so in the hands of a translator.  I have accordingly on various occasions aimed at epigram and pungency when there was nothing epigrammatic or pungent in the Latin, in full confidence that any trifling additions which may be made in this way to the general sum of liveliness will be far more than compensated by the heavy outgoings which must of necessity be the lot of every translator, and more particularly of myself. [Footnote:  Cowper himself has some remarks bearing on this point:  “That is epigrammatic and witty in Latin which would be perfectly insipid in English; and a translator of Bourne would frequently find himself obliged to supply what is called the turn, which is in fact the most difficult and the most expensive part of the whole composition, and could not perhaps, in many instances, be done with any tolerable success.  If a Latin poem is neat, elegant and musical, it is enough; but English readers are not so easily satisfied.  To quote myself, you will find, in comparing the Jackdaw with the original, that I was obliged to sharpen a point which, though smart enough in the Latin, would in English have appeared as plain and as blunt as the tag of a lace.” —­Letter to Unwin, May 23, 1781 (Southey’s Cowper, ed. 1836, vol. iv. p. 97).] All translation, as has been pointed out over and over again, must proceed more or less on the principle of compensation; a translator who is conscious of having lost ground in one place is not to blame if he tries to recover it in another, so that he does not consciously depart from what he believes to be the spirit of the original:  the question he has to ask himself is not so much whether he has conformed to the requirements of this or that line, most important as such conformity is where it can be realized without a sacrifice of higher things, as whether he has conformed to the requirements of the whole sentence, or even of the whole paragraph; whether the general effect produced by all the combined elements in the English lines answers in any degree to that produced by the Latin.  Often and often, while engaged on this translation, I have been reminded of Johnson’s words in his Life of Dryden:  “It is not by comparing line with line that the merit of works is to be estimated, but by their general effects and ultimate result.  It is easy to note a weak line and write one more vigorous in its place, to find a happiness of expression in the original and transplant it by force into the version; but what is given to the parts may be subducted from the whole, and the reader may be weary, though the critic may commend.  That book is good in vain which the reader throws away.” [Footnote:  Compare his parallel between Pitt’s and Dryden’s Aeneid in his Life of Pitt.] I will only add that if these remarks are true of translation in general, they apply with special force to the translation of an original like the present, where the Latin is nothing if it is not idiomatic, and the English in consequence, if it is to be anything, must be idiomatic also.