must be properties of time and verbs that have them, are relations between events and properties. The adjectival passive is an aspectual operator manipulating both the reference time and the aspectual class.

2001: Schlenker's Monsters (pdf. 32KB)

Talk given at UCLA on April 4, 2001 in the Philosophy department. I restate Philippe Schlenker's theory of attitudes and indexicals in a Kaplanian framework. To hold an attitude toward a character is means being in the de-se attitude with respect to the diagonal proposition obtained from the character. It follows that all Kaplanian demonstratives are abstracted away in these contexts. There are three kinds of pronouns: logophoric pronouns remain in situ and are shifed in attitude contexts. Kaplanian demonstratives are non-shiftable and underly a sort of Principle C saying that they cannot be bound. One has to give them widest scope. At third class of pronouns are shiftable with respect to an attitude operator, but they need not be shifted. These pronouns may have narrow or wide scope with respect to an attitude.

1999
1999: Semantics and Syntax of Perfect and Future Constructions (pdf 112KB)

This is a 58 pages handout of my course given at the LOT summer school at Potsdam in July 1999. The manuscript covers the interaction of temporal adverbs with tense, perfect and aspect mainly for German. Description of of the program.

1998

Some Remarks on Ch. Kennedy's Syntax and Semantics of Comparison (pdf 96KB)

Handout of a talk presented at the colloquium on the comparative in Berlin, 27. - 29. November 1998.

1992

1992/93: LFs für Fragesätze, WS92/93 (pdf 112KB)

LFs für Fragesätze: This is the handout of the seminar on interrogatives given in the winter semester 1992/93. It is quoted in the literature and people sometimes ask me for a copy. Therefore I make it accessible. One of the features is the ambiguity between alternative and yes/no-questions is explained by a difference in scope. We get the alternative question by considering the or-clause as the interrogative quantifier which is moved to SpecC. The nucleus of the question contains a bare propositional variable. Some of the stuff is published in my paper "Against LF pied-piping" and in the handbook article "Die Aufgaben der Syntax".
Reconstruction and Logical Form (pdf 116KB)

1992: Reconstruction and Logical Form

This is an early paper on semantic methods of reconstruction, the talk was given in December 92 at Stuttgart. The paper contains a method of quantifying into questions. We apply a variable to the question meaning which converts it into typ t. Then we do the quantfying in. Then we lambda-abstract over the variable.