Ambiguity

« in which this is possible.

Semantic ambiguity is one such way, but there are others: homonymy (mentioned in §1),vagueness, relativity, indexicality, nonliterality, indirection and inexplicitness.

All these other phenomena illustratesomething distinct from multiplicity of linguistic meaning.

An expression is vague if it admits of borderline cases (seeVagueness ).

Terms like ‘bald', ‘heavy' and ‘old' are obvious examples, and their vagueness is explained by the fact that they apply to items on fuzzy regions of a scale.

Terms that express cluster concepts, like ‘intelligent', ‘athletic'and ‘just', are vague because their instances are determined by the application of several criteria, no one of whichis decisive.

Relativity is illustrated by the words ‘heavy' and ‘old' (these are vague as well).

Heavy people are lighterthan nonheavy elephants, and old cats can be younger than some young people.

A different sort of relativity occurswith sentences like ‘Jane is finished' and ‘John will be late'.

Obviously one cannot be finished or late simpliciter but only finished with something or late for something.

This does not show that the words ‘finished' and ‘late' areambiguous (if they were, they would be ambiguous in as many ways as there are things one can be finished with orthings one can be late for), but only that such a sentence is semantically underdeterminate - it must be used tomean more than what the sentence means.

Indexical terms, like ‘you', ‘here' and ‘tomorrow', have fixed meaning butvariable reference.

For example, the meaning of the word ‘tomorrow' does not change from one day to the next,though of course its reference does (see Demonstratives and indexicals ).

Nonliterality, indirection and inexplicitness are further ways in which what a speaker means is not uniquely determined by what their words mean (see Speech acts §4 ).

They can give rise to unclarity in communication, as might happen with utterances of ‘You're the icing on my cake', ‘I wish you could sing longer and louder', and ‘Nothing is on television tonight'.

These are not cases oflinguistic ambiguity but can be confused with it because speakers are often said to be ambiguous.

3 Philosophical relevance Philosophical distinctions can be obscured by unnoticed ambiguities.

So it is important to identify terms that do double duty.

For example, there is a kind of ambiguity, often described as the ‘act/object' or the‘process/product' ambiguity, exhibited by everyday terms like ‘building', ‘shot' and ‘writing'.

Confusions in philosophyof language and mind can result from overlooking this ambiguity in terms like ‘inference', ‘statement' and ‘thought'.Another common philosophical ambiguity is the type/token distinction.

Everyday terms like ‘animal', ‘book' and ‘car'apply both to types and to instances (tokens) of those types.

The same is true of linguistic terms like ‘sentence',‘word' and ‘letter' and of philosophically important terms like ‘concept', ‘event' and ‘mental state' (see Type/token distinction ).

Although unnoticed ambiguities can create philosophical problems, ambiguity is philosophically important also because philosophers often make spurious claims of it.

Indeed, the linguist Charles Ruhl ( 1989 ) has argued that certain ostensible ambiguities, including act/object and type/token, are really cases of lexical underdetermination.Kripke (1977 ) laments the common stratagem, which he calls ‘the lazy man's approach in philosophy', of appealing to ambiguity to escape from a philosophical quandary, and Grice (1967 ) urges philosophers to hone ‘Modified Occam's Razor: senses are not to be multiplied beyond necessity'.

He illustrates its value by shaving a sense off the logicalconnective ‘or', often thought to have both an inclusive and exclusive sense.

Grice argues that, given its inclusivemeaning, its exclusive use can be explained entirely on pragmatic grounds (see Implicature §6 ; Pragmatics §12 ). Another example, prominent in modern philosophy of language, is the ambiguity alleged to arise from the distinctionbetween referential and attributive uses of definite descriptions (see Descriptions §5 ).

Less prominent but not uncommon is the suggestion that pronouns are ambiguous as between their anaphoric and deictic use.

So, forexample, it is suggested that a sentence like ‘Oedipus loves his mother' has two ‘readings' - that is, it is ambiguous- because it can be used to mean either that Oedipus loves his own mother or that Oedipus loves the mother ofsome contextually specified male.

However, this seems to be an insufficient basis for the claim of ambiguity.

Afterall, being previously mentioned is just another way of being contextually specified.

Accordingly, there is nothingsemantically special in this example about the use of ‘his' to refer to Oedipus.

Claims of structural ambiguity can alsobe controversial.

Of particular importance are claims of scope ambiguity, which are commonly made but rarelydefended (see Scope ).

A sentence like ‘Everybody loves somebody' is said to exhibit a scope ambiguity because it can be used to mean either that for each person, there is somebody that that person loves or (however unlikely)that there is somebody that everybody loves.

These uses may be represented, respectively, by the logical formulas‘ (8x)(9y)(Lxy)' and ‘ (9y)(8x)(Lxy)'.

It is generally assumed that, because different logical formulas are needed torepresent the different ways in which an utterance of such a sentence can be taken, the sentence itself has twodistinct logical forms (see Logical form ).

Sustaining this claim of ambiguity requires identifying a level of linguistic description at which the sentence can be assigned two distinct structures.

Some grammarians have posited a levelof LF, corresponding to what philosophers call logical form, at which relative scope of quantified noun phrases maybe represented.

However, LF of this kind does not explain scope ambiguities that philosophers attribute tosentences containing modal operators and psychological verbs, such as ‘The next president might be a woman' and‘Ralph wants a sloop'.

An utterance of such a sentence can be taken in either of two ways, but it is arguable thatthe sentence is not ambiguous but merely semantically underdeterminate with respect to its two alleged ‘readings'.Notwithstanding the frequency in philosophy of unwarranted and often arbitrary claims of ambiguity, it cannot bedenied that some terms really are ambiguous.

The nouns ‘bank' and ‘suit' are clear examples and so are the verbs‘bank' and ‘file'.

Philosophers sometimes lament the prevalence of ambiguity in natural languages and yearn for anideal language in which it is absent.

But ambiguity is a fact of linguistic life.

Despite the potentially endless supply ofwords, many words do double duty or more.

And despite the unlimited number of sentences, many have severalmeanings and their utterance must be disambiguated in the light of the speaker's likely intentions.. »