When was academic quality?

A professor of mine at UC Irvine once told me that when he was working on his doctoral dissertation in the 1960s, he obediently went back to the earlier scholarship on his topic--the 20s, the 30s, the 40s, etc.--and read through it.  At a certain point, however, he arrived at a welcome or unwelcome conclusion, whichever way one chose to look at it: the older scholarship was terrible.

Even if one doesn't wish to conceal decades of material under an uncomfortable woolen blanket of condemnation, it soon becomes hard to avoid noticing key differences in approach that can make assessment difficult.  In English, for example, you have a lot of hand-wavy aesthetic judgments that, pace those who want to bring aesthetic judgments back, are frequently pretty useless.  When Joseph Ellis Baker tells us in 1932 that Lady Georgiana Fullerton's Ellen Middleton is "false, inflated," but William Sewell's Hawkstone "still might afford amusement to the average reader" (The Novel and the Oxford Movement, 19), the twenty-first century scholar is left with any number of questions, starting with the question of Baker's taste.  Where on earth does one locate the average reader who wants to slog through the prolix Hawkstone? And how does one imagine such a person into being? More to the point, Baker's approach in this monograph, which relies heavily on plot summary and coverage but has no analysis to speak of, is typical of literary history from this period; ironically, the emphasis on extensive coverage of a constrained topic, borrowed from the German research model, is why the book is still extremely useful as a bibliographical resource, even though what Baker has to say frequently has not aged well.  Now, readers expect a sharper historical narrative with a well-defined theoretical focus (although plot summary remains unavoidable). 

This is a bit of a workaround to Rob Warren's recent post about his stint as editor of Sociology of Education, where he discovered that lots of the submissions were kind of, well, not good.  When I was at Modern Philology for a year in the mid-90s, I also had the eye-opening experience of realizing that many academics, even those at tony R1 institutions, were not producing A+ work.  (That's before we got to the book reviews, which, as far as I could tell, were often written by academics who expected the editors to wallop their prose into shape.)  My own experience of reading earlier published work, as well as reading submissions from two decades ago, leaves me with this question: how could we know that most or many submissions in any field--not the publications, the submissions--have ever been "high quality" according to the methodological criteria of their own time?  Is it that scholarship has entered a methodological decline? Or, rather, that scholarship has always been, in the main, fairly bad (cue Sturgeon's Law), but that the circumstances producing a particular form of badness have changed?