Truth and facts

The following is an essay from a philosophy module, I think on Logic and Language. I haven’t re-read it to see to what extent I still agree with what I wrote then.

Is the truth of a proposition a matter of its corresponding with a fact?

In this essay I will discuss the correspondence theory of truth, and the claim that the truth of a proposition is a matter of its corresponding with a fact. To do this I will first outline Bertrand Russell’s criteria for a theory of truth. I will then outline the central claim common to correspondence theories. Then we shall see Russell’s structurally isomorphic version of the correspondence theory. I will then discuss Austin’s theory of conventions. Then I will compare a performative account, which is not a correspondence theory. Finally, I will conclude with my own beliefs on the subject.

Criteria for a theory of truth.

Russell was moved toward a correspondence theory, as he saw it as the only theory that could satisfy his criteria for a theory of truth. In the first criterion he claimed that a theory of truth must allow for the opposite of truth, namely falsehood. This was a reaction to ‘theories according to which all our thinking must be true’[1], which seemed to him a bold and erroneous claim to make. If all our thoughts were true, the word ‘true’ would become meaningless, expressing nothing of substance. As Russell saw claims of truth to be meaningful, there must be something that it is to be false.

The second criterion any theory of truth should satisfy is that ‘truth and falsehood must be properties of beliefs and statements’[2]. This means that we can only attribute truth or falsehood to beliefs and statements. It has been suggested that beliefs are true in virtue of the propositions to which they refer, and that it is the propositions themselves that carry the truth-value. For the purposes of this essay, I will accept the use of the terms Russell prefers, as this discussion is beyond my scope here. I will use terms such as ‘proposition’, ‘belief’, ‘sentence’, or ‘statement’ throughout, with the intended reference being to that which you wish to carry truth value.

The third criterion is that whatever it is that makes a proposition true must be a factor independent of the belief itself. If I believe something which happens to be the case, ‘I believe truly, not because of any intrinsic quality of my belief, but because of an historical event’[3]. If I believe something that is not the case, no matter what other properties the belief has, it will have the property of falsehood. ‘Hence, although truth and falsehood are properties of beliefs, they are properties dependant upon the relations of the beliefs to other things, not upon any internal quality of the beliefs.’[4]

We shall see that the correspondence theory of truth meets all these criteria, and it appears to be the only one that does. In fact, the third criterion appears to assume correspondence as the only valid theory. To say that the only way a truth theory can be valid is if a proposition corresponds to a fact, and that correspondence theory is correct, is circular. To break this circularity a justification for how propositions and facts correspond must be given.

What constitutes a correspondence theory?

For a theory of truth to be considered a correspondence theory, it must contain the central idea ‘that truth consists in a relation between propositions and the way things are in the world’[5]. This means that a proposition, or belief as Russell would have it, is true if and only if it corresponds to something in the world. ‘To say this belief is true is to say that there is in the Universe a fact to which it corresponds; and that to say it is false is to say that there is not in the Universe any fact to which it corresponds’[6].

If a correspondence theory is going to be successful, it must justify the central claim. Although this simple correspondence theory ‘has the advantage of conforming to our ‘common sense’ understanding of what it is to say something is true’[7], it is not satisfactory. For a theory to claim there is a correspondence between language and the world, there must be some description of how the two correspond.

Russell and structural isomorphism.

Many correspondence theories, including that of Bertrand Russell, are structurally isomorphic. This means that when he says true propositions correspond with facts, he is claiming that there is an exact relationship between a proposition on the one hand, and a worldly fact on the other. A true proposition will describe, accurately and without loss, the state of affairs to which it pertains, and from a state of affairs, we can draw the appropriate propositions, again without loss of meaning. Moreover, there is also the claim that the structures of propositions and facts are the same.

In the view of Russell, sense-data are logical atoms. These objects of direct acquaintance are atomic in the sense that they cannot be broken down into further constituent parts. These are equitable with, and are signified by, designators. The patch of green I see in my visual field is picked out by ‘this green’, when I am saying or thinking ‘this green’ whilst looking at that patch. Every experience I have of green will correspond with a new designation of ‘this green’. Therefore, the designator and logical atom are mutually necessitating. This relationship is true not only of definite designators, but any designator, such as the generalisation ‘green’. On any occasion where an object of sense perception is picked out, there is an accompanying designator.

Propositions are constituted by the relationships between elementary propositions, which are in turn constituted by definite designators. In the same structure, facts are made up of states of affairs, and ultimately logical simples such as sense-data. A proposition or belief claims the existence of a certain relationship between definite designators. The designators correspond to the direct objects of acquaintance. When these objects of acquaintance stand in a relationship, they become a fact. When this fact is accurately depicted by a proposition, the proposition is said to be true, and false if otherwise. It is this mirroring at each level, and the necessary mirroring of the structure, that correlates the world with true statements.

However, this form of the correspondence theory is problematic. If the facts and propositions are linked, or the same, in the manner suggested, we would expect both language and the world to contain the same number and type of constituents. For example, the statement ‘the monkey is in the cage’ consists of three obvious components. There is the monkey, then we have the cage, and the relationship in which they stand. However, the worldly state that would correspond to this statement seems to have only two constituent things, namely the monkey and the cage. Even if we admit that ‘the spatial relation that obtains between them is a third thing, then it is not the same kind of thing as the first two.’[8] Here we see that there is a difficulty in matching language with worldly items. The problem is highlighted further in the consideration of a possible language in which the meaning ‘the monkey in the cage’ is contained by one word. This would allow a proposition with only one obvious constituent, to correspond with two, or three depending on how we interpret the spatial relationship, worldly items. Language and the state of affairs cannot reasonably be said to have the same structure if it is possible for them to be so discordant.

Austin and conventional correspondence.

Austin attempted to show a version of correspondence that did not contain the structural isomorphism of Russell. He believed that the correspondence between fact and proposition was a matter of conventional relationships. In Austin’s theory, any word, as long as it is used and understood consistently, correlates ‘with types of situations to be found in the world.’[9] This form of convention is a descriptive one, and applies to sentences, rather than actually issued statements. The sentence ‘snow is white’ would, thanks to convention, describe the type of situation in which we would experience snow as white.

The second form of convention put forward by Austin is that of the demonstrative convention. A demonstrative statement is one that is issued, and makes a claim about a particular item on a particular occasion. For example, the statement ‘this snow is white’ is demonstrative in that it picks out a certain patch of snow at a certain point in time, and then makes a claim about it.

For a statement to be true it must relate to the actual situation. This relationship works purely by the conventions of language. ‘This snow is white’ is true if there is a state of affairs that correlates with the words used in the demonstrative conventions, and ‘is of the type correlated with those words by descriptive conventions.’[10] So for ‘this snow is white’ to be true, that snow must be white, the definite terms must be understood, and it must be the type of circumstance correlated with those words generally.

Austin avoids structural isomorphism by making it explicit that the conventions used are entirely arbitrary. The word-choice and even grammar do not affect the relationship between the statement and the world, as they appear to for Russell. Language is given meaning only by the way in which it is commonly used, rather than through the necessary relationship implied by Russell. Here we see that the structure of language may alter, but we have no reason to believe this is in line with correlating changes in the world.

The first problem to be raised against Austin regards his concentration on historic events. It appears that only indexical statements, those that are tied to a certain time and place, can be addressed by this theory. These ‘are statements which, in virtue of not being explicitly referential as general or indefinite statements are, cannot be employed in different situations’[11]. Due to this, demonstrative conventions would be irrelevant for this form of statement. As Austin’s account needs both forms of convention to work, this could be seen as a fundamental flaw.

However, a defender of Austin’s theory could argue that, although indexical statements appear to present the problem relating to demonstrative conventions, this is to misunderstand how indexical statements should be treated. The truth of indexical statements may not rest in their correlation with conventions, but upon the truth of more basic component statements. The truth of these basic statements, instead, may be judged by the method Austin proposes.

A more challenging problem for Austin comes when we look at just what it is his theory actually shows. It could well be the case that he gives a description of how we use conventions in language; ‘what they do is to determine what is being said.’[12] The conventions are such that they render formulations of arbitrary words meaningful to those that use the conventions. They do not, however, describe the form of the correspondence, in that they give no description of how a proposition fits a worldly fact. Strawson believes if the truth of a statement is judged on its fitting with the conventions, ‘it follows that in predicating truth of some statement we are either talking about the meanings of the words used by the speaker, or we are saying the speaker used the words properly.’[13] Austin has not moved the correspondence theory of truth any further than the central tenet ‘that the statement or proposition is true if, and only if, things are as it says they are.’[14] As we have seen, this is insufficient to establish that the truth of a proposition is a case of it corresponding with the facts.

A performative alternative.

A performative account of truth, such as Strawson’s, would not meet Russell’s criteria for a truth theory. Deflationary theories, of which the performative account is an example, state ‘that truth is not a substantive notion; there is nothing more to truth… than the use of the predicate ‘…is true’’[15]. In Strawson’s view, the predicate ‘…is true’ has only a performative role. ‘[W]e use ‘true’ to express our endorsement of a statement’[16], rather than making claims about the properties of an object or situation. If I say ‘snow is white is true’, all the truth predicate is adding is the affirmation that I agree with or believe that ‘snow is white’.

Performative theory is different to correspondence theory in that it shifts the focus of attention onto the question of how we use the truth predicate, rather than when we use it. It claims ‘that there really is no separate problem of truth but merely a linguistic muddle’[17]. Once questions regarding meaning and content are answered, there will be no mystery remaining regarding the nature of truth.


I believe that the correspondence theory has failed to justify a belief that the truth of a proposition is a matter of it corresponding to the facts. If one were to be successful, there must be a proper explanation of the relationship in which they stand. The claim of structural isomorphism that the structure of facts and propositions are the same is problematic. Propositions and facts appear to, at least on some occasions, have different numbers of constituent parts.

The theory of conventions put by Austin also fails to show a relationship. It turns out to describe how we come to understand the content and meaning of statements. This is not instructive as to how the statement can be said to correlate with the world, and must ultimately rest on the correspondence intuition that a statement is true if and only if what it states is the case.

The performative argument is one example of a position that rejects the criteria put forward by Russell. The most obvious question arises from the lingering correspondence intuition that if I say that a proposition is true, I am not just saying that I agree, but that I agree and believe the proposition corresponds to the world. I am not convinced that the discussion of truth can be solved by the discussion of content and meaning, as I feel answering the Strawson question may well fail to answer the question that drives the truth theories, without showing it to be empty.

Whilst I believe that the forms of correspondence theory shown here are inconclusive, I do not believe this rules out the possibility that facts correspond with propositions. It is only the case that the form of the relationship must be satisfactorily explained. However, I feel that Strawson raises the valid point that the question addressed by correspondence theories misses the way in which the truth predicate is used. In many, if not all, cases, the claim that something is true is closer to an affirmation than a statement of correspondence with a state of affairs. I am still undecided as to whether the argument that truth will become clear as content and meaning are explained holds weight.


Bertrand Russell, The Problems of Philosophy (Guernsey: Oxford University Press: 1912)

Anthony Harrison-Barbet, Mastering Philosophy (London: Macmillan Education: 1990)

A.C. Grayling, An Introduction to Philosophical Logic Third Edition, (Oxford: Blackwell Publishing: 1997)

G.E. Moore, Some Main Problems of Philosophy, (London: George Allen & Unwin Ltd: 1965)

Bob Hale and Crispin Wright, A Companion to the Philosophy of Language, (Oxford: Blackwell Publishing: 1997)

Thomas Mautner, Dictionary of Philosophy, (London: Penguin Group: 1996)

[1] Russell p.70

[2] Harrison-Barbet p. 81

[3] Russell p. 70

[4] Russell p. 70

[5] A. C. Grayling p. 138

[6] Moore p.277

[7] Harrison-Barbet p.81

[8] Grayling p.140

[9] Grayling p.141

[10] Grayling p.141

[11] Grayling p. 141

[12] Hale and Wright p. 321

[13] Grayling p. 142

[14] Hale and Wright p. 321

[15] Grayling p. 147

[16] Mautner p. 543

[17] Ramsey in Grayling p. 160

  1. Hi,

    I don’t really understand this whole business about predicates, content and meaning, but it seems to me that Agrippa’s trilemma forecloses the possibility of an objective truth, i.e., correspondence to ‘facts’.

    To argue meaningfully about what is true, we must agree on enough things so that we can have a common basis for discussion. In other words, when discussing ‘truth’ we are really discussing coherence of certain ideas with other ideas – those latter ideas being those we agree upon. Thus, truth is coherence. A claim is not true or false by itself but only in relation to a body of accepted ideas. When we say that some idea is true we mean that it coheres with what we believe is a body widely accepted ideas.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: