Truth – The Minimal Theory - notes on Horwich

Truth or being true is unsusceptible to scientific or conceptual analysis. There is no such underlying nature of truth.

For example, “the proposition that quarks really exist is true if and only if quarks really exist, the proposition that lying is bad is true if and only if lying is bad.” (page 5)

Nothing more about truth need be assumed. To attempt to analyse some special quality of truth is to analyse a psudo-problem based on syntactic overgeneralization. (page 5)

The minimal theory therefore is: It is true that p if and only if p.

He says that truth as we normally think of it (some special property) is actually irrelevant to most philosophical and scientific problems. It’s verifiability that matters – is what I think he’s saying.

Alternative theories:


Truth is the property of corresponding with reality. But what if truth is simply indescribable or a device for semantic ascent – as the minimal theory states.


A system of beliefs is coherent when its elements are consistent with one another and when it displays a certain overall simplicity. According to coherence theory the whole system and each of its elements are true… if they are coherent. They use idealized justification – verification and truth are not the same. Our evidence may point on occasion incontrovertibly towards a conclusion that happens to be false. (page 9)


Truth is utility, true assumptions are those that work best. Horwich criticises this by saying the association between truth of a belief and its tendency to facilitate successful activity should not be exaggerated. He says that actions based on true belief can “nonetheless work out badly.” This doesn’t make sense. What is he saying truth means here? If something worked out ‘badly’ then by definition a true belief was not acted upon, or the appearance of it having gone ‘badly’ is the untrue part. I think I am a pragmatist when it comes to scientific and philosophical theories. Truth is indescribable, but when people use it they are almost always referring to something they believe to be true and they hold that belief due to its usefulness – i.e. its utility. It is true for them for as long as it can be, then a more useful theory arrives that disproves the other. Truth does not imply eternal or universal truth, it can be localized and subjective.

These theories don’t deny the correctness of the minimal theory, but they suggest it is incomplete.

P is true if and only if p – it seems very incomplete now having read more, surely it is circular reasoning? – but is an important point.

Horwich says the other theories are inflated and would look like this: ‘a is F’ is true if and only iff there exists an object x such that ‘a’ refers to x and ‘F’ is satisfied by x’.

He says that is unnecessary and can be broken down to the minimal theory. Minimalism denies the need of a conceptual analysis of truth. Minimalism denies that reference and satisfaction are complex or naturalistic properties.

One may try to formulate a finite theory of truth itself in a succinct body of statements that can be applied to other theories (in physics, maths etc) but according to minimalism there is no such thing. In the basic theory of truth there is an infinity of biconditionals of the form ‘that p is true if and only if p’. We can’t formulate it specifically because there are too many axioms.

One may propose an account linking truth with other matters such as assertion, verification, reference, meaning, success or logical entailment. Minimalism entails truth as a certain purity and is fairly independent of other ideas.

Therefore the minimal theory is complete and is only about truth, there is no extra content. Equivalence biconditionals – minimal theory, does say the other theorise are false, but they fail to explain truth.

Other difficulties with minimalism:

There is a problem that the minimal theory must be missing something – the naturalistic character that provides truth with its causal properties. A deflationary/minimal account of truth also entails an anti-realist perspective on science.

The truth of propositions may depend on the context, such as, ‘I am hungry’.

The correspondence intuition – that a representation is true if it corresponds to reality. - interesting objection as that is all about objective/subjective reality.

At the end of chapter 1 Horwich has gotten nowhere. He starts off saying that asking ‘what is truth?’ is a rhetorical question that implies answering to be futile. He says he will find a simple minimal explanation and concludes that ‘a statement is true if and only if it is true’, which he weakly disguises in a silly formula. I am not convinced anything has been explained at all.

Chapter 2: The proper formulation

(IMPORTANT – make sure your theory doesn’t fall fowl) There are three common problems with truth theories:

1. Isn’t said if the theory concerns the nature of truth itself, or merely the meaning of the word ‘true’.

2. Isn’t said what the theory is supposed to do. Adequacy conditions are left unclear.

3. Commitment to blatenrly implausible theses, such as ‘being true’ is not a property at all, or that every instance of p is true if and only if p. (I disagree with him completely. Being true is NOT a property of a statement that corresponds to reality. It is a property of reality, of which a statement is a part of. Someone says, 2+2=4 is true, that is the equivalent of them saying, ‘I believe 2+2=4.’ If and only if the first 2 (of something – in maths you can never know what it is you are referring to) is absolutely identical to the second 2 then it is true that is adding their value together will produce 4. Someone is saying if you try it, I believe you will be produce 4 and nothing else because they believe you are both part of a consensus reality Where the symbols represent the same value. We learn to count with objects as children, we are not just told that 2+2=4 on a blackboard, we are shown it with ‘physical’ objects that the symbols represent. Saying 2+2=4 is also meaningless. It is as useful as saying 1=1. It is the same as saying, ‘there is unity’ or ‘if things are the same, then things are the same’. It is saying, ‘this is a circle of thought – it joins up, the story makes sense’. In a more complex example: Some boys are playing football but when they stopped one of them kicked it and it smashed a window. One of them didn’t see it kicked but saw the lad take a run up and heard the smash at the right time. He say it is true that the other boy kicked the ball into the window. He is saying ‘if these are the events, which I believe they are, then they are the events’, or ‘if the same thing happened as I think happened then what I think happened happened and so my statement about it is true’. So someone saying something is true is just saying what they believe, which is to say a circular (and therefore baseless) set of reasons that satisfy curiosity. I say the beauty of a sunrise reflecting from a crisp mountain lake is truth itself, but I cannot justify my belief in that and so therefore truth is not a property of a statement but a property of reality, of which a statement is but a futile attempt to describe. Truth itself and the word true are very different things.

I have gone against all 3 of the rules of ‘the proper formulation’. But the problem is with the rules. The nature of truth itself and the word truth immediately become entangled. It cannot simply be stated that a theory is about one or the other, because it not clear what distinction the author is trying to make until they have explained themselves – which they will do. Therefore it is not a common problem of truth theories, rather it is part of the problem they are dealing with.

Regarding what a theory is supposed to do, this cannot be simply stated either. We pretend to ourselves and others that we are write something ‘hands free’ leaving little or no trace of our selves, when actually this is impossible. How can we know why we are writing something? Surely the purposes of a theory closely or identically align with what the creator believes their purposes to be. The purpose of the creation of this theory, for example, is to exercise my brain, help towards a degree and huge a range of purposes, such as because once I started I estimated I would feel better if I continued rather than stopped and that I don’t actually know why this theory is being created at all, yet along what I predict it to “do”. Therefore the ‘purpose’ of a theory cannot be said to be a separate thing from its creator. The theory is in fact is inseparable from its creator as consciousness is inseparable from existence. I wasn’t aware that words ‘do’ anything other than transmit meaning, i.e. communicate. The purpose of a theory being shared therefore is to communicate the theory – which is to communicate the ‘person’. Hence his second problem is not a problem. A reader judges for themselves how adequate a theory is. A theory is clearly meant to explain something, to communicate something. If he says there is a problem with the purpose of truth theories then there is a problem with the whole theory – the whole communicative act, not a problem that can be solved completely by attempting to state a simple purpose for a theory, although that may help improve clarity somewhat.

His third problem of the commitment to blatantly implausible theories is an exaggerated and incomplete summary.

Realism and Normativity

QR Code
QR Code horwich_minimal_theory (generated for current page)