Truth-Bearers and Truth-Makers
Tom Wetzel
William Alston's book "A Realist Conception of Truth"
got me to thinking about the questions,
a. What are the entities that may be true or false?
b. What is it for one of these entities to be true?
Alston formulates his "minimalist" view of truth via
what he calls a T-schema:
(T) The proposition that P is true if and only if P.
He thinks this is sufficient for what he calls a realist
"theory" of truth. This assumes of course that "propositions"
are the primary bearers of truth-value.
You might suppose that if his account aspires to be a "theory", it
should provide an account of what propositions are, given
the important role they play in his theory. He begs off doing
that, though.
Of course, we know what entities count as "propositions",
and Alston takes that to be sufficient for his purposes.
"Proposition" is a kind of technical term in philosophy.
We can take it that, by definition, "proposition" refers to whatever
it is (if anything) that is designated by expressions of the
form "that P" (where P is a declarative sentence) where such
expressions are used to indicate the target or "intentional object"
of cognitive states such as believing or suddenly realizing
(aptly named "propositional attitudes"), or of a linguistic act (such as
saying, stating, asserting) that gives vent to such
cognitive states. By extension, we can also say that any sentence, used
as a vehicle to communicate or express any such state or act, may
express or delineate a proposition in its use on that occasion.
Alston points out that there are, roughly, three views or "theories" as
to what "propositions" actually *are*:
i. Mind-dependent entities that exist as aspects of "propositional
attitudes".
"Internalists" such as Searle and Chisholm seem to favor
this view. This view has a problem accounting for how George and Jack
can agree on something, which seems to assume they accept the same
proposition. Also, the proposition a person assents to or entertains
might also be something that could be the case even if they had not
existed.
ii. Mind-independent abstract entities that are eternal and necessarily
existing.
On this view, the proposition that Bill Clinton lives in New
York has no necessary relationship to Bill Clinton since the latter
exists contingently whereas the proposition does not. Frege comes to mind
as a theorist who seems to favor a view of this sort. This theory has
a problem with accounting for how brains can somehow pick out a particular
proposition given its eternal, causally-inert existence independent of
physical things.
iii. States of affairs. Those that obtain or are the case are called
"facts".
On one view of states of affairs, these are entities that
have intrinsic logical structures and that
can have physical properties and physical particulars as constituents.
This seems to make them contingent entities that can enter into causal
relationships, and thus more accessible to animal cognitive systems.
On this view, then, if Jack believes that this road is 60 feet wide,
the state of affairs
(S) the road's being 60 feet wide
is the target of that belief, it is
what is designated by the "that"-clause "that the road is 60 feet wide."
I. Are Propositions Truth-Bearers?
Why does Alston think that propositions are the entities that, first and
foremost, are true or false? Why not say that beliefs or statements
(acts of stating) or sentences are truth-bearers?
He uses the familiar argument against sentences being the truth-bearers
based on their ability to be used to say different things in different
contexts. If Jack says
(P) I am hungry
he says something different than if George uses (P) to say something
on that occasion. And what Jack says might be true even though George
is faking it, and isn't really hungry.
Of course, one could say that truth or falsity only attaches to
sentences on some particular occasion of use, or in a particular
context. We need not suppose that this be a sentence that is
generated on the spot by a human. Suppose you encounter an exit
gate in a rapid transit system that flashes an LED display "You did
not pay enough fare" when you insert your ticket into the exit
turnstile. The referent of "You" is fixed by the context, that is,
you are the person who inserted that ticket, it is your ticket.
Of course this sentence might not be true. Maybe you did pay enough
fare but the machine is malfunctioning. But we understand what
that machine-generated sentence token is "saying" on that occasion.
It was designed to say that by its human designers.
Another argument against sentences as bearers of truth is based on
the ability of two speakers to say the same thing even though
they say it in different languages. If Angie says "el gato esta' negro"
and Rene says "the cat is black", we'd say they "said the same thing."
But the sentences are distinct. So what is this "thing" they both said?
Alston bases his case on the "what X said" locution. In other words,
If Jack says, "Twinkletoes is pregnant" (referring to a neighborhood
cat), George might say
(Q) What Jack said is true.
So George is predicating truth of something, something designated by
the phrase "what Jack said." Well, what is this referring to? Is it
referring to the sentence he mouthed on that occasion? Or the proposition
that Twinkletoes is pregnant?
Alston argues that "what A said" is most plausibly interpreted as
referring to a proposition. So, in this last case, "what Jack said"
refers to the proposition that Twinkeltoes is pregnant.
Alston's argument, then, can be summarized this way:
1. "What A said" is most plausibly interpreted as referring to the entity
designated by the "that"-clause in "A says that P".
2. We attribute truth or falsity to "what someone says."
3. The entity designated by the "that"-clause in "A says that P" is, by
definition, a proposition.
4. Hence, we attribute truth or falsity to propositions.
However, Alston admits that nouns like "belief", "statement" and "assertion"
are systematically ambiguous. That is, "belief" can be used to refer either
to a person's believing something or to what it is she believes.
"Statement",
likewise, can be used to refer either to the act of stating something or to
what it is they state on that occasion.
Moreover, it seems to me there is a similar ambiguity in talking about
what someone "says" or "asserts", between the proposition they have
expressed or asserted, on the one hand, and the sentence they have used to
state it.
For this reason, I think that Alston's reliance on the "what X said"
locution
makes his argument inconclusive.
For example, if Jack says "Gargon garbles gold funxations",
I might respond by saying to Jack, "What you said makes no sense."
I'm not saying I have grasped what he said but think it is
a ridiculous claim to make or that it is obviously false -- I'm
saying I have no idea what it is he is trying to say. His words
have not succeeded in delineating some claim that I can evaluate
as to its being true or false.
So by "what you said" I must be referring to the sentence or wording
he used, the noises that seemed to have the form of a linguistic
utterance but which I can't fathom.
"Saying" is thus ambiguous between the proposition expressed or stated
(if any) versus the words or sounds or inscription used to say or
express that proposition on that occasion.
When we talk about the "sayings" of famous people, we usually have in
mind partiuclar quotes, which are stated in a particular form of
words.
It's true of course that the proposition that was stated using a
particular sentence has only an accidental relationship to that
sentence used on that occasion to state it. Words occurring in that
sentence might be able to have a different reference or interpretation
in different contexts, as with the example of "You" generated by
the ticket machine. And a natural language evolves and it is only
accidental that particular words are linked to the particular features
or entities in the world they are linked to in current social practice.
A sentence abstracted from any context of use thus may have no determinate
truth-value. The truth-value it has seems to be derivative from
something else, what it is that someone wants to say on a particular
occasion.
And this, again, seems to support the view that it is the proposition --
what
it is that the speaker or writer wanted to say when using it -- that has
the more basic relationship to truth or falsity.
II. An Argument Against Propositions As Truth-Bearers
But I'm going to suggest that propositions can't be truth-bearers.
It seems to me that the word "true" is relational. To say that
something is true or false is to say it has or lacks some
appropriate relationship to something else. If Cynthia files
for divorce, claiming that Joe wasn't "true to her", she is
asserting that a certain relation didn't hold between her and
Joe.
If Jack says that the road is 60 feet wide, Jack has described
the road as being a certain way, as having a certain width.
He represented that chunk of reality as having a certain feature.
And I think we judge "what Jack said" as being true or false
depending upon whether the road is as Jack represented or
described it as being. If we get out a tape measure and
determine that the road is only 54 feet wide, we'd say
he was mistaken.
Thus, if we take "what Jack said" as referring to something that
has truth-value, then "what Jack said" must refer to something
that represents or presents or describes a way something is, and
we judge it true or false depending upon whether that thing is
as "Jack says it is".
Now, the reason I think propositions cannot be truth-bearers
is that they don't represent or describe anything. Either it is the
case that the road is 60 feet wide or not.
If we consider a simple, positive subject/predicate statement of
the form "A is F", it seems to me that this represents A as being
F. We judge this statement to be true if A is in fact F, and false
if A is not F. But this implies that there is a distinction between
A's being F -- the way the world is -- and the representation or
description "A is F". It is the representation or description which
we judge to be true or false.
If Jack weighs 40 pounds, it isn't his weighing 40 pounds that we
would say is true or false -- that is just the way he is.
But if someone says "Jack weighs 40 pounds", then we can judge
that true or false because it does represent Jack as having a
certain property. And if he has that property, we say that
statement is true; false otherwise.
So, a problem with the view that propositions are truth-bearers is
that propositions do not represent or describe the world as being
a certain way. As a mind-independent abstract entity, a proposition
is not a human artifact and thus does not describe or represent
or assert anything. It is only humans or their artifacts (such as
sentences on an occasion of use) that may be said to represent
or describe or assert that things are thus-and-so.
Alston had conceded that "belief", "statement", and "assertion"
are ambiguous between a state or act, on the one hand, and the
proposition that is its intentional object, on the other hand.
I have added that "saying", "statement", and phrases of the
form "what X says", are also ambiguous between propositions
and the words used to express or state that proposition.
I suggest, then, that it is open to us to take truth-values to
be properties of cognitive states (believing) and linguistic
acts like asserting or saying. But a statement or assertion
can only be made by using a sentence, I suppose, and a belief
can only be communicated to others using a sentence. And thus
the sentence on a given occasion of use will tend to have a
proposition uniquely associated with it (if we leave out
the question of failures of reference), namely, the "intentional
object" of the particular statement being made, or of the belief
being communicated.
Thus, sentences, on a particular occasion of use in a particular
context, do have a truth-value. Hence, sentences can be regarded
as bearers of truth-value, in a derivative sense. Of course, it
is also true that the sentence used by the speaker on a given
occasion may have an interpretation that diverges from her
"intentional object" -- the person didn't say what she meant.
Again, the possible divergence between speaker meaning and
conventional meaning supports the thesis that a sentence may be a
bearer of truth-value, for if it were only the intentional
object of the act of speaking that was relevant in determining
truth-value, there could be no such divergence of truth-value
between what they *actually* said and what they meant to say.
On the view that I'm suggesting here, when we use phrases
of the form "what X said", where this refers to something
that may be said to be true or false, what we're referring
to is either the sentence X used on that occasion, or their
act of saying, of representing the world as being thus-and-so.
There is still a problem with this interpretation, though.
If we say
(U) What Jones said is the same thing that Bill said yesterday.
we are not committing ourselves to saying that Jones uttered
exactly the same sentence that Bill did the day before.
We are saying, I think, that they are both representing or
claiming the same thing as far as the way the world is.
And they might have used somewhat different language to do that.
It may seem that the most natural way to interpret (U) would be
to say that (U) is implying that the proposition stated by Jones
is the same as that stated by Bill. That is because the words
used by Bill on that occasion assert the same proposition as
did Jones' words the previous day.
I think this ambiguity -- between the proposition stated and the
act of representing the world as being a certain way -- is a
characteristic of our talk of "saying."
III. Propositions As States of Affairs
I mentioned at the outset that one of the theories of
what propositions are that Alston mentions is the view that
they are states of affairs. This should set off alarm bells
right away, given that Alston takes propositions to be
truth-bearers.
Amongst those philosophers who countenance states of affairs,
states of affairs are usually regarded as truth-makers. That is
one of the main reasons for countenancing these entities.
So, if Jack says that the road here is 60 feet wide, according
to Alston, the proposition
(V) that the road is 60 feet wide
is supposed to be true. But if this is true, this presumably is
to made true by the world's being the way (V) says it is, that is,
by (S) being the case:
(S) the road's being 60 feet wide
But on theory (iii), that propositions *are* states of affairs,
(V) and (S) are the same entity. Presumably, (V) is (S) qua
object of cognition or qua intentional object of a propositional
attitude.
But, if so, then it is hard to interpret (V) as representing the
world as being a certain way or as desribing or claiming something
about it, which would seem to be part of the idea of a truth-bearer.
Part of that idea, I've suggested, is that "true" is implicitly
relational, that there is something that the truth-bearer is true to.
But if (V) just *is* (S), then it just *is* its truth-maker, in which
case there is nothing for it to be true *to*, where this implies
something other than itself, which it is articulating or claiming to hold.
It seems to me this is another reason for rejecting Alston's claim
that propositions are truth-bearers. For, theory (iii), as a theory
of what propositions are, has certain virtues (such as
ontological simplicity). But if propositions
are truth-bearers, it's hard to make sense of the identification
of propositions with states of affairs.
This suggests to me then, that if one is attracted to theory (iii),
as a theory of what propositions are, one should reject the view
that propositions are truth-bearers. This suggests an alternative
view, according to which beliefs, statements, and sentences on an
occasion of use, are true or false depending upon whether the
associated proposition (the intentional object of the belief, or
what the sentence expresses) is the case (is identical with a state of
affairs that is the case). And a sentence, on an occasion of use,
may also be said to be true if the sentence is being used on
that occasion to articulate a proposition that is identical with a state
of affairs that is the case.
Tom Wetzel