W. K. Clifford is often quoted for his asseveration that "it is wrong always, everywhere, and for anyone, to believe anything upon insufficient evidence." Now one of my firmest beliefs is that I am an actual individual, not a merely possible individual. A second is my belief that while there is an infinity of possible worlds, there is exactly one actual world and that this world of me and my world mates is the world that happens to be actual. (Think of the actual world as the total way things are, and of a merely possible world as a total way things could have been. For a quick and dirty primer, see Some Theses on Possible Worlds.)
But not only do I have insufficient evidence for these two beliefs, it looks as if I have no evidence at all. And yet I feel wholly entitled to my acceptance of them and in breach of no plausible ethics of belief, assuming there is such a subject as the ethics of belief.
Consider the following argument that I adapt from D. M. Armstrong, who borrowed it from Donald C. Williams:
1. Exactly one of the infinity of possible worlds is actual.
2. This world of me and my mates is a possible world.
Therefore, very probably,
3. This world of me and my mates is merely possible.
This is an inductive argument, but a very strong one. While it does not necessitate its conclusion, it renders the conclusion exceedingly likely. For if there is an infinity of worlds, how likely is it that mine is the lucky one?
And yet the conclusion is absurd, or to be precise: manifestly false. Is it not perfectly obvious that this world of ours and everything in it is actual? I am convinced that I am actual, and that all this stuff I am interacting with is actual. I am sitting in an actual chair in an actual room which is lit by an actual sun, etc.
But how do I know this? What is my evidence? There are no facts known to me that are better known than the fact that I am actual (that I actually exist). So my evidence cannot consist of other facts. Is it self-evident that I am actual? You could say this, but how do I know, given the above argument, that my actuality is objectively self-evident as opposed to merely subjectively self-evident? Subjective self-evidence is epistemically worthless, while objective self-evidence is not to be had in the teeth of the above argument. No doubt I seem to myself to be actual. But that subjective seeming does not get the length of objective self-evidence. I now argue as follows:
4. If it is wrong to believe anything on insufficient evidence, then it is wrong to believe anything on no evidence.
5. I have no evidence that me and my world are actual.
6. It is not wrong to believe what is obviously true.
7. It is obviously true that I am actual.
Therefore, contra Clifford,
8. There are some things it is not wrong to believe on insufficient evidence.
This is not a compelling argument, but it is a very powerful one. Not compelling because the Cliffordian extremist could bite the bullet by denying (7). He might say that the ethics of belief enjoins us to suspend belief on the question whether one is actual.
Now this is psychologically impossible, for me anyway. But apart from this impossibility, it is surely better known that I am actual than that Clifford's extreme thesis is true.
There are other obvious problems with the thesis. Any tyro in philosophy should see right away that it is self-vitiating. If it is wrong to believe anything on insufficient evidence, then it is wrong to believe Clifford's thesis on insufficient evidence. But what conceivable evidence could one have for it? None that I can see. It is not only a normative claim, but one stuffed with universal quantifiers. Good luck! If you say that the thesis needn't be taken as applying to itself, then other problems will arise that you can work out for yourself. Why do I have to do all the thinking?