However that may be, the question, in spite of so many attempts, still remains unanswered. If any one wishes to give a scientific proof of primitive communism, these are the conditions on which he may perhaps succeed: 1. He must find definite and exact authorities; which he must translate, not approximately, but with absolute correctness, according to the literal signification of the words. 2. He must abstain from adducing facts which are comparatively modern in support of an institution which he ascribes to the beginning of things, as has been done in the case of the German mark, the island of Java and the Russian _mir_. 3. He must not content himself with collecting a few isolated facts which may be exceptional; but he must study phenomena which are general, normal, and far-spreading; of these he will find the evidence principally in legal records, and to a small extent in early religious customs. 4. He will be careful not to confuse agrarian communism with family ownership, which may in time become village ownership without ceasing to be a real proprietorship. 5. He will not mistake undivided tenancies on a domain belonging to a proprietor for community in land. The fact that _villani_, who were not the owners of any land at all, often cultivated the soil in common for a lord, or annually divided it amongst themselves, has no connection with agrarian communism, and is in fact directly opposed to it. 6. He will be careful not to confuse the question by introducing village commons, unless he has first of all succeeded in proving that such commons are derived from a primitive communism. This has never yet been proved, and all that has hitherto been ascertained about commons is that they are an appendage of private property. On these conditions alone can the work be done scientifically; short of this the only result will be a confused picture of the fancy. If any one, after taking all these precautions against gross error, discovers a body of facts and evidence in support of a theory of communism, he will have settled the question historically. Till then, do not invoke history in its favour. Present your theory as an abstract idea which may be valuable, but with which history has nothing to do. Let us not have sham learning. In saying this I have at heart the interests of historical science. There is danger lest, from love of a theory, a whole series of errors should be forcibly thrust into history. What I fear is not the theory itself; it will not affect the progress of human events; but it is the method employed to secure its acceptance. I distrust this pretended application of learning, this practice of forcing documents to say the very opposite of what they really say, this superficial habit of talking about all the nations of the world without having studied a single one. Never have “original authorities” been so much lauded as to-day; never have they been used with so much levity. THE END.