As a mode of assurance, expectation is clearly marked off from memory, and is not explainable by means of this. It is a fundamentally distinct kind of conviction. So far as we are capable of analyzing it, we may say that its peculiarity is its essentially active character. To expect a thing is to have stirred the active impulses, including the powers of attention; it is to be on the alert for it, to have the attention already focussed for it, and to begin to rehearse the actions which the actual happening of the event--for example, the approach of a welcome object--would excite. It thus stands in marked contrast to memory, which is a passive attitude of mind, becoming active only when it gives rise to the expectation of a recurrence of the event.[138] And now let us pass to the question whether expectation ever takes the form of immediate knowledge. It may, perhaps, be objected that the anticipation of something future cannot be knowledge at all in the sense in which the perception of something present or the recollection of something past is knowledge. But this objection, when examined closely, appears to be frivolous. Because the future fact has not yet come into the sphere of actual existence, it is none the less the object of a perfect assurance.[139] But, even if it is conceded that expectation is knowledge, the objection may still be urged that it cannot be immediate, since it is the very nature of expectation to ground itself on memory. I have already hinted that this is not the case, and I shall now try to show that what is called expectation covers much that is indistinguishable from immediate intuitive certainty, and consequently offers room for an illusory form of error. Let us set out with the simplest kind of expectation, the anticipation of something about to happen within the region of our personal experience, and similar to what has happened before. And let the coming of the event be first of all suggested by some present external fact or sign. Suppose, for example, that the sky is heavy, the air sultry, and that I have a bad headache; I confidently anticipate a thunderstorm. It would commonly be said that such an expectation is a kind of inference from the past. I remember that these appearances have been followed by a thunderstorm very often, and I infer that they will in this new case be so followed.