Learnability and Cognition
Steven Pinker, 1989

Chapter 1. A Learnability Paradox
Why does He gave them a book sound natural, but He donated them a book sound odd? With this question, Pinker embarks on a quest to uncover the paradox of language learing. The problem is the acquisition of lexical arguments. Several initial solutions turn up negative. This is the paradox.

1.1 Argument Structure and the Lexicon
Certain verbs require certain combinations of surrounding noun phrases, some take only a subject, some, subject and object, and some take both of those as well as a "recipient". How do children learn these patterns given incomplete evidence?

1.2 The Logical Problem of Language Acquisition
Use of positive evidence is to build a hypothesis which covers all of the known cases. Use of negative evidence is to restrict the hypothesis so as to exclude known invalid cases. But the evidence shows that parents rarely communicate negative evidence. And yet, all speakers form essentially the same language hypothesis.

1.3 Baker's Paradox
Many (3-argument) verbs, which take a dative "to" PP, also can be used with reversed object and recipient ans some cannot. John gave a dish to Sam becomes John gave Sam a dish. John reported the accident to the police does not become *John reported the police the accident. Without negative evidence, how can the child know this? There is a similar problem with certain two-argument verbs.

1.4 Attempted Solutions to Baker's Paradox
Several "nonsolutions" are like saying you can get rich by buying low and selling high. Some people have argued that some form of negative evidence may be either available or unnecessary. Some have argued that the child's grammar is strictly conservative (possible cases are deemed invalid until instantiated). Some have claimed that argument structure might be transmitted by a subtle syntactic correlate. Each of these "solutions" fails on closer inspection.

Chapter 2. Constraints on Lexical Rules

2.1 Morphological and Phonological Constraints

2.2 Semantic Constraints

2.3 How Semantic and Morphological Constraints Might Resolve Baker's Paradox

2.4 Evidence for Criteria-Governed Productivity

2.5 Problems for the Criteria-Governed Productivity Theory

Chapter 3. Constraints and the Nature of Argument Structure

3.1 Overview: Why Lexical Rules Carry Semantic Constraints

3.2 Constraints on Lexical Rules as Manifestations of More General Phenomena

3.3 A Theory of Argument Structure

3.4 On Universality

Chapter 4. Possible and Actual Forms

4.1 The Problem of Negative Exceptions

4.2 Transitive Action Verbs as Evidence for Narrow Subclasses

4.3 The Nature of Narrow Conflation Classes

4.4 Defining and Motivating Subclasses of Verbs Licensing the Four Alternations

4.5 The Relation Between Narrow-Range and Broad-Range Rules

Chapter 5. Representation

5.1 The Need for a Theory of Lexicosemantic Representation

5.2 Is a Theory of Lexical Semantics Feasible?

5.3 Evidence for a Semantic Subsystem Underlying Verb Meanings

5.4 A Cross-Linguistic Inventory of Components of Verb Meaning

5.5 A Theory of the Representation of Grammatically Relevant Semantic Structures

5.6 Explicit Representations of Lexical Rules and Lexicosemantic Structures

5.7 Summary

Chapter 6. Learning

6.1 Linking Rules

6.2 Lexical Semantic Structures

6.3 Broad Conflation Classes (Thematic Cores) and Broad-Range Lexical Rules

6.4 Narrow Conflation Classes and Narrow-Range Lexical Rules

6.5 Summary of Learning Mechanisms

Chapter 7. Development

7.1 Developmental Sequence for Argument Structure Alternations

7.2 The Unlearning Problem

7.3 Children's Argument Structure Changing Rules Are Always Semantically Conditioned

7.4 Do Children's Error's Have the Same Cause as Adult's

7.5 Acquisition of Verb Meaning and Errors in Argument Structure

7.6 Some Predictions About the Acquisition of Narrow-Range Rules

7.7 Summary of Development

Chapter 8. Conclusions

8.1 A Brief Summary of the Resolution of the Paradox

8.2 Argument Structure as a Pointer Between Syntactic Structure and Propositions

8.3 The Autonomy of Semantic Represenation

8.4 Implications for the Semantic Bootstrapping Hypothesis

8.5 Conservatism, Listedness, and the Lexicon

8.6 Spatial Themas and Abstract Thought


Top of Page | LCAAS Opinion | Sort by Topic | Sort by Title | Sort by Author