TWO KINDS OF POSSIBILITY

I first consider two pairs of propositions: their implications, and the kind of observation and argument by which their truth or their falsity would normally be established, if they were challenged.

The first pair

(a1) 'It will not happen now'

(a2) 'It cannot happen now'

The second pair

(b1) 'He will not do it now'

(b2) 'He cannot do it now'

To bring out the contrast between the pairs, one might choose more specific propositions as examples of the two types:

(A1) The gas will not escape now'

(A2) The gas cannot escape now'

(B1) 'Jones will not escape now'

(B2) 'Jones cannot escape now'

I am concerned with the difference between "will not' and the 'cannot' in the two cases: I shall argue that it is a different difference.

Consider the first pair first: to make the transition from 'It will not happen now' to 'It cannot happen now,' from the weaker to the stronger statement, is, at the very least, to claim that a certain type of evidence, or reason for believing, or source of knowledge, that the thing will not happen is available: not only will it not happen, but it is impossible that it should. Someone might agree that the thing won't happen, but he might at the same time dispute the stronger statement that it cannot happen, that it is impossible that it should. 'I agree with you,' he says, 'that in fact it will not happen now; but, all the same, it might; it is not altogether impossible. Your statement is too strong.'

We see that the contradictory of 'It cannot happen here and now' is not only 'It can, or could, happen here and now,' but also 'It may, or it might, happen here.' These are alternative ways of representing the possibility of its happening, when someone has asserted the impossibility. When the two disputants agree about the future course of events — that in fact the thing will not happen — they may still disagree about the possibility of its happening. If it is certain that it will not happen, there is no possibility that it will.

I do not myself know of any entirely adequate account of the logic of statements of the form 'It is certain that so-and-so will not happen.' But at least it is clear that he who asserts that it is certain that it will not happen here, or alternatively asserts that it cannot happen here, must show that there are very strong grounds for believing, or that there is some sure source of knowledge, that it will not happen here, when the truth of his statement is challenged. A man may say 'It will not happen here,' (e.g. This horse will not win the race'), and his statement may be true, and the utterance may be an entirely justifiable utterance in the circumstances, even though he was not in a position to support the statement with good grounds, or good evidence, or by citing a source of knowledge. But 'It cannot happen here,' which denies that it might happen, asserts, at the very least, that there are very good reasons for believing that it won't. Perhaps it is sometimes, or even generally, used to assert, or imply, much more than this: e.g. that there is some well-established natural law covering this happening; perhaps it asserts, or implies, that there is some cause, or sufficient condition, of its not happening. However this may be, at least it asserts that there are grounds for believing that it will not happen, which are strong enough to justify the assertion that it cannot happen. If no such grounds exist, the statement 'It can't happen here' is unsustainable and incorrect, even if the statement 'It won't happen here' is acceptable; for an objector could say 'Well, I admit that it didn't happen: but still it might have happened, and you were quite wrong when you said that it couldn't happen.'

We are familiar with the series: (1) 'It might happen here,' which might also be expressed as 'It could happen here;' (2) 'It is unlikely to happen here,' but 'It might' or 'It could,' and (3) 'It is certain that it won't happen here,' which might be expressed as 'It can't.'

Turn now to the second pair of the original propositions: either to the more general 'He won't do it now' and 'He can't do it now': or to the more specific 'He won't escape' and 'He can't escape.' The transition from 'He won't' to 'He can't' may here be altogether different; for when we say that he can't do it, we are not ordinarily saying that it can't be the case that he will do it. Given an appropriate verb of action, we would not ordinarily be denying that he might do it, that there is a possibility that he will. We are rather saying that he is not able to do it, that he lacks the means, or the authority, or the opportunity, to do it. The transition from 'He won't escape now' to 'He can't escape now' would not ordinarily be a transition from a weaker to a stronger statement about a future happening — about a possible eventuality, his escape. The conjunction of 'can' with a verb representing voluntary action would ordinarily preclude this interpretation; only in the past tense 'He can't have escaped' would ordinarily have as its contradictory 'He might have escaped,' where this is equivalent to 'It is not impossible that he did.' 'He could have escaped,' is ambiguous when taken out of context. It might be intended to be equivalent to 'It could be the case that he escaped', and 'It is not impossible that he did,' and 'He might have escaped;' or it might be intended to be equivalent to 'He had the ability, the means, and the opportunity, to escape.'

'He can't escape' may be informative as giving a reason or explanation why in fact he won't. He won't escape, because he can't. It is informative as a reason or explanation, because it may be the case that he would escape if he could, but he can't. Contrast this with the other pair: of the gas it is not similarly informative to explain the fact that it won't escape now by saying that it can't. It is not clear what it would mean to say of a gas that it would escape if it could. It is just conceivable that some non-anthropomorphic, literal sense might be given to this; for example, that it is a gas which has a tendency to escape from any container, but that now its escape is obstructed. But 'It would, if it could, but it can't' is not a literal and natural, and even less a scientific way, of speaking of the behaviour of a gas; more natural would be 'It would escape if it wasn't for the lead in the container, which makes it impossible,' Here 'It's impossible' does not actually give the reason why it won't escape; it indicates only that there is a specific cause of its not escaping, in spite of its tendency to escape. If I am asked to give you a reason for believing that it won't escape, I may explain to you why it can't. But I cannot quote its inability to escape as a reason for its not escaping. On the other hand, 'He would if he could, but he can't' is a very natural, and informative, way of speaking of the behaviour of a person. We now know in very general terms why he won't escape: he lacks the ability, or the means, or the opportunity to escape.

'He can't do it' (the governor is speaking about the prospect of his escape from prison) gives one of two possible...