The death of Donald Davidson last
year, shortly after the publication of this book, brought to an end one of the
most fruitful and influential philosophical careers of the last hundred years.
Along with Quine (with whom he is most often compared), Davidson profoundly
affected the direction of analytical philosophy in the second half of the
twentieth-century. Although he never published a monograph, and his papers
spanned a wide range of philosophical problems, Davidson’s work formed a
unified system that established its author as one of the driving forces in many
areas, most notably, the philosophies of language, mind, and action, all of
which were revolutionized by his seminal publications.

Donald Davidson is a
collection of seven new papers by distinguished philosophers on each of the
various aspects of Davidson’s work. A particularly welcome feature of this
excellent book is the effort made by the contributors to reflect the unity of
Davidson’s philosophy. This is not only aided by the connections made between
various areas of Davidson’s work in each article, but also by an excellent
editor’s introduction and the inclusion of papers on some of the elements which
give Davidson’s philosophy its unity (for example, Davidson’s event ontology,
detailed in Paul Pietroski’s essay "Semantic and Metaphysics of Events").
The papers are more expository than critical and the result is a first rate
introduction to Davidson’s philosophy which will be most beneficial to
students. The book is sufficiently accessible for undergraduates to digest but
will also interest graduate students and those with a professional interest in
Davidson’s philosophy.

The first paper, by Ernest Lepore and Kirk Ludwig,
presents a particularly clear account of Davidson’s truth-theoretic philosophy
of language. Davidson’s revolutionary claim is that a theory of truth (along
the lines of Alfred Tarski’s axiomatic truth-theory for formal languages) can,
with suitable modifications, double up as a theory of meaning. Tarski’s
recursive definition of truth for a language L begins by establishing
some basic axioms that assign semantic values to the terms of L so as to
issue in theorems of the form:

(T) "S" is T
in L if and only if S.

The quoted letter is here understood as standing for a
structural description of a sentence (S) of L. (T),
however, is not a sentence of L but a sentence of a metalanguage
used to talk about L. Evidently, the T-predicate in such theorems
will be co-extensive with a truth-predicate for sentences of L.
Accordingly, "Convention T" amounts to a necessary and
sufficient condition for predicating truth of any sentence in a given language.
So, for example, we have:

            (T1) "Snow is white" is true in
L if and only if Snow is white,

as an instance of the theorem schema (T). The
apparent circularity of the convention stems from the fact that the language in
which (T1) is given (the metalanguage) here shares terms of the language
(the object language) from which the sentence named by the quotation device is
taken. Had the metalanguage been English, and L been German, we would
have:

            (T2) "Schnee ist wiess" is true
in L if and only if snow is white.

The fact that (T2) is genuinely informative to an
English speaker who does not speak German illustrates that the apparent
circularity of (T1) is illusory. Specifying the terms and their
assignments is, of course, to be taken care of by the axioms used. Davidson’s
claim is that a complete theory of meaning for L will coincide, in all
relevant ways, with a theory of truth for L. Lepore and Ludwig present
the details of this technical thesis with impressive clarity.  The problems for
a Davidsonian theory of meaning, and the objections that have been raised
against it, are explained alongside the rejoinders to them.

The informative nature of (T2)
plays an important part in Davidson’s extension of his theory from the realm of
the purely theoretical to the empirical. Davidson claims that a theory of the
kind just sketched could be used to "radically interpret" the speech
of a previously unknown language. If confronted by a community whose language
was completely new to us we could ascertain what they meant by their statements
if we could determine the truth-conditions of those utterances, according to
Davidson. This claim, explored in Piers Rawlings’ contribution to the volume ("Radical
Interpretation") is clearly intended to both support and be supported by
the claims made regarding truth and meaning. If a Davidsonian theory of meaning
really is sufficient for the project of radical interpretation, this would
surely confirm the adequacy of the theory of meaning: the theory would seem,
after all, to be sufficient to allow us to understand anything said in the
language under scrutiny. Rawling subjects Davidson’s argument to detailed
analysis, identifying the premises required and exploring the roots of the
enterprise in the works of Ramsey, Tarski, and Quine, as well as pointing out
the difficulties that still stand in the way of the claim.

Davidson’s work in the philosophies
of mind and action is of at least equal importance to his philosophy of
language. Alfred Mele’s article on Davidson’s philosophy of action and Jaegwon
Kim’s paper on his philosophy of mind each provide excellent accounts of
Davidson’s contributions in these fields and, in doing so, provide enough
material to introduce the subjects to those unfamiliar with them. It is in
these two areas that Davidson’s event ontology is most explicitly called on.
His theory of "Anomalous Monism" in the philosophy of mind seeks to
draw on the insights of identity theories without abandoning an account of
mental causation by holding that all mental events are physical but not all
physical events are mental. This theory has been widely and vigorously
discussed and the objections to the theory are explained clearly by Kim. In the
philosophy of action, Davidson has instigated perhaps even more discussion by
claiming that only by treating actions as events can the semantics of action
sentences be explained. His further claim that reasons for actions are causes
of those events has been so influential as to become the orthodox position in
the field, as Mele illustrates.

The other contributions to the
volume examine Davidson’s work in epistemology (Ernest Sosa) and even the
consequences of his work for literary theory (Samuel C. Wheeler III). The field
of literature on Davidson is already substantial and, as his work is
posthumously collected, likely to continue growing. The quality and clarity of
the papers in this volume, however, ought to guarantee it a high-ranking place
in that field.

© 2004 Graham Stevens

Graham Stevens, Ph.D. lectures in
philosophy at the University of Manchester and has also taught at the University
of Southampton. He is the author of articles in the philosophies of language,
logic and mathematics.

Categories: Philosophical