[ Pobierz całość w formacie PDF ]
own metaphysics. Thus, he might reinterpret the folk s use of now as being
associated with the concept now*. Yet even if Mellor revises which concept
he associates with that word, it is unclear how this bears on the conceptual
practices of the folk. The folk are ignorant of these philosophical views and
revisions. If they were in error before, they remain in error. So even if the
replacement procedure enables clued-up philosophers not to speak falsehoods
when they use such words as now , later , soon and the like, it leaves the
folk in the lurch.
Lastly, a complication: some connotations admit of degree. C causes E
has the connotations that C is evidence for E, that C explains E and that
C is a means to E. Causation can be probabilistic, but the closer C comes
to determining E & the stronger the evidence is that C provides for E, the
better C explains E and the more useful C is as a means to E (Mellor 1995:
93). Can this be reconciled with the fact that the connotations of causation
are semantic entailments of C causes E ? Here is one way. Probability admits
of degree. So too do evidence, explanation and the usefulness of a means to
an end. The degree of probability is commensurate with these other degrees:
C s being a cause with probability p of E entails that C provides evidence for
E with a degree commensurate to p, that C explains E to a commensurate
degree, and so on.
56 Chris Daly
Having outlined Mellor s method, I turn to its application to the issue of
communication.
3 Mellor s theory of communication
A theory of communication tells us what communication is, and thereby what
we need to do to communicate with others. Mellor offers a theory of com-
munication of which linguistic communication is a special case. He makes a
truth-theoretic component central to his theory. Mellor (1990: 81) offers two
important truisms about truth. The first draws on Aristotle s Metaphysics (1928:
7, Book IV). It is that to believe or say truly is to believe or say, of what is, that
it is, or of what is not, that it is not (Mellor 1990: 82). The second draws on
Ramsey (1990b).4 It is that truth is that property of our beliefs which ensures
that the actions they make us perform will succeed (Mellor 1990: 82).
A connotation may be obvious or unobvious. Presumably, truism is Mellor s
term for an obvious connotation. I assume that Mellor takes the above truisms
to express necessary truths. Mellor (1991c: 275) elsewhere says that I expect
truth itself to be defined as the property of full beliefs that guarantees the
success of actions based on them, probability providing the weaker assurance
of success that expected utility spells out . Philosophical orthodoxy takes (cor-
rect) philosophical definitions to state necessary truths. So I take Mellor to be
stating a (purported) necessary truth about the property of truth. Mellor also
takes his theory of communication to provide the truth conditions of beliefs:
we can t equate a belief s truth conditions with those in which every action
it helps to cause succeeds. But we can if we restrict the actions to those
caused just by it and some desire. Then its truth conditions are what I
shall call its utility conditions : those in which all such actions would
achieve the desired end.
(Mellor 1991d: 23)
Specifying that a belief s truth conditions are its utility conditions is,
presumably, stating a necessary truth about those truth conditions.
Now the theory. Mellor takes communication to be a form of observation.
Suppose you observe some fact, and so acquire a belief about it. What makes
your observation a good one? Mellor s answer is that there has to be a causal
link between the fact observed and your belief about it, such that your belief
is true because the fact which makes that belief true has caused you to get
that belief. Some observations are direct, others are indirect. To make an
indirect observation of something, Æ, a learnable correlation between Æ and
something else that we can observe directly is needed. This is a sign of Æ. We
make an indirect observation by first making a direct observation of a sign,
and then making an inference from that to what we believe the sign signifies
(Mellor 1990: 86).
Suppose Sam Spade hears the doorbell ring (i.e. directly observes it). If his
Truth and the theory of communication 57
observation is a good one, then his belief that the doorbell is ringing is true.
Suppose he believes that the ringing is a sign of a client, and that a client is
ringing the doorbell. Then Sam indirectly observes a client. For Sam s indirect
observation to be a good one, his direct observation of the doorbell ringing must
be good, and the inference from the premises (his beliefs that the doorbell has
rung and that the doorbell s ringing is a sign of a client, and the fact that the
doorbell s ringing is a sign of a client) to the conclusion (that there is a client)
must be good. That is, the inference must transmit truth from its premises
to its conclusion. If Sam s indirect observation of a client is a good one, then
his belief that there is a client is true.
The case of communication is similar if more complicated. Suppose Sam
hears his secretary saying there s a client . Suppose his observation is a good
one. Then his belief that the secretary said there s a client is true. Suppose
Sam believes that what she said is a sign that she believes that there is a client.
Sam then infers that she believes that there is a client. Lastly, suppose Sam
believes that her believing that there is a client is a sign that there is a client.
Sam then infers that there is a client.
What distinguishes communication from the doorbell case is that X gets
the belief that p from what Y believes (namely, from Y s belief that p). But X
does not infer that p directly from what Y says. X infers p indirectly via what X
believes Y believes. Suppose Rabbit wants to tell Pooh the truth about honey,
and believes the truth to be that there is honey in the pot. Then:
Rabbit doesn t just want to say what s true: he wants to make Pooh believe
it. And as an experienced [informant], he knows that Pooh will only believe
what he says if Pooh believes that he believes it too. So Rabbit s immediate
desire is to give Pooh a true belief about what he, Rabbit, believes. So
what Rabbit will tell Pooh is not necessarily what he actually believes,
but what he believes he believes.
(Mellor 1990: 92)
Communication, then, is the production in the audience of beliefs about
what the speaker believes he believes (i.e. to produce in the audience beliefs
about some of the speaker s second-order beliefs).
Lastly, we want true beliefs because:
What is generally and inherently good about getting true beliefs is that
they re useful, in the following sense: truth is that property of our beliefs which
[ Pobierz całość w formacie PDF ]