it would be better to relieve them of history altogether, and place the subject as a whole in the hands of a distinct examiner.  I would still allow merit for a literary examination in French, German, and Italian, but would strike off the languages, and let the candidate get up the literature as he chose.  The basis of a candidate’s literary knowledge, and his first introduction to literature, ought to be his own language:  but he may extend his discrimination and his power by other literatures, either in translations or in originals, as he pleases; still the examination, as before, should test the discrimination and the power, and not the vocabulary of the languages themselves.

In order to do full justice to classical antiquity, I would allow markings at the rate of 500 for Political Institutions and History, and 250 for Literature.  Some day this will be thought too much; but political philosophy or sociology may become more systematic than at present, and history questions will then take a different form.

In like manner, I would abolish the language-examination in modern languages, and give 250 marks for the literature of each of the three modern languages—­French, German, Italian.  The history would be taken as Modern History, with an adequate total value.

The objections to this proposal will mainly resolve themselves into its revolutionary character.  The remark will at once be made that the classical languages would cease to be taught, and even the modern languages discouraged.  The meaning of this I take to be, that, if such teaching is judged solely by its fruits, it must necessarily be condemned.

The only way to fence this unpalatable conclusion, is to maintain that the results could not be fully tested in an examination as suggested.  Some of these are so fine, impalpable, and spiritual in their texture, that they cannot be seized by any questions that can be put; and would be dropped out if the present system were changed.  But results so untraceable cannot be proved to exist at all.

[LANGUAGE QUESTIONS TAKE THE PLACE OF THE SUBJECT.]

So far from the results being missed by disusing the exercises of translation, one might contend that they would only begin to be appreciated fairly when the whole stress of the examination is put upon them.  If an examiner sets a paper in Roman Law, containing long Latin extracts to be translated, he is starving the examination in Law by substituting for it an examination in Latin.  Whatever knowledge of Latin terminology is necessary to the knowledge of Law should be required, and no more.  So, it is not an examination in Aristotle to require long translations from the Greek; only by dispensing with all this, does the main subject receive proper attention.