Not being more than glancingly familiar with Klemperer’s prewar recordings (and even that would give me only a fragmentary impression), I cannot say how his conducting may have changed between the 1920s and the 1960s. But, since his philosophy of music-making seems to have remained the same, I think it’s better than sheer speculation to propose that there does seem to have been a close aesthetic bond between what we would have seen and what we would have heard at the Kroll’s 1929 Holländer. If so, the production had its own integrity—its parts agreed with one another—and for that reason alone was much to be preferred to the intentionally de-constructed wreckage of our own increasingly normalized avant-garde. I’m also sure that had I been young in Berlin at that time, while I would not have wanted to forego the playing and singing of the ensembles under Bruno Walter at the Städtische Oper and under Erich Kleiber and Leo Blech at the Linden Oper, I would have been excited by the rebellious alternative being offered at the Kroll, and in the Kroll’s corner against the political attacks on it from the right. We are nonetheless left with these questions: Granted that this production and others at the Kroll had their intramural integrity, did this integrity extend to a bond with the work itself? Did its modernist aesthetic serve the creator’s vision? (Shouldn’t it? Why not?) Did modernism finally prove to be a productive step for opera to take? Was this “reform” truly a reform, not only in the sense that any new tax bill is “tax reform,” or in the always-useful one of purging decadence, but in the ultimate sense of opening a path to something more nourishing than what had gone before?

Had Klemperer’s recordings of Don Giovanni and Fliegende Hollander benefited from better casts—let’s say that of Antal Dorati’s recording for RCA Victor (Leonie Rysanek, George London, Giorgio Tozzi, Karl Liebl), or that of the 1950 Met revival under Fritz Reiner (Astrid Varnay, Hans Hotter, Sven Nilsson, Set Svanholm), then of course his structuralist reading would have been less determinative. (But what part, I wonder, did O.K. play in the selection of singers for these recordings? Did he want these singers, for their inability or reluctance to engage in “exaggeration or emotional indulgence?” Some of the complaints about the Kroll, surely not all politically motivated, were that the conducting exerted too rigid a grip, and that the staging, though fresh in terms of ideas about the characters, in practice reduced them to “cramped puppets.” We learn next to nothing from either Heyworth or Carnegy about the actual effect of the onstage performances.) So I pulled out those two recordings, and a number of others from the late 1920s into the early ’40s (samplings of key scenes, not complete hearings), in search of individual performances that might fulfill my own feel for the opera and its characters. I was particularly interested in singing that seemed to capture the tonal caste, the color and mood, of the Dutchman/Senta pairing, as against that of Daland, Erik, and the others—in short, of the two worlds I spoke of in the first part of this article.