LINGUIST List 13.3354

Wed Dec 18 2002

Review: Semantics: Lepore (2000) reposting of 13.3273

Editor for this issue: Terence Langendoen <terrylinguistlist.org>


What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in. If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Simin Karimi at siminlinguistlist.org.

Directory

  1. Alice G. B. ter Meulen, Lepore (2000) Meaning and Argument (reposting of 13.3273)

Message 1: Lepore (2000) Meaning and Argument (reposting of 13.3273)

Date: Tue, 10 Dec 2002 13:17:17 +0000
From: Alice G. B. ter Meulen <atmlet.rug.nl>
Subject: Lepore (2000) Meaning and Argument (reposting of 13.3273)

Lepore Ernest (2000) Meaning and Argument: An Introduction to Logic
through Language. Blackwell Publishers, 418pp, hardback ISBN
0631205810, $70.95, paperback ISBN 0631205829, $33.95.

Book Announcement on Linguist:
http://linguistlist.org/get-book.html?BookID=3030

Alice G.B. ter Meulen, Center for Language and Cognition,
University of Groningen, NL

[This is a reposting of the review that appeared at
http://linguistlist.org/issues/13/13-3273.html
which corrects the character encoding errors in that issue.
We apologize for the errors in the previous posting. --Eds.]

This first introduction to logic, written by a renowned philosopher on
the faculty of Rutgers University, caters to a general audience of
undergraduate students with no special background in mathematics or
computation, nor in linguistics. The comprehensive textbook consists of
17 chapters and an appendix (55 pp.) offering many interesting insights
into the relation between linguistic structure and logical form, as
well as a very useful, exhaustive list of answers to the exercises. For
linguists, the appendix may well present the most interesting part of
the book, as it discusses the need for event arguments, the semantics
of adverbs, anaphora, comparison and alternative semantics of
conditionals, conjunction, negation and relative clauses.

The main tenet of the author is the claim (p. 39) that translating
ordinary English to truth-conditional logical formulas often requires
thoughtful reflection on possible meanings in context, which will never
be reducible to a mere matter of automatically applying computational
algorithms. Although this categorical statement as such may not be
controversial, it could have served the intended audience if references
had been included to the semantics of natural language, in spite of its
limitations to only smaller fragments of English. The index mentions
logicians of historical repute as Quine, Russell and Carroll, though
Donald Davidson is absent, surprisingly so, given his formative
influence over the author, admitted in the acknowledgements (p. xiv).
Given the subtitle of this book, it is truly regrettable that no
systematic references are provided to more contemporary logicians and
the semantics of natural language in the wake of Richard Montague's
pioneering research in the early 1970s, though references at the end of
some chapters do include a few selected papers by contemporary
semanticists. Early on the student is made aware of the important
distinction between conversational inference, which is non-monotonic,
i.e. cancellable when information is added, and monotonic, non-
cancellable logical deductions, including analytically necessary
truths. Accordingly, the chapters on the logic of statements
(propositional logic) pay lots of attention to detecting deductive
arguments in actual English texts, while developing truth tables and
the nice tree style presentation of tableau techniques, which often
appeals to students and trains their analytic insights considerably.

Linguists may find that sometimes odd views are proclaimed: only NPs
can be the subject of sentences (p. 20), dismissing clausal subjects as
in 'That John smokes bothers Mary', and the really misleading claim
that 'is a dog' is a common noun, and 'is brown' an adjective (p. 135).
Property predicates are defined as the result of omitting a singular
term from a statement (p. 135), but this is later not generalised to
allow for quantified statements. In fact, variables are not introduced
until binary relational predicates are (p. 198). Lepore's language of
monadic predicate logic is at first unusually limited without
variables, although individual constants are admitted, but with
superscripts indicating the number of arguments a predicate requires.
Accordingly (p. 181), 'John drinks water' is translated to D1j, whereas
'Someone drinks water' is symbolized as E D1 (where "E" is the
existential quantifier) and 'Everybody drinks water' as A (P1 --> D1)
(where "A" is the universal quantifier; note also that the number 1 is
superscripted in Lepore's notation). This logic at once confusingly
indicates argument structure, but does not analyze object NPs, departing
radically from linguistic syntactic structure, and it makes monadic
predicates syntactically ambiguous. As teacher of logic classes you
would like to have seen a justification of this peculiar choice of
notation, intended perhaps as a simplification serving a pedagogical
purpose.

Another matter of considerable importance to linguists is the issue
whether indefinite NPs are referential or quantificational. Although it
certainly is commendable that this book addresses it so explicitly, and
offers an account of unselective binding, the arguments adduced to
convince the reader that they are quantificational are shaky at best.
Lepore claims (p. 138) that 'something' is quantificational because the
two statements 'something is a dog' and 'something is not a dog' do not
constitute a contradiction, as they would, had their subjects been
singular referring terms. The student must be puzzled when this
argument would lead one to conclude erroneously that universal
quantifiers, introduced on the following page, must hence also be
referential. But then a more convincing test is presented, appealing to
the fact that universal quantifiers cannot be referential like singular
terms, since they cannot be substituted for reflexive pronouns, salve
veritate. This would have been a natural point to introduce variables
as contextual names or symbolization of deictic pronouns, but for this
to happen the student must live with this at best awkward logical
language for another 54 pages. The semantic notion of a variable
assignment function and a proper definition of truth and satisfaction
are regrettably beyond the scope of this introductory textbook.

The key advantage of this book over other logic textbooks resides it
its appendix, where Lepore shows remarkable breadth and awareness of
linguistic research. He presents interesting analyses of comparatives,
inviting the student to reflect on a choice between a pragmatic account
where uniqueness of definite descriptions with superlatives is not
required and a semantic one which does require uniqueness. His
discussion of gerunds and infinitives, different kinds of adjectives,
bare infinitives in perception reports, and non standard quantifiers
present some important developments of the semantics of natural
language that have taken place over the past 40 years. It would have
been appropriate here to include the proper references to the original
sources.

ABOUT THE REVIEWER
Alice ter Meulen is the co-author of Mathematical Methods in
Linguistics (Partee, ter Meulen and Wall, Studies in Linguistics and
Philosophy, Kluwer Academic Publ. 1990). She has a Ph.D. in Philosophy
of Language from Stanford University (1980), is currently working on
the dynamic semantics of tense, aspect and temporal reasoning, and is
appointed as chair of English Linguistics at the University of
Groningen, the Netherlands and research member of the Center for
Language and Cognition.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue