[143] for instance about biblical literalism^a, taking suddenly a biblical word literally, while in the totality of his thinking literalism would be excluded---then you have to ask: now what does that mean? This cannot be understood in terms of his whole system, so it must be understood in terms of remnants of the past, or of a special historical situation in which he was forced into such a thing---e.g., the Marburg Confession [discussion?]. But be very cautious in this method; always be prejudiced towards each other, and toward any figure in the past, that he knows what he is talking about and that he MEANS it, and that it has some place in the whole of his thinking. And refute him in terms of arguments, but not in terms of sociological or psychological analysis, except in extreme cases where it is obvious that you cannot deal with him in terms of rational discussions. Of course, nothing in the world is absolutely obvious, so there is always the possibility that you make a mistake, but in any case, this is the practical, the pragmatic criterion which I give you in all such kinds of discussions. For this reason, I suppose it is not entirely useless to listen now to the next lecture, even after everybody completely analyzed [the situation] sociologically and psychologically. Historical truth^b has a quite different character from scientific truth^c. History^d reports unique events. Every historical event is once, unique, and never can be repeated. Every physical fact can be repeated as often as we can produce the external conditions. You can never produce the external conditions for a historical constellation. It has gone forever. Therefore you can [sic.] test historical truth by repetition. You can test it certainly by OTHER methods. Scientific processes, you can test again and again, and it is one of the great advantages of the sciences that every special scientist