[Sidenote: _Its merits_] A first attempt of its sort (it cannot be said here with too much frankness and conviction) can even less than any other book be faultless: and it is almost a sufficient proof of Hallam’s greatness that his faults are not greater. Some things, indeed, that seem to me faults may not even seem to be so at all to others. He was aware that he must “pass over or partially touch” some departments of at any rate so-called literature; but his preference or rejection may seem somewhat remarkable. Few will quarrel, at least from my point of view, with the very large space given to mere “scholars,” but it is surely strange that a historian should have thought History of secondary importance, while according ample space not only to Philosophy and Theology, but even to Anatomy and Mathematics. A more serious and a more indisputable blemish is the scanty and second-hand character of his account of mediæval literature, which he might almost as well have omitted altogether. It cannot be too peremptorily laid down that second-hand accounts of literature are absolutely devoid of any value whatever:—the best and latest authorities become equally “not evidence” with the stalest and worst. [Sidenote: _and defects._] Hallam was aware of this principle to some extent, and he almost states it, though of course in his own more measured way, and with reference to quotation mainly, in his preface. But his first chapter is really nothing but a tissue of references to Herder and Eichhorn, Meiners and Fleury, with original remarks which do not console us. The account of Boethius at the very beginning is a pretty piece of rhetoric, but, as the Germans would say, not in the least “ingoing.” It is a horrible heresy to say[559] that “It is sufficient to look at any extracts” from the Dark Ages “to see the justice of this censure,” for no collection of extracts will justify the _formation_ of any critical opinion whatsoever, though it may support, or at least illustrate, one formed from reading whole works. [Sidenote: _In general distribution and treatment._] Further, in a note of Hallam’s[560] I think may be found the origin of Mr Arnold’s too exclusive preference for “the best and principal” things and his disparagement of the historic estimate, though I trust that Mr Arnold[561] would not have shared Hallam’s contempt, equally superfine and superficial, for the “barbarous Latin” of the Dark Ages. Finally, it is difficult to conceive a more inadequate reference to one of the most epoch-making of European poems (which is at the same time in its earlier part one of not the least charming) than the words “A very celebrated poem, the _Roman de la Rose_, had introduced an unfortunate taste for allegory in verse, from which France did not extricate herself for several generations.” It is all the worse because nothing in it is positively untrue.