Adding to his previous explanation of the two fallacies (the
historic estimate and the
personal estimate), Arnold explains that the historical estimate is especially likely to be a problem when readers confront ancient works.
Matthew Arnold acknowledges that there is not a great deal of harm in overrating ancient poets, but doing so degrades the language used to describe poetry and to make distinctions. For example, Arnold points out, a French critic (
M. Vitet), in his determination to elevate the French epic poem
The Song of Roland to the highest rank of poetry, uses language that is only appropriate for
Homer. To show his readers the travesty of describing
The Song of Roland in terms reserved for Homer, Arnold produces excerpts from the former and the latter. The difference in quality, he insists, is stark: “[With Homer] we are here in another world, another order of poetry altogether.”