Publications

CV and List of Publications

 

 

Papers on Server

2011

Atle Grønn & Arnim von Stechow: The Temporal Organisation of Indicative Conditionals

We develop a tensed version of Kratzer's account of indicative conditionals. Much care is devoted to licensing the observed temporal morphology. To theory is based on our system of feature transmission via semantic binding.

Atle Grønn & Arnim von Stechow: Future vs. Present in English and Russian Adjunct clause

The revised version of our 2010 paper. The progress made with respect to the former paper is this. (a) The standard account for Russian embedded relative clauses is that their tense must be deictic, i.e. indipendent of the matrix tense. We show that is not true. When we embedd a relative clause with a future tense under a matrix past, it can have a bound reading; sentences of the type: "At the university, the girl got to know Bill Clinton who later would become her husband". The other progress is that we assume that a TPRO that stand for the "subjective now" in the complement of attitude has the present feature i-n in non-SOT languages. This enables us to get rid of the relative present for non-SOT languages and the last residue case for Ogihara's rule of tense deletion. We are able to analyse each main tense in a temporal adjunct under the assumption that its temporal centre ("the perspective point") is an anaphoric pronount Tpro bound by a higher tense.

2010

Atle Grønn & Arnim von Stechow: Complement Tense in Contrast. The SOT-Parameter in Russian and English.

In an SOT-language like English, ‘past under past’ may have a simultaneous interpretation, i.e., we have temporal agreement. In a non-SOT language like Russian, we only have the shifted interpretation. In English, the temporal morphology of the embedded verb is determined by the matrix tense via a binding chain through verbal quantifiers such as ‘say’ or ‘think’. In Russian, these attitude verbs break the binding chain. The morphology of the embedded verb is determined locally by an embedded relative PRESENT, FUTURE or PAST. We propose that the difference between English and Russian is derived from:

The SOT-parameter

A language L is an SOT-language if and only if the verbal quantifiers of L transmit temporal features.

Verbal quantifiers quantify over times (e.g. fut. will) or world-times (e.g. verba dicendi).

The paper will take up a recent challenge by Daniel Altschuler and Olga Khomitsevich against existing accounts: verbs of perception and, occasionally, factive verbs in Russian may express simultaneity by ‘past under past’. We will show that the problem is in fact non-existent when the complement is imperfective. Concerning factives, however, we argue that the complement tense is an independent de re past. Finally, perception verbs are normally not verbal quantifiers and hence not subject to the SOT-parameter.

Atle Grønn & Arnim von Stechow: Future vs. Present in Russian and English Temporal Adjunct Clauses

1. English has a simultaneous reading in Present adjuncts embedded under will. Russian Present adjuncts under budet or the synthetic perfective future can only have a deictic interpretation. This follows from our SOT parameter.
2. The syntax of Russian temporal adjunct clauses (do/posle togo kak…) shows overt parts that had to be stipulated for English as covert in earlier papers. We present a neat and straightforward analysis of Russian temporal adjuncts. Contrary to Russian, English has Present under Future in before-clauses. It would be nice to relate this feature to the SOT parameter, but the construction seems to be an idiosyncrasy in the domain of subordinate tense.

2009

Tenses in Compositional Semantics. To be published in W. Klein (ed). The Expression Time in Language. To appear. The paper presents a theory of Tense in which the temporal feature tensed forms is checked by semantic Tense via a chain of semantic binding. The paper treats tense in simple sentences (the auxiliary system), the interaction of Tense with temporal adverbs, Tense in subordinate constructions (relative clauses, before/after-clauses, complements of attitudes).

2008

Syntax and Semantics: An Overview. 128 Pages. A corrected draft of a handbook article (corrections from August 16th 2009). The article discusses the syntax of a broad range of interesting syntactic constructions and shows how their LFs are obtained. The LFs are expressions of a typed language with abstraction. The LF-language has a precise interpretation.

Doris Penka & Arnim von Stechow: Phrasal before and after

We treat cases lake John arrived berfor/after Mary. We give a compositional semantics for the adverb, but it requires countercyclical Merge at LF, i.e. the insertion between a moved phrase and the the lambda-abstract created by LF-movement.

2007

.

Schritte zur Satzsemantik I. 143 S. 1,8 MB. Die ersten 9 Kapitel meiner Einführung in die Semantik. (1. Einleitung, 2. WB-Semantik, 3. Tutorium: Mengen und Funktionen, 4. Interpretation einfacher Sätze, 5. Syntax, 6. Typengetriebene Interpretation, 7. Semantik einiger Konjunktionen, 8. Semantische Systeme, 9. Abstaktion und Variablenbindung)

Schritte zu Satzsemantik II, 169 S., 1,9 MB. Weitere 8 Kapitel meiner Einführung in die Semantik (2. Präsuppositionen, 3. Kaplansche Charaktere: Deixis und Tempus, 4. QR in lokalen und temporalen Verbergänzungen, 5. QR von PRO: Quantoren in XPs, 6. QR von PRO: Relativsätze, 7. Dekompositionstheorie I: Kausale Verben, 8. Dekompositionstheorie II: Schöpfungsverben)

Schritte zur Satzsemantik III. 108 S., 1,3 M (1. Bindungstheorie, 2. Eselssätze)

The temporal degree adjectives früh(er)/spät(er) 'early(er)'/'late(r)' and the semantics of the positive.

In: A. Giannakidou and M. Rathert (eds.). Quantification, Definiteness and Nominalization. Oxford University Press. 2009. p. 214-233.

Revised version of Times as Degrees.

 

2006

Arnim von Stechow. Times as Degrees (28 pages)

The paper give a semantics for früh(er) 'early(ier)' and später 'late(r)'. The presuppositions of the phase particles schon 'already'/noch 'still' and schon 'already'/erst 'only' are stated in terms of 'early'/'late'.

 

Penka, Krasikova, von Stechow. Anankastic Conditionals again.

The object of our investigation is expressing necessary conditions in natural language, particularly in a certain kind of conditional sentences, the so-called Anankastic Conditionals (ACs) , a topic brought into the linguistic discussion by the seminal papers (Sæbø, 1986) and (Sæbø, 2001). A typical AC is the following sentence, Sæbø’s standard example:
(1) If you want to go to Harlem, you have to take the A train.
Sæbø analyses the sentence by means of the modal theory in (Kratzer, 1981), according to which a modal has two contextual parameters, a modal base f(w) and an ordering source g(w). The modal base contains relevant facts and the ordering source contains an ideal like wishes, moral laws and the like. Normally, the antecedent of a necessity-conditional is added to the modal base. Sæbø’s new proposal for the analysis of the AC is that the antecedent without the information ‘you want’, called inner antecedent, is added to the ordering source.
Sæbø’s analysis had remained almost unnoticed in the literature for more than a decade. But recently, quite a number of semanticists have discussed his theory. Every alternative account contains one or other material modification of Sæbø’s theory.
Our proposal will be this. The inner antecedent is not added to the ordering source. It rather is the antecedent of a Lewisonian necessity-conditional. Equivalently, it can be added to a circumstantial modal base that contains all the facts compatible with the antecedent. Furthermore, the construction is analyzed as a conditional speech act: the sentence expresses an assertion in a context in which you want to go to Harlem.
The structure of the paper is as follows. In section 2 we will have a closer look at the data, section 3 will be a review of the literature on ACs and in section 4 we will present a counterfactual account.

 


2005

 

Sigrid Beck & Arnim von Stechow. Dog After Dog Revisited (version January 2006)

The paper is about the meaning of certain pluriactional adverbs like "ein Hund nach dem anderen", "über einander", "dog after dog", "stückweise". It incorporates these adverbs into Beck's theory of the relational plural. The adverbs restrict the cover variable.

Semantisches und Morphologisches Tempus: Zur Temporalen Orientierung von Modalen (Version August 2005)

Dieser Aufsatz gibt den Stoff der Vorlesungen wieder, die ich im August/September 2004 in Kyoto gehalten habe. Ich untersuche vor allem die temporale Orientierung von Modalen. Es gibt zwei Möglichkeiten der Analyse: (a) unter Modale wird in LF optional ein kovertes Futur eingesetzt um die futurische Orientierung zu erhalten; (b) Modale betten grundsätzlich eine "uneigentliche Zukunft" ein, also ein Intervall, dass mit der Referenzzeit beginnt und sich in eine unbestimmte Zukunft erstreckt. Man muss dann Temporaladverben intersektiv deuten: sie schneiden aus dem Intervall ein Stück heraus. Das Futur "wird" wird am besten als zweistelliges Modal interpretiert. Dies hat den Vorteil, dass man das Tempus im Nebensatz semantisch ignorieren kann.Analoges gilt für kontrafaktischeh Konditionalsätze. Man gelangt so zu einem System, dass neueren Vorschlägen von Condoravdi ähnelt. Das Papier wird in der japanischen Zeitschrift "Neue Beiträge zur germanistischen Linguistik" publiziert. Bd.4, Heft 2, 2005

 

 

Arnim von Stechow, Sveta Krasikova, & Doris Penka. Anankastic conditionals (version of April 2, 2005)

We propose a new analysis for anankastic conditionals. In anankastic conditionals, the truth of the consequent is a necessary condition for the truth of the antecedent:

a. You have to take the A train if you want to go to Harlem.

As first discussed in Sæbø (1986), (a) is truth conditionally equivalent to the following sentences:

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.
The hardest problem is a compositional analysis of (a): what is the role played by "want" in the antecedent? Sæbø (1986) and recent proposals by von Fintel & Iatridou (2004) and Huitink (2004) assume that "want" indicates that the ordering source of the modal is teleologic. But these approaches are problematic. Our proposal starts from the observation that (a) is elliptical. Its overt forms is (a'):

a'. If you want to go to Harlem, you have to take the A train to go to Harlem.

The consequent in (a') doesn't contain "want" and is in fact (c). We propose an analysis for (c) along the lines of Lewis' (1973). Sentence (c) is true iff the nearest worlds where you will go to Harlem are worlds where you take the A train.

The want-antecedent in (a) adds a felicity condtion: the sentence can be uttered appropriately only in contexts whose conversational background is compatible with the proposition that you want to go to Harlem. Compared to competing modal approaches the analysis is more precise because it says what the modal base contains, viz. the facts that are true in the best antecedent worlds.

The paper has problems with overgeneration. Why is the following sentence odd?

d. For kangaroos to have no tails, they have to topple over.

Problems of this sort are not overcome, but all the problems that have been discussed in the literature about anankastic conditionals.are solved.


2004

Roumyana Pancheva & Arnim von Stechow. On the Present Perfect Puzzle (15 pages. pdf 144KB)

This is the text of our NELS34 talk. We derive the present perfect puzzle by a new theory of semantic competition. The Perfect means the same in languages such as English and German. It denotes a time span that has no part located after the speech time but may be before the speech time or include it. The Present is different in English and German. In English, it denotes the speech time, in German a time not before the speech time. The Perfect operator is located at the place of the participle at d-structure. At LF it moves to the perfect auxiliary (if there is one). If the auxiliary is finite and located in the tense node, the complex tense Present + Perfect competes with the more specific Past in English. The result is that the Present Perfect has an Extended Now reading in this configuration and semantically excludes temporal frame adverbs specifying a past time. No such competition arises in German or for the Pluiperfect. In certain other configurations of English, e.g., non-finite Perfects, the competition doesn't arise either.

Arnim von Stechow & Ede Thomas Zimmermann. A Problem for a Compositional Treatment of De Re-Attitudes.

Abstract

Arnim von Stechow & Doris Penka: Too and enough

I can't remember at which occasion this talk was presented. We analyse too as a comparatve with a "can"-complement and eough as an equative with a "must"-complement. We compare the analysis with that of Cecile Meier.

 

2003

February 2003: Feature Deletion under Semantic Binding (25 pages, pdf, 256KB) This is the actual text of my NELS33 talk. It covers the same material as my 2002-paper Binding by Verbs, but it is shorter and contains, hopefully, less errors and typos. The paper concentrates on temporal adverbs and the sequence of tenses in different languages.

May 2003: Postcript to Feature Deletion under Semantic Binding (5 pages, pdf 136KB) After I had delivered the paper (Stechow, 2003), a colleague wrote to me that the system outlined was virtually identical with (Kratzer, 1998), and that this article had not been cited. The longer version (Stechow, 2002 (to appear)) quotes (Kratzer, 1998), and the reference has been deleted by my automatic bibliography program when I rewrote and shortened the paper. I am sorry for that.The paper contains some remarks on the relation between (Stechow, 2003) and (Kratzer, 1998).

June 2003: Alexiadou, Rathert, von Stechow: The modules of Perfect constructions (pdf, 412KB)

Introduction to the book Perfect investigations. To be published by de Gruyter.

July 2003: Paslawska-Stechow: Tempus-Aspekt-Architektur aus tplologischer Sicht

2002

Eva Gerö & Arnim von Stechow. Tense in Time: The Greek Perfect (229 KB)
This is a revision of a paper written in 1999. It contains a diachronic analysis of Greek tense and aspect. We give a detailed account of the tense/aspect architecture of Greek and follow its changes through history. The synthetic Perfect has a resultative interpretation from the beginning on, but we argue that it has an Extended Now interpretation as well from the classic period on. We try to prove this point by a detailed analysis of pertinent examples.

Temporal Prepositional Phrases with Quantifiers (pdf.240KB)
I comment on a recent paper by Pratt & Francez in L&P. I treat sentences like John called every Monday and John was in Boston today. I accept P&F's idea that temporal nouns contain a temporal variable which is restricted by the tense of the clause. On the other hand, the temporal variable belonging to the verb is introduced by an aspectual relation like PERFECTIVE, PERFECT or PROGRESSIVE. This variable is bound by the tense as well. The main idea of the paper is that the objects of temporal PPs are quantfied in and may take different scopes. Languages like German show typical inverse linking configurations on the surface. I argue that sentence such as Wolfgang hat in keiner Woche an jedem Trag Tennis gespielt are generated from the deep structure an jedem Tag in keiner Woche Wolfgang Tennis gespielt hat by Scrambling plus Pied-Piping, in other words, the paper contains a detailed semantics and syntax of cascades of temporal quantifiers. In addition, the paper contains a rather elaborated account of the archtecture of tense and aspect . (In the previous version of the download, the name Francez was consistenty mispelled as Franzes. I am very sorry for that.)

Alla Paslawska & Arnim von Stechow: Perfect Readings in Russian (54 pages,pdf 320KB)
This is a revision of a paper from 1999. The main idea is that the so-called perfective morphology in Slavonic languages does two things: it selects for a "telic" actionsart (accomplishment/achievement) and it licenses an aspectual relation, viz. INCLUSION of the event-time in the reference time or PRECEDENCE of the event time with respect to the even time. Aktionsarten are what has been called situation aspect and aspectual relations are what has been called viewpoint aspect (C. Smith). The former relation is what Klein (94) calls PERFECTIVE, the latter is what he calls PERFECT. The thesis therefore is that there is no unique meaning of perfective morphology. The so-called imperfective morphology is a privative opposition with respect to the perfective morphology and has many uses. So, a fortiori there is no unique meaning for the so-called imperfective.This is a reconstruction of Forsyth's (1970) views.

The paper gives a compositional semantics of telic versus atetlic verbs. It discusses the interaction of temporal adverbs and the semantic aspects.

The disscussion is influcenced in a crucial way by Schoorlemmer (1995). We give a detailed review of Schoorlemmer's (1995) theory of the tense-aspect relation in Russian. We show that certain shortcomings of Schoorlemmer's theory stem from Reichenbach's theory of tense, which treats times as points and not as intervals. Interval semantics is crucial for getting the facts of Russian right. The paper contains a syntax semantics for the Russian past passive participles, which reconstructs Schoorlemmer's insights and seeks to overcome some deficiencies of her system.

German seit "since" and the ambiguity of the German Perfect (pdf 392 KB)

There is a curious division of labour between the semantics of tenses and temporal adverbs of duration. In languages such as English, Swedish or Danish we say such things as Mary has lived in Amsterdam since 1990 by means of the Present Perfect, whereas in German we express this most naturally by a Present statement: Marie wohnt sei 1990 in Amsterdam. Like many others I say that the English Present Perfect non-compositionally expresses an Extended Now (XN) and the perfect level adverbial since 1990 specifies the beginning of that XN. German is different. The durative adverbial seit 1990 introduces an XN starting with 1990 and ending with the tense time, and the VP is predicated thereof. In English, the reference time of the sentence is the XN, in German it is the speech time. The German Pluperfect (and the Infinitve Perfect) behaves exactly like its English counterpart, but the Present Perfect does not. Whereas the English and the Scandinavian languages have that puzzling XN-reading, the German Perfect is peculiar in so far as it can can both express the simple past and have a resultative reading as well. The paper, which is published in the Festschrift for Dieter Wunderlich, gives many empirical examplesfrom real life.

Perfect of Result (pdf. 88 KB)

This is a section of a rvesion of the previous paper. I contains a revised version of the section "Perfect of Result". The former, and in fact printed version, contained some typos and the theory of result modification did not work, a fact discovered by Daniel Büring. I have been trying to overcome the shortcomings.(April 9, 2003).I try to overcome a difficulty of Kratzer's theory of the Perfect of Result, viz. the inner modification of adjectival passives by functional adverbs such as wieder/again. This is achieved by itroducing a furter argument for telic verbs, viz. a property of states.

Binding by Verbs: Tense, Person and Mood under Attitudes (49 pages, pdf 240KB)

This is the text unerlying the NELS talk in november 2002. The present text was presented at a DGFS-meeting in Mannheim and published in a volume of the conference. See my list of publications under "Biographisches".

Abstract:

The paper presents an integrated theory of tense, pronouns and mood under attitudes on the basis of the theory in (Schlenker, 1999), which will be elaborated and improved in certain details. The centre of the paper will be the analysis of certain temporal phenomena that have been worrying me for the last ten years. The coverage of the theory is more general, however. The central ideas of the approach are these.
1 . Each finite verb has a world, a time and a person argument, whose semantic properties are checked by the person, tense and mood morphology of the verb. Call them the egocentric arguments of the verb.
2. With the possible exception of the third person, the said arguments of the verb are variables. Variables with features will always be interpreted with respect to the context of utterance. The result is that variables with these features are deictic expressions.
3. Features of semantically bound variables are deleted at LF and therefore not interpreted (Heim’s principle of feature transmission; see e.g. (Heim, 2001)).
4. Attitudes (modals and certain temporal auxiliaries) may bind the egocentric variables of their verbal complement under agreement and therefore may produce bound person/tense/mood readings. Under these circumstances the variables lose their referential character entirely.
That certain verbs (as opposed to noun phrases) can bind variables is crucial for understanding the functioning of de se pronouns (e.g. PRO, logophors, Engl. I under attitudes), sequence of tense rules, certain readings of temporal adverbs such as in two days, shiftable deictic modals such as ought or should.
The talk tries to clarify the following point: while (Kaplan, 1979) is wrong in his claim that pronouns such as I or you are always directly referential, Schlenker’s (1999) criticism of Kaplan’s Prohibition against Monsters is not warranted. Schlenker’s attitudes are arguably purely intensional operators. There is no expression of natural language that denotes a genuine monster such as Stalnaker’s diagonal operator.
B. Data. Here are some of the data that will be discussed in the talk:
(1) Amharic I under says (Schlenker, 2001)
Why can the English sentence John says that he is a hero be expressed as John says that I am a hero in Amharic?
(2) Russian Present under thought (folklore)
Why is Engl. John thought that Mary was sick pronounced as John thought that Mary is sick in Russian?
( 3) in 10 minutes under promise : How do we explain the following contrast?
a. #Ich traf Irene um 5. Sie rief mich in genau 10 Minuten an. (German)
‘I met Irene at 5. She called me in exactly 10 minutes’
b. Irene versprach mir, mich in 10 genau Minuten anzurufen.
‘Irene promised me to call me up in exactly 10 minutes’
(4) ought under thought: What about this?
a. *When John was a schoolboy, he ought to study more. (Abusch, 1993)
b. When John was a schoolboy, he thought that he ought to study more.
(5) Germ. Konjunktiv under thought: Subjunctive in than-clause generates contradiction
a. Ich dachte, ihre Yacht sei/wäre (subj.) länger als sie ist/war (ind.).
(Russell, 1905) (Stechow, 1984)
I thought your yacht be-SUBJ longer than it be-IND.
c. #Ich dachte, ihre Yacht sei/wäre (subj.) länger als sie sei/wäre (subj).
I thought your yacht be-SUBJ longer than it be-SUBJ.
C. LF-movement. The LF of the de se reading of ( 6a) will be ( 6b), where the features of the variables bound by thought are deleted at LF under agreement with the features of thought.
( 6) a. I thought I was a hero.
b. w1ind tpast x1st [VP thought*ind*past*1st* l<w2ind t2past x21st> [CP w2ind t2past x21st was a hero]]
The paper addresses the problem of how the relation between thought and the variables it binds can be conceived of as LF-movement. There are several possible theories. The lambdas at the left edge of CP are operators moved at LF. The relation between thought and the bound variables would then be a sort of generalised chain in the GB-style; cf. (Percus and Sauerland, 2002). Another possible option is to literally move the verbal quantifier out of an argument position of the subordinate verb; cf. (Heim, 2001). Both variants lead to an area of LF that has not been investigated much so far.

 2001

Temporally Opaque Arguments in Verbs of Creation (pdf 268KB)

Verbs of creation (create, make, paint) are not transparent. The object created does not exist during the event time but only thereafter. We may call this type of opacity temporal opacity. I is to be distinguished from modal opacity, which is found in verbs like owe or seek. (Dowty, 1979) offers two analyses of creation verbs. One analysis predicts that no object of the sort created exists before the time of the creation. The other analysis says that the object exists throughout the act of creation. I investigate three theories: Theory I says that no object of the sort created and which is caused by the very act of creation exists before the creation. In this theory, verbs of creation must embed a property. Theory II can regard the indefinite object of a creation verb as a quantifier and gives it wide scope with respect to the verb. The theory has to make sure that the objects quantified over exist only after the event. While Theory I and II start from the assumption that the extension of all nouns depend on time, Theory III says that Individual Level predicates do not depend on time. This ontology will enable us to treat verbs of creation as first order relations. The theory will entail that a picture does not mean the same as there is a picture. The paper discusses various approaches to the problem: Krifka, Parsons, Landman, Kratzer and Zucchi.

Doris Penka & Arnim von Stechow: Negative Indefinita unter Modalen
Ausgehend von Daten zu Bechs Kohäsion wird eine Theorie der negativen Indefinita im Deutschen vorgestellt. Sie werden als ganz spezielle NPIs analysiert, nämlich als solche, die im unmittelbaren Skopus einer abstrakten Negation stehen. Die stärksten Argumente für eine solche Analyse kommen von N-Wörtern in Idiomen (Maria kann so auf keinen grünen Zweig kommen). Das Objekt kann als Bestandteil des Idioms nicht QRt werden. Trotzdem hat die Negation weiten Skopus über das Modal. Damit ist LF-Bewegung von N-Phrasen nach SpecNeg, wie sie von Haegemann und Zanuttini angenommen wird (NEG-Kriterium), aus semantischen Gründen für das Deutsche (und auch für slawische Sprachen) nicht haltbar. Das Papier gibt eine präzise Syntax und Semantik für einschlägige Konstruktionen. Es steht im Einklang mit Blaszczaks Behandlung der Negation im Slawischen und geht sowohl auf die Wurzeln der NPI-Konstruktion in der Sprachgeschichte des Deutschen als auch auf Negative Concord imBairischen ein. Mögliche Gegenbeispiele gegen die Analyse (it. Non ha detto quasi niente/ dt. er hat fast nichts gesagt) werden diskutiert, aber nicht endgültig gelöst.

How are Results Represented and Modified? Remarks on Jäger & Blutner's Anti-Decomposition (pdf.280IKB)
The paper is the contritubution to the Oslo Workshop on adverbs (October 99) and investigates a recent proposal to resultativity by G. Jäger and R. Blutner (J & B). J & B say that the representation of result states of accomplishments by means of CAUSE and BECOME is not correct and should not be done in the syntax in terms of decomposition. They develop an axiomatic approach where each accomplishment/achievement is related to its result by a particular axiom. Modification of the result by "again" makes use of these axioms and the restitutive/resultative ambiguity is a matter of lexical ambiguity or polysemy. They argue that the classical decomposition theory cannot treat the restitutive reading of A Delaware settled in New Jersey again (there had been Delawares in New Jersey but not this particular one; and those earlier Delawares never moved to New Jersey but were borne there). I discuss (and dispute) these data and compare the two theories. J & B's contains an OT-part dealing with the disambiguatiing role of stress. While the decomposition theory cannot deal with the datat mentioned, it can integrate the OT-part of J & B's theory.
The paper was corrected in October 2001, and section 5 was completely rewritten. There I give an overview about my actual views about the representation of result states. There are two kinds of telic VPs. One kind has a syntactically accessible result state; these VPs are relations between events and properties of times. The properties are the result states and can be modified by adverbs like noch immer "still" , 2 Stunden lang "for 2 hours" or wieder "again" . For adjectival passives we will have a modification of the result state. Other verbs don't have accessible result states. Their adjectival passives only have a repetitiven reading if they are modified by wieder, and the modification by noch immer and 2 Stunden lang is not possible at all. There is cross-speaker variation.

2000

Sauerland-Stechow: Syntax-Semantics Interface (pdf 188KB)
This small handbook article tells you about the most important things that you should know about the interaction of syntax and semantics. It is a good starter for beginners.

von Stechow. Some remarks on choice-functions.

1999

Alla Paslawska/A. von Stechow: Perfekt und Futurkonstruktionen im Ukrainischen (und Russischen) (pdf 472KB, 60 Seiten)Das Ukrainische ist eine der wenigen slavischen Sprachen, die ein Perfekt und ein Plusquamperfekt haben. Die Konstruktion wir im Rahmen einer Extended Now-Theorie im Sinn von McCoard analysiert. Das System fasst Perfekt als ein subordiniertes Tempus (Tempus2) auf, und erst unterhalb der Perfektkopfs ist der eigentlich "view point" Aspekt angesiedelt.

Irene Rapp/A. von Stechow: Fast "Almost" and the Visibility Parameter for Functional Adverbs (pdf 240KB)
Adverbs can be classified as to their capacity to look into lexical decomposition ("D-adverbs"). The paper provides a detailed investigation of German fast "almost". Resultative readings, which exist for English, are not found in German. The Visibility Parameter gives a systematic explanation of the crosslinguistic contrast. The paper has appeared in the Journal of Semantics in 2001.

1999/2002: Partial Wh-Movement, Scope Marking, and Transparent Logical Form (pdf 300KB)
The Bausteine Syntaktischen Wissens by von Stechow & Sternefeld (1988) contains one of the first systematic descriptions that accounts for most of the the German data exhibiting partial wh-movement. The paper is written in English and makes the result accessible to the Linguists who do not read German. The article was written in 1999. The article was published in 2001.

Remarks on Kamp and Reyles Analysis of the English Perfect (pdf 196KB)
Kamp and Reyle try to reduce the English Present Perfect to a resultative reading where the Present Perfect denotes the result state of an event. I show the difficulties that arise when it comes to an interpretation of sentences like Mary has lived in Paris for five years and Mary has ben to Paris since 1998. I argue that an analysis in terms of an Extended Now in the sense of McCoard and Dowty is the better one.

1998

German Participles II in Distributed Morphphology (pdf 240KB)
This is my contribution to the Bergamo conference held in July 1998. There are 4 participle II constructions in German: the passive, the predicative construction, the HAVE-perfect construction and the attributive construction. All these have different properties when the interact with temporal
locating adverbials ("vor zwei Stunden") or measure adverbials ("in zwei Stunden").
I defend the thesis that the participle II morphology carries no meaning in itself. The different meanings are induced by the auxiliaries and the syntax, which introduces abstract heads with an own semantics. The adverbs can scope with respect to these heads. The most complicated construction is the attributive one. I can have a resultative meaning or a perfect passive meaning. Both interpretations involve a rather different syntax.
The theory is formulated in the framework of Distributed Morphology

Eine erweiterte Extended Now-Theorie für Perfekt und Futur (ps 958 KB)
Ich gebe ein kompositionale Semantik für das Perfekt und Futur des Deutschen an, die auf der Extented Now (XN) Theorie McCoards basiert. Das Perfekt liefert ein XNP(r)-Intervall, dessen rechte Grenze die Referenzzeit gerade einschließt. Das Futur liefert das Spiegelbild XNF(r) dazu. Präsens und Präteritum werden als temporale Pronomina gedeutet, die im Komplementsatz abgebunden werden. Das XN-Intervall wird nicht konjunktiv mit der VP verbunden, sondern durch ein Quantifikationsadverb (z.B. immer oder ein unsichtbares Adverb), deses Restriktion es bildet. Die Auxiliare haben und werden sind also semantisch keine Funktoren, sondern Argumente. Perfekt und Futur sind keine Aspekte. Aspekte sind immer unter ihnen, direkt über der VP angesiedelt. Das Papier enthält eine kompositionale Semantik für das Auxiliarsystem von Chomskys Syntactic Structures und weist die breite Anwendbarkeit der Theorie auf verschiedene Sprachen hin.

1996

The different readings of germ. "wieder" (again). A structural account. (pdf 280KB)
This is a preprint version of the paper about German wieder "again", which was paublished 1996 in the Journal of Semantics. It argues that we have to locate decomposition in the syntax in order to deal with scope ambiguities arising with wieder . German syntax reveals certain structural facts not visible in English.

Stechow-Geuder: Morphologische Kategorien und ihre Interpretation (pdf 136 KB)
Im Wesentlichen geht es um Syntax, Semantik und Typologie der Negation. Das Papier verteidigt meine seit vielen Jahren vertretene Auffassung, daß deutsche negative Indefinita in den meisten Fällen negative Polaritätselemente sein müssen, welche keinerlei Negation beinhalten sondern nur Indefinita sind. Der Vortrag gibt eine Übersicht über die Typologie der Negation in verschiedenen Sprachen mitsamt deren Deutung.

Against LF-Pied Piping

Discussion of Nishigauchi's theory of LF-Pied Piping. Paper published in NALS.

1995

Claudia Nohl & Arnim von Stechow: Die Interpretation syntaktischer Strukturen (155 Seiten, pdf 400KB)Vorlesunngsmanuskript zur Einführung in die Semantik

Tense in Intensional Contexts: Two Semantic Accounts of Abusch's Theory of Tense (pdf 172KB)
This is a formal reconstruction of an earlier version of Abusch's account of the Sequence of Tense. The paper dates from 94, is rather long (and tedious) and covers many different topics not treated by Abusch.

On the Proper Treatment of Tense (pdf 116KB)
This is the 1995 SALT-paper. It builds on the former paper and contains my theory of zero tenses. It is easier to read than the former paper and covers similar topics. The main idea is that the temporal argument of complements of attitudes is l-bound, i.e. abstracted away. There could be no relation between a time "denoted" by a time argument of the embedded main predicate can have any relation to the speech time. It is related to the "subjective now". Suppose, Mary calls me at 6, but she thinks that it is four o'clock. She gives me a call and says:"I will meet you at 5". I can tell Wolfgang about this call by the sentence: "At six Mary told me that would meet me at 5". The point of examples like these is that there cannot be deictic tenses in complement of attitudes. In the paper I speak of zero tense.

1995/2000: Some remarks on choice functions and LF-movement ( pdf 232KB)

This paper was given at the Konstanz conference on choice function in 1995 (?) it has been published in the mean time. I discuss some possible restictions for the distance of the existential quantifier that binds the choice function variable. The paper exhibits a rather skeptic attitutude towards the explanatory power of choice functions. Most arguments that have been given for choice functions are not really stringent.

1991

1991: Intensionale Semantik - eingeführt anhand der Temporalität: 227 S. (pdf 488KB)

Temporale Konstruktionen werden in verschiedenen formalen Sprachen analysiert. Es handelt sich um intensionale und um extensionale Sprachen. Die Monographie enthält einen ersten Entwurf von Strukturen, die später transparente LFs genannt worden sind. Die Spezialität der hier entwickelten Tempussemantik ist, daß TemporaQuantifikatiionsadverben einschränken können.

Focussing an Backgrounding operators, 22 pages.

(thanks to Floris Roelofsen for the scan)

1990

Layered Traces

Talk given at Revfülöp, Hungary.

 

1984

Comparing Semantic Theories of Comparison (pdf 8.3 MB)

My Reaction to Cresswell, Hellan, Hoeksema and Seuren

 

1982

Structured Propositions (pdf 440KB)
This is one of the first papers in which I introduced the use of structured propositions into semantics. It has never been printed but has been circulating as a report of the Konstanz SFB. Many people have asked me to make it accessible. Udo Klein has brought the old manuscipt to an internet version by scanning it,, correcting it etc. Thanks.
The paper treats the de re- and de se-attitudes. It combines Lewis's method of de se-attitudes with a theory of structured propositons. Furthermore it deals with apparent paradoxes of Binding Theory in these contexts.

Wolfgang Klein & Arnim von Stechow: Intonation und Bedeutung von Fokus (pdf 47 MB)

Dieses nie veröffentlichte SFB-Papier enthält eine ausführliche Diskussion von Informationsverläufen deutscher Sätze, die maschinell aufgezeichnet sind. Dazu gibt es eine explizite Semantik, die mit eine einer Fokus-Hintergrund Semantik arbeitet, also einer Version von strukturierten Propositionen.

Structured Propositions and Essential Indexicals

This is the talk given on the 4th Amsterdam Colloquium. Essential indexicals (e.g. de se pronouns) are analysed es egocentric descriptions in the theory of structured propositions. The paper builds on David Lewis' Attitudes de se. Again this is an early paper introducing structured propositions into semantics.

 

 

Festschrift

These are the preprints of the contributions to the Festschrift on the occasion of my 60th's birthday.