Scepticism

1. The philosophical problem of scepticism

Most contemporary discussions of scepticism have focused on scepticism concerning the external world. We can use this type of scepticism to illustrate the broader philosophical problem, as many of the arguments we consider can be applied mutatis mutandis to other types of scepticism.

One type of scepticism denies that we know anything about the external world. The view is not simply that, for example, by gathering more evidence we could come to know. Rather, it is that we are unable to attain knowledge. On the plausible assumption that knowledge entails justified belief, scepticism concerning knowledge follows from scepticism concerning justified belief – the view that justified belief about the external world is unattainable.

Scepticism is of philosophical interest because there appear to be very strong arguments that support it. This presents us with the problem of how to respond to these arguments. One way would be to accept their conclusion. Of course, very few philosophers are willing to do this. There are very few actual sceptics. So the problem of scepticism is how to refute or in some way neutralize or deflate the force of these arguments.

In the history of philosophy, some sceptical arguments have been based on the unreliability or relativity of our senses (see Pyrrhonism), or upon the inability of reason to produce non-question begging arguments for our beliefs (see Hume, D. §2). Nearly all sceptical arguments exploit sceptical hypotheses or alternatives. Sceptical alternatives suppose that the world is very different from what we would normally believe on the basis of our sensory evidence. This entails that our sensory evidence is radically misleading. More precisely, suppose we claim to know a proposition q on the basis of evidence e. Let (proposition) h be an alternative to q just in case h is incompatible with q (q and h cannot both be true). Then h is a sceptical alternative to q provided h is an alternative to q compatible with e. An alternative of this kind has sceptical force precisely because it is compatible with the evidence we claim gives us knowledge of q. For example, ordinarily, I would claim to know on the basis of my visual evidence that I am currently looking at my computer monitor. One sceptical alternative, introduced by Descartes (1641), is that the world of familiar objects does not exist and that I am being deceived into thinking it does by a powerful demon. The demon causes me to have just the sensory experiences I would have if the world of familiar objects existed (see Descartes, R. §4). According to a modern version of this alternative, I am a brain-in-a-vat being artificially stimulated to have all the experiences I would have if I had a body and interacted, in the normal way, with the world of familiar objects. These alternatives are incompatible with what I claim to know about the familiar world around me since according to those alternatives, that world does not exist. Moreover, since these alternatives entail that it appears to me as if that world exists, they are compatible with my evidence.

Sceptical alternatives provide the basis for very powerful sceptical arguments. Exactly how they do this is a matter of some controversy. The quickest route to scepticism is through what I will call the entailment principle:

Since a sceptical alternative is, by definition, a proposition incompatible with q but compatible with e, it follows from the mere existence of sceptical alternatives of the kind we have been considering that we do not know those empirical propositions we ordinarily claim to know. But, this argument is only as good as the entailment principle. Should we accept this principle? In effect, the principle says I can know p only if my evidence precludes the possibility of error. Though many philosophers concede that this principle has considerable intuitive force, most have thought, in the end, that it should be rejected. This position is sometimes called fallibilism (see Commonsensism §§1–2; Fallibilism). Of course, few philosophers believe that scepticism should be avoided at all costs. But when given a choice between scepticism and fallibilism, most philosophers opt for fallibilism (at the expense of the entailment principle).

Does fallibilism beg the question against scepticism? After all, precisely what the sceptic claims is that the existence of alternatives consistent with our evidence undermines our claims to know. Fallibilists merely respond that the alternatives the sceptic has invoked do not undermine our knowledge claims: that is, we can know even when there are such alternatives. Since this is the point at issue, fallibilists seem to need an argument in support of this crucial claim.

Here, fallibilists can appeal to our strong intuition that in many cases we do know things, despite the existence of sceptical alternatives. And it is not clear that the sceptic can undermine those intuitions except by appealing to the entailment principle – which is itself undermined by those very intuitions. Thus neither side of the debate may be able to defend its position without begging the question.

Unfortunately scepticism is not so easily dispatched. The sceptic can turn the appeal to our ordinary intuitions against fallibilism. For some of those intuitions can provide the basis for a new sceptical argument. This argument begins by claiming, quite plausibly, that whatever else we may say about the significance of sceptical alternatives, we cannot claim, plausibly, to know they are false. For example, we cannot claim, plausibly, to know that we are not brains-in-a-vat being artificially stimulated to have exactly the same experience we would have as normal human beings. None of our evidence counts against this hypothesis since if it were true, we would have precisely that evidence.

But how, exactly, does this permit the sceptic to conclude we do not know the propositions we ordinarily claim to know? At this point, the sceptic appeals to a very intuitive principle that is weaker than the entailment principle. This principle says that the set of known (by S) propositions is closed under known (by S) entailment:

While one could quibble with some details about this principle, it (or something very much like it) seems compelling (see Deductive closure principle). From this principle and the claim that we fail to know sceptical alternatives are false, it follows that we fail to know the propositions we ordinarily claim to know (since we know those propositions entail the falsity of sceptical alternatives).