- 教官: Hana Filip
Genericity
Dieses Seminar widmet sich der Bedeutung und dem Ausdruck der generischen Sätze wie (i) Hunde bellen, Eine Orange enthält im Schnitt 30mg Vitamin C, Silke fährt (gewöhnlich) mit dem Rad zur Schule (charakterisierenden Sätze), und (ii) Die Kartoffel wurde bereits 7000 Jahre v. Ch. von den Inkas in Südamerika angebaut (Artenreferenz). Charakterisierende Sätze besagen etwas über, oder `charakterisieren', eine Art und auch ihre individuellen Mitglieder, und wir bewerten sie als wahr, auch wenn nicht alle Mitglieder die relevante charakterisierende Eigenschaft besitzen. Z.b. Hunde bellen ist wahr, auch wenn es Hunde oder Hunderassen (Basenjis, u.a.) gibt, die nicht bellen; dies sind also Ausnahmen zu der charakterisierenden Eigenschaft von bellen, die auf die Mehrheit der Hunde zutrifft. Allerdings ist das `Mehrheit'-Kriterium nicht das richtige Kriterium, um die Ausnahmen, die charakterisierende Sätze `tolerieren' in Griff zu bekommen. Es gibt auch charakterisierende Sätze wie Haie greifen Menschen an, die wahr sind auch wenn die charakterisierende Eigenschaft von `Menschen angreifen' für die überwiegende Mehrheit von Haien nicht zutrifft. Wie kann man also die Ausnahme-tolerierende Eigenschaft von charakterisierenden Sätzen richtig erfassen? Mit solchen und ähnlichen Fragen, die charakterisierende Sätze aufwerfen, werden wir uns beschäftigen. Generische Sätze erlauben uns über Regularitäten in der Welt zu sprechen, d.h., dwas wir als `normal', `erwartet', `üblich' halten, und daher geben sie uns einen wichtigen Einblick in die Art und Weise, wie wir die Sachverhalte in der Welt um uns verstehen, kategorisieren, und schliesslich Entscheidungen über unser Handeln treffen, usw. Sie reflektieren nicht nur Fakten, die unbestreitbar sind (z.B. Die Erde dreht sich um die Sonne), sondern auch Vorurteile (Die Briten sagen immer, dass Deutsche keinen Humor haben) und falsche Einschätzungen.
Going plural: compositional routes to multiplicity in natural language
In this class we'll develop a compositional approach to plurality using concepts from Boolean algebra (Winter, 2001). We'll discuss topics including collectivity, distributive readings, and plural quantification. Time permitting, we'll discuss plural ontologies beyond individuals, such as events and times.
Continuations and friends: what linguists can learn from programming language theorists
Richard Montague famously asserted "There is in my opinion no
important theoretical difference between
natural languages and the artificial languages of logicians" (1970,
p222). In this course we'll be exploring a natural extension of
Montague's conjecture. exploring parallels between natural languages and
the artificial languages of computer programmers. The main focus
will be on techniques developed by functional programmers for
programming languages grounded in the lambda calculus, and their utility
for analyzing phenomena in natural language semantics. Topics covered
will include continuations and scope-taking, monads and exceptional scope, as well as applicative functors and enriched semantic composition.
- 教官: Yichi Zhang
Topics in Logic and Semantics: Conditionals
The truth condition for material conditionals states that a material conditional is true whenever its antecedent is false or its consequent true. Suppose you didn't raise an eyebrow when you first learned this. On the material implication analysis, it's true that if you DID raise an eyebrow, the sun will explode! This surely is an odd result. Can we do better? In this course, we will examine ways to better theorize about conditionals in natural language. We will cover both indicative conditionals (e.g., "If Oswald didn't kill Kennedy, someone else did"), and counterfactuals (e.g., "If Oswald hadn’t killed Kennedy, someone else would have"). We shall survey a wide range of analyses: from the Gricean pragmatic defense of material implication to the Lewis-Stalnaker similarity semantics, to premise semantics, and to the more recent dynamic approach to conditionals. Along the way, you will be introduced to basic modal logics and conditional logics. In the second half of this course, we will explore the connection between conditional
- 教官: Sonia Ramotowska
- 教官: Jacopo Romoli
All the things you could have said: theories and experiments on implicit meanings
A core part of the study of our knowledge of meaning has to do with the inferences we draw from sentences, with the assumption that understanding the meaning of a sentence involves understanding what you can conclude from it. One of the most important of such inferences are 'implicatures,' an example of which is the conclusion that not all of the students did well on the exam, from a sentence like (1).
(1) Some of the students did well on the exam.
In its traditional conception, going back to the seminal work of the philosopher Paul Grice, implicatures arise on the pragmatic side of meaning, on the basis of implicit reasoning over what the speaker said and what she could have said instead. More recently, implicatures have been argued to arise more on the semantics side. In both approaches, alternative sentences which were not uttered by the speaker but could have been, play a crucial in the derivation of implicatures. In this course, we will systematically review the debate about the different approaches to this crucial type of inference we draw from sentences, combining both theoretical and experimental literature.