Having worked our way through the Pauline Alpert saga in the last few
days, here's a little something more to think about.
Denis Hall has been investigating the coding techniques of Duo-Art
rolls, from the very earliest which were simply added on to existing
Metro-Art rolls, through various intermediate means of going from the
recording console to the finished roll, to the final method which
involved reducing information to terraced dynamic levels and then re-
applying details from the original recording. This was briefly
described in notes from me and Dan Wilson a few weeks ago.
Looking at the various ages of English-made rolls, it is clear that a
lot of recoding took place. In particular, it seems as if there was a
major recoding exercise when rolls were chosen for the World's Music
Series. In other words, rolls recorded in the 1914-1925 period were
revised for reissue in late 1926.
One of the examples which has come to light is a very substantial
rework indeed, where the decision about which notes are theme and which
are accompaniment has been changed. The coding levels have then been
substantially reworked to reflect this change.
Denis feels that this is the work of Reginald Reynolds -- who coded the
original UK rolls anyway. He was known to revise coding sometimes
after issue, in any case, and certainly revised American rolls before
their UK issue. For the World's Music Series (WMS), the major recoding
appears to be on poorer rolls. Perhaps Percy Scholes (the editor of
the WMS rolls and presumably the one who chose the titles) selected
pieces on title only, and Reynolds had to resurrect the worse
performances in light of later and better recordings.
This is an interesting area when we are considering authenticity of
performance. For a start, is this just a UK-Aeolian quirk? Is there
any evidence of American rolls being reworked in this way? (This is
something Wayne Stahnke can probably tell us from his experience with
scanning them into the computer.) This evidence adds to our analysis
of the attitudes Aeolian had regarding 'authenticity' versus a
marketable commercial product. What is obvious is that recoding years
after the event can hardly have been in reference to the original
performance!
Julian Dyer
|