Whether literary history is conceived as a reference archive or as a narrative, there remains the question of what elements of a literary field, such as works, authors, modes, and genres, are to be archived or narrativized. What role does a notion such as masterpiece play? Scholars have elaborated rich structures of significance to motivate their research into works from the past that no present reader would otherwise willingly attend to, but without some such notion as masterpiece, how can a student aspire to read distant, difficult work? What alternative categories of importance, or value, can we propose? I do not think the old, positivistic category of “representative selections” any longer wins much assent.(Arac, "What Good Can Literary History Do?" American Literary History 20.1 (Spring/Summer 2008): 3)
I, for one, do not find "representative selections" a very useful term, but Arac's question allows me a moment to reflect on the degree to which I still find something like "representative selection" or "masterpiece" to be a "motivat[ing]" influence in the selection of the texts I analyze, or the degree to which I believe that it might cause "a student [to] aspire to read distant, difficult work."
My initial reaction is mostly skeptical on the first count, but my skepticism is somewhat more qualified on the second count. I don't teach yet, so I can't speak to that perspective, but as an undergraduate I remember choosing literature classes more by period than by the individual selections, although I can remember course catalog entries clearly trying to play the "masterpiece" angle in advertising for a course on Tolstoy, or on Mann, or on Ulysses. I don't know how effective these appeals were, or if they were necessary.
I would be interested in other people's experiences—both as teachers and as students. To what degree is the idea of "masterpiece" used as a motivator toward venturing into "distant, difficult work," and to what degree is it effective? And then, the other question Arac asks, "What alternative categories of importance, or value, can we propose?" And, finally, do we (pragmatically) need any?