When Aristotle’s _Constitution of Athens_ was brought to light and printed, I made some observations on it in the _Athenæum_, 7 February 1891, a few days after it came out. The editor had cited the passage in chapter 54 which shows that it was written after 329 B.C., but had overlooked a passage in chapter 46 which shows that it was written before 325 B.C.; and I cited this passage, and thereby limited the date. Amongst other things I proposed to read _architheô_ ... for _archiereô_ ... in chapter 56, as I happened to remember that _architheôrois_ came in the same context in an inscription of that period. I think my reading was right--all subsequent editions have adopted it--but I should very much like to know if the word went wrong in copying or dictation, or was a slip of Aristotle’s own pen. I fancy that the Greek and Latin authors wrote the wrong word now and then, and never noticed it. That is not the view of textual critics and editors: they ascribe all errors to the men who copied out the manuscripts. But this limits them to errors that might arise in copying, and thus restricts the choice of emendations far too much. Take such an emendation as _Isara_ for _Arar_ in Livy, xxi. 31. This makes Livy say that the river was the Isère, not the Saône; but the context requires him to say it was the Durance, otherwise he would be saying ‘right’ instead of ‘left’ a few lines further on. A copyist might easily write _arar_ for _isara_, so this emendation is accepted, although it does not suit. Such emendations are deceitful things. In this case they make Livy say the Isère, and make Polybios say it also, iii. 49, though he says something else; and then Members of the Alpine Club go saying that the river must have been the Isère, since Livy and Polybios agree in saying that it was. Other folk may say it does not matter what the river was; but that is a reason for leaving the whole thing alone, not for getting it wrong. If you take it up at all, you should not risk the sort of snubbing that Westbury gave the herald after cross-examination--“Go away, you silly man: you don’t even understand your own silly science.” My father used to tell me of Westbury’s methods at the Bar. A judge would put a question that seemed to be a poser. Westbury would pause, and then he would not only answer it convincingly, but would put the point with such lucidity that you could not understand how anyone had failed to see it. And the judge would turn quite red, feeling that he had asked a foolish question, and people in court would titter and guffaw, though half of them could never have answered it at all.