A. MEINONG, Ueber d. Erfahrungsgrundlagen unseres Wissens. 413

in section iv., seems to me insufficiently explained ; at any rate I have failed to grasp what is meant by it.) A judgment of the sort considered is not to be called a perception unless it is true, i.e., unless the object exists. Now, if we are to have any reason for believing in our judgments rather than disbelieving, two things are necessary : (1) that there should be a kind of judgments in whose nature it lies to be true ; (2) that by means of judgments of this kind we should be able to recognise when judgments are of this cind. These two conditions are more or less fulfilled by self-evi- ent judgments. It is self-evident that a self-eyident judgment cannot be false, which is the first requirement ; but it is not always self-evident whether a judgment is self-evident or not, so that the second requirement is only partially fulfilled. (May it not be doubted whether the first is fulfilled, if self-evident is taken in a jurely psychological sense ? And if it is not, there is danger of tautology.) If perception is to be knowledge, it must be self- evident knowledge. Hence we reach the conclusion (pp. 35-36) : A perception is "an immediately evident affirmative judgment )f existence concerning a present thing, based on a perceptive presentation (Wahrnehmungsvorstellung) (or a suitable substitute one) ". Here a "perceptive presentation" is a presentation whose object is judged to exist in a perception, or in any judg- lent which is psychologically like a perception, i.e., like, except at lost as regards its truth or its self-evidence. The second section introduces the word aspection (Asnekf) for rhat is just like a perception except at most as regards truth and ilf-evidence. Thus the question of the " trustworthiness of per- jption " becomes the question as to when and how far aspections ire perceptions, since perceptions have been defined as true. The inswer to this generally depends upon whether the object exists. ?hus relations cannot be perceived, because they cannot exist ; )lours might exist, but there are reasons for doubting whether ley do exist, and therefore whether aspections having colours as )bjects are perceptions. The usual reasons for doubting the exist- ance of both primary and secondary qualities are reviewed, and the Drovisional conclusion is reached that it is doubtful whether there such a thing as external perception. The third section, on internal perception, decides that in this ise there can be no doubt that some aspections are perceptions, lough here also observation is not infallible. When we hear )unds or see sights we can be quite certain that we are hearing >r seeing them. A man who has a toothache is quite sure that has it. (Prof. Meinong denies self-evidence to hallucinations, ind would, I suppose, exclude hysterical pains on this ground. Jut if self-evidence is defined without explicitly including truth, is hard to see any psychological justification for denying it in lese cases.) An element of uncertainty is introduced, even with iternal aspections, by the fact that they are never quite simul-

ineous with their objects. That they are not always simultaneous