Arguments move from premises to conclusions. The premises state things taken temporally for granted; if the argument works, the premises provide grounds for affirming the conclusion. A valid deductive argument is one in which the premises necessitate, that is, entail, the conclusion. (It would involve a self-contradiction to assert the premises but deny the conclusion.) What I shall call a ‘correct’ inductive argument is one in which the premises in some degree probabilify the conclusion, but do not necessitate it. More precisely, in what I shall call a correct P-inductive argument the premises make the conclusion probable (i.e. more probable than not); in what I shall call a correct C-inductive argument, the premises add to the probability of the conclusion (i.e. confirm it, make it more probable than it was; but do not necessarily make it overall probable). Arguments only show their conclusions to be true if they start from true premises; arguments of the above types which work (i.e. are valid or correct) and do start from such premises I will call sound arguments. Arguments are only of use to show to an individual that the conclusion is true if he already knows the premises to be true. Most of what I shall have to say today concerns arguments with respect to which there is no doubt that the premises are true.