Paul Elbourne: Meaning: A Slim Guide to Semantics.
Hansen, Nat
Paul Elbourne
Meaning: A Slim Guide to Semantics.
Oxford and New York: Oxford University Press 2011.
viii + 174 pages
$24.95 (paper ISBN: 978-0-19-969662-8)
Philosophy of language has recently undergone a 'linguistics
turn', in which the most interesting contemporary philosophical
work on language has been informed by, or is in some cases
indistinguishable from, research in contemporary linguistics.
Elbourne's book is in the 'Oxford Linguistics' series and
it provides an excellent introduction to central topics in the
philosophy of language shaped by the linguistics turn. Chapters 1-3
concern features of word meaning, Chapters 4-6 consider sentence
meaning, and Chapters 7 and 8 discuss the interaction of meaning and
context and the relation between language and thought, respectively.
Philosophers unfamiliar with the linguistics turn in the philosophy
of language will find some aspects of Elbourne's introduction to
meaning unorthodox. For example, he favors an internalist account of
word meaning according to which word meanings are concepts, located in
our heads, and he doesn't flinch from the standard worry that such
a view poses problems for shared meaning and communication. On the
contrary, the fact that the internalist theory does not guarantee
communication seems to align with evidence of both interpersonal and
intrapersonal differences in grasp of the meanings of words (30-2). And
some topics that once dominated discussions concerning the philosophy of
language are only briefly mentioned. For example, skepticism about
synonymy is dispatched in a paragraph after the introduction of a few
examples of what appear to be synonymous expressions ('napkin'
and 'serviette', 'gorse' and 'furze', 34).
But this 'slim guide's' shift in perspective on the
nature of meaning is refreshing. It allows Elbourne both to discuss some
standard topics such as the dispute between 'referentialist'
and 'internalist' theories of word meaning and to examine some
fascinating interactions between the study of meaning and neighboring
disciplines like experimental psychology and neuroscience. In what
follows, I will briefly summarize the contents of the book as a whole
while dwelling longer on the elements that will appear most novel to
philosophers.
The more familiar elements of Elbourne's investigation of word
meaning in chapters 1-3 cover the inadequacy of dictionary definitions
as a systematic account of meaning (Ch. 1), and the advantages and
disadvantages of 'referentialist' and 'internalist'
theories of word meanings (Ch. 2), which includes a sketch of the tangle
of metaphysical categories caught up in debates about what words might
refer to (abstract objects, properties, etc.). Chapter 3 shifts focus
from the debate over where to look in general for the meanings of words
to consider the nature of synonymy, vagueness, and ambiguity of word
meanings (whatever those meanings might be).
It is in his examination of ambiguity in Chapter 3 that
Elbourne's empirically informed approach to the study of meaning
clearly emerges (38-42). There is an intuition that the multiple
meanings of certain ambiguous words, like 'column', are more
closely related to one another than the meanings of other ambiguous
words like 'bank'. Elbourne says that dictionaries capture the
intuition by using separate entries for the unrelated meanings of
'bank' while the related meanings of 'column' have
'just one entry with lots of subdivisions' (37). An
internalist about word meanings might maintain that a similar difference
in structure exists in the mind, but it isn't obvious what evidence
could support such a claim. But, surprisingly, Elbourne says that
'light has been shed on this issue by brain imaging' (38). He
cites recent studies of 'semantic priming', where processing
of word meaning is sped up when one sees a semantically related but
phonologically unrelated word ('paper'/'magazine'),
'phonological inhibition', where processing is slowed down
when one sees a homonym of particular word ('river
bank'/'savings bank', e.g.), and 'repetition
priming', when processing is sped up when one sees the same word in
quick succession, which provide evidence of a difference in structure.
In brief, processing times for pairs of phrases like 'lined
paper'/'liberal paper' were quicker than when
'liberal paper' was preceded with an unrelated control. That
provides confirmation of the intuition that the occurrences of
'paper' in both the phrase 'lined paper' and the
phrase 'liberal paper' are not mere homonyms (since processing
of a word is slowed down relative to a control when one sees a homonym
of that word): instead, they are a single word with related meanings.
This is a compelling demonstration of the relevance of findings from
neighboring areas of inquiry to the study of meaning.
After discussing word meanings in Chapters 1-3, Elbourne turns to
sentence meanings in Chapters 4-6. Chapter 4 concerns two candidates for
the meanings of sentences: Russellian propositions and sets of possible
worlds. Chapter 5 deals with entailment, presupposition (focusing on
definite descriptions), and structural ambiguity. Chapter 6 tackles the
nature of the compositionality of meaning, and explains how the meaning
of certain expressions can be modeled as functions which take the
meaning of expressions as inputs and yield the meaning of larger phrases
as their outputs. As with the chapters on word meaning, Elbourne brings
recent empirical discoveries in linguistics and related fields to bear
on topics central to the philosophy of language. For example,
traditional conceptual arguments for and against the idea that the
meaning of sentences can be modeled as sets of possible worlds are
combined with an argument that thinking of meaning as sets of possible
worlds contributes to a powerful explanation of the behavior of
so-called 'negative polarity items' (NPIs).
An NPI has to appear in a sentence together with an NPI-licensing
phrase for the sentence to be grammatical. Compare the following
sentences (the examples are from pp. 56-7):
(1) No gods show any mercy to mortals
(2) *Some gods show any mercy to mortals.
The unacceptability of (2) is due to the presence of the NPI
'any' without the presence of an NPI-licensing expression like
'no'. There are other NPI-licensing expressions in English
such as 'at most three' or 'it is never the case
that'. Elbourne describes an account (due to William Ladusaw) of
what unites the phrases that serve as NPI-licensors in terms of downward
entailment, which in turn is spelled out using the machinery of possible
worlds (60-1). If successful, this explanation provides an example of
how the ability to successfully predict a surprisingly diverse array of
phenomena can lend some empirical support to abstract philosophical
commitments.
Another thing to like about Elbourne's introduction is his use
of lively examples (drawn from law and politics as well as Buffy the
Vampire Slayer) to illustrate abstruse issues like the structural
ambiguity of sentences and the interaction of meaning and context. For
example, the discussion of structural ambiguity in Chapter 5 concludes
with a demonstration of the practical importance of semantics through a
fascinating discussion of the crucial role played by a structural
ambiguity in the Treason Act of 1351, one interpretation of which led to
the hanging of Sir Richard David Casement in 1916. The discussion of
different theories of how context shapes the semantic content of
sentences in Chapter 7 is similarly made concrete by reference to the
supreme court case of Smith v. United States, in which a drug dealer
traded an automatic submachine gun to an undercover police officer for
two ounces of cocaine (129) (the relevance of this case for debates
about context sensitivity was first noted by Stephen Neale). The drug
dealer was subject to a statute that held that crimes that involve a
defendant who 'uses' a machine gun 'during and in
relation to' drug trafficking is subject to a harsh thirty-year
sentence. So did the defendant 'use' a machine gun when he
traded it for drugs? The defense argued that the statute should be
understood as prescribing the use of a firearm as a weapon, even though
the words 'as a weapon' do not appear in the statute. This
claim aligns with certain views of 'implicit content', where,
in the right context, unpronounced material can be part of the content
of what is said when a sentence is asserted. The Supreme Court, however,
disagreed with the defense in a 6-3 decision. Might the court have been
swayed with a more persuasive account of the complex ways that meaning
and context interact?
Given that philosophy of language has taken the linguistics turn,
philosophers wanting an updated look at central topics in philosophy of
language should read this book. When combined with some of the further
reading Elbourne suggests, it could also be used as a way of bringing an
introductory philosophy of language course up to date with reference to
current research and controversies. Elbourne writes lively prose and he
conveys the sense that the contemporary study of meaning is rich,
exciting and developing rapidly. My only substantial complaint is that I
wanted more--philosophers would benefit from an empirically informed
introduction to dynamic semantics, experimental pragmatics, the meaning
of questions, imperatives, and so on. But such an introduction could no
longer be a 'slim guide'.
Nat Hansen
University of Reading
