This is a revised version of a post I put up a while ago and then took down when someone made me realize it didn’t quite work. I think it works now.
You can hear Hilary Lawson debating with Barry Smith here. Lawson promotes a view he calls ‘non-realism’ (or sometimes ‘post-post-modernism’). This is meant to be a revolutionary take on how we see the world and our relation to it, getting rid of ideas that no longer serve us. The idea of truth is among these. Although there are many rhetorical flourishes, this is what I think it all comes down to.
Lawson’s theory is that the world is fundamentally open, meaning that nothing really true can be said of it. What we say about the world is all ‘metaphors’. These are not metaphors in the normal sense, according to which they figuratively describe real things, since for Lawson there are no real things. Hence I put the word in scare quotes. We use ‘metaphors’ to try to enclose the open reality. But reality resists closure. We should give up on trying to say true things and accept that everything we believe is just ‘metaphors’ – not truths, not even errors, just ultimately doomed attempts to enclose the world.
How then do we decide which metaphors to use? We should use the metaphors that suit our needs and interests. Lawson is no fool; he realizes that in saying all this he seems to be saying things that he wants us to believe are true. But he checks himself. No, even his theory – even the idea that the world is ‘open’ – is just more metaphors! He’s not asking us to believe that it’s true; he’s asking us to accept that it’s a set of metaphors more germane to our real interests. We should endorse these metaphors because if we do we’ll live better – more creatively, more openly, more artistically. Not because the ideas expressed in the metaphors are actually getting at truth.
I think this is a very bad idea, but things become so tricky when it comes to debating with Lawson. Any apparently truth-apt claim his opponents make is reinterpreted as the proposal of a metaphor. One is reminded of Karl Popper’s stories about Freudians claiming that all criticisms of Freudian psychology are clear evidence of an Oedipus complex. But it’s much worse with Lawson. Understanding a proposition involves grasping its truth conditions. If there are no truth conditions, genuine communication – including the communication of arguments – is impossible. A speaker’s words may evoke feelings in the hearer, but there is nothing to guarantee that anything in the mind of the speaker will be there in the mind of the hearer. The only thing that could guarantee this is the possibility of language serving as a vehicle for meaning. Yet the only remotely successful explanation of how language can serve as such a vehicle makes indispensible reference to the notion of truth: the sense of a sentence is the way in which its truth or falsity is determined.
Nevertheless, I think it’s worth the effort to engage with Lawson. Ideas like his are not just wrong; they are dangerous. Listen to the podcast. Somewhere in the debate section he actually refuses to admit that it’s just false that the Nazis were responding to a genuine threat posed by a parasitic and conspiratorial race. Of course he could say that believing such a thing won’t serve our real interests. But who are we? Surely some people think their interests are very much served by people believing in the Jewish conspiracy. And what if they happen to be stronger than us? Then that will become our guiding metaphor, no more or less valid than the ‘metaphor’ that the Earth is round.
This is toxic stuff, and it’s the business of philosophers to prove it wrong. Wittgenstein once said that philosophers were slum landlords and his job was to put them out of business. Being himself a philosopher, he must have meant that other philosophers are slum landlords. And putting them out of business seems like a legitimate and useful application of philosophy. So. Let’s put Lawson out of business.
What is the reason for believing anything, if nothing is true and everything is just ‘metaphors’? Lawson says we should believe propositions (or embrace ‘metaphors’) when it somehow suits us, on some pragmatic criterion, to do so. This runs into a logical problem, and I don’t think there’s any way out of it. Bertrand Russell’s criticisms of the pragmatic theory of truth very much apply here.
Here is something I believe:
Call this proposition D. Now I believe D, and I think my reason for believing it is that D is true. I’m glad it’s true, since I like dogs, especially when they exist. Entitlement to enclose the world in the metaphor of D is a poor substitute for a big shaggy dog. Still, Lawson wants me to endorse D, not because it’s true, but because it somehow helps me get along in the world. So now call the following proposition G:
Endorsing D helps me to get along in the world.
Why should I endorse G? Again, for Lawson it can’t be because G is true. So it must be that endorsing G helps me get along in the world. So call the next proposition G1:
Endorsing G helps me to get along in the world.
But then another belief, G2, will be needed for endorsing G1, and this will just go on. You will get to a whole infinite series of ‘metaphors’, (G, G1, G2, …), each Gn+1 providing a reason for endorsing Gn. But why should you endorse this series as a whole? Perhaps because you believe that doing so will help you to get along. Call that belief H.
Now is H a member of the series (G, G1, …)? If it is, then it must justify itself, since H is what justifies the series and H is ex hypothesi part of the series. But is it possible for a belief to justify itself? How could it, on Lawson’s criterion? Take the following sentence:
Believing this sentence helps me to get along in the world.
Even if we accept Lawson’s pragmatic criterion of justification, belief in this sentence in no way justifies itself, since that criterion is only satisfied if the sentence is true. And according to Lawson it can’t be true. The case is obviously no better for the following sentence:
Believing that this sentence helps me to get along in the world helps me to get along in the world.
Thus it seems that Lawson’s justificatory schema doesn’t allow any beliefs to justify themselves – any belief needs a further belief to justify it. This is to be expected. In the background of Lawson’s thinking is the notion that we take things to be certain ways rather than them actually being those ways. Applying this to justification, it follows that beliefs can’t be justified in themselves; we must take them to be justified, and this must mean forming other beliefs about them.
Thus, in the above case, H can’t be a member of the series (G, G1, …). It must justify the series from outside. So how do we justify H? Clearly by forming another belief, H1, to the effect that H helps one to get along in the world. Soon we will arrive at another infinite series (H, H1, …) with the same structure as (G, G1, …). This series will require a further belief for its justification, I, which will have to be justified by I1, leading to the series (I, I1, …), and so on.
Now the series of series thus generated might somehow end up closing itself into a circle, if one belief could be simultaneously a member of two series, though I’m not sure how that could work. Or it might simply run to an infinite series of infinite series. In either case, take the whole shebang – the series of series including D and every belief ultimately necessary for justifying D. By ‘ultimately necessary’ I mean necessary given the determination to believe nothing that is not explicitly justified – thus we cannot justify D by simply assuming G. How do we justify belief in the whole shebang? By forming a belief, of course, about how the whole shebang helps us to get by. But that belief can’t be within the whole shebang, since we’ve seen that beliefs can’t justify themselves on Lawson’s criterion. Yet nor can it be outside of it, for then the whole shebang isn’t the whole shebang – it doesn’t contain everything ultimately required for justifying D, since while it is required for justifying D it doesn’t contain what is required to justify itself. The only viable conclusion seems to be that there is no whole shebang. D is not ultimately justifiable at all.
The structure of this argument is, of course, similar to that of the third of St. Thomas’ ‘Five Ways’ of proving God’s existence. This was based on the idea of an infinite chain of dependence among contingent beings. The options are to accept that there is at least one non-contingent being or that there is some absolute contingency about the universe as a whole. Following the analogy to Lawson’s argument, these options come to much the same thing: either there is at least one unjustified belief among those required to justify D or it is the case that what, as a whole, justifies D is not itself justified. Either way D is not ultimately justified.
But what applies to D applies to most beliefs of similar form. This includes belief in most of Lawson’s own statements. So if Lawson is right, there is no reason to believe that he is right, and indeed the supposition that there is such a reason leads to paradox. Good – let’s not believe him then. So what should we believe?
I propose that a good reason for believing something is that it is true. I admit we can never know with certainty that something is true. But I’m pretty sure, for example, that D is true, and if asked why I believe D my reply will point to that certainty. I give the grounds for my certainty, and in doing so I of course assert other beliefs. If D turns out to be false – if all the dogs I see are just hallucinations or something – then I don’t really have a reason for believing D, I just think that I do. But there is a big difference between a theory according to which I might not actually have a reason to believe something when I think I have a reason to believe it (surely anybody has to accept that) and a theory according to which I can’t possibly have a reason for believing anything, including the theory that led me to think that in the first place. (Lawson’s other points, that objects can be described in an infinite number of ways, are just nothing to the point here, since if he is right we can’t have any reason at all to believe any of this.)
Perhaps what Lawson really means to say that there is no difference between asserting something and claiming that it’s useful to believe it. But then his claims become relatively banal. If saying ‘it is useful to believe that the world is thus and so’ just means the same thing as saying that ‘the world is thus and so’, then Lawson isn’t really challenging realism or the notion of truth at all. He’s just saying how things are, in a flowery language including the utterly redundant words ‘it is useful to believe…’. We understand what he is saying only because we understand that ‘it is useful to believe’ means simply ‘it is true’, and we know what that means.
To explain the last point, consider that if we don’t understand that assertions are aimed at truth it is unclear what we should do with any assertion upon reading or hearing it. Any explanation of what we should do, such as the one Lawson offers, will not help, because it will consist of assertions that, again, we don’t know what to do with. If we did not understand that to claim that P is to claim P as true, we would have no hope of understanding the claim at all. The act of stating that P would be, to adapt Anscombe’s example, like that of pointing at a table and a chair and saying: ‘Suppose that this is Socrates, and that is Plato.’ The likely reaction in that case would be: ‘Yes, and?’ Only when we knew what the table-and-chair model was meant to represent – that of which it was meant to be a true picture – would we have any hope of knowing what the speaker was getting at. But imagine if the speaker replied: ‘The clock is the condition of Socrates being the chair and Plato being the table, and the time it shows is the condition of you getting along better in the world.’ This is more gibberish, unless we know what that model is supposed to truly depict. Without an implicit grasp of truth, we have no hope of making any sense out of anything the speaker says. If we try to do without the concept of truth, as Lawson suggests, we are in the same predicament of utter incomprehension as somebody would be in if she had no idea that assertions were attempts to say what is true.
Lawson’s theory should, even by its own lights, be rejected. Assertions can be true or false. Communication is thus possible, since the concepts of meaning and truth cannot really be separated. This might seem a tawdry truth compared with the wild excesses of Lawson’s book. But in fact there is something remarkable and almost miraculous about it. Trees and birdsongs are rich in meaning and convey all sorts of things. But they don’t carry distinct meanings the way that sentences do. A birdsong does not distinguish between ‘had we but world enough and time, this coyness lady were no crime’ and ‘I am currently sexually available’. But language does. A non-truth-apt ‘metaphor’ is no more a vehicle for the communication of distinct meaning than a tree. Without truth, we could commune with each other, but we couldn’t communicate. Dogs exist, and I can also say that dogs exist. Both are wonderful facts.