[ Ref. Piotr Barcz in 25.10.20 MMDigest ]

Piotr Barcz raises good questions about faithfulness to the original
when recutting rolls. Those could be understood as issues of technical
quality (rough synchronization, missing or undersized punches), versus
taste-based revising of the arrangement (adding or subtracting runs
and counter-melodies, altering syncopation, deleting tremolos).

As we know, since the dawn of the player industry, even rolls attributed
to famous artists were routinely reworked by editors, not only for the
sake of compatibility with player design, but often dressing up the
arrangement with doubled melodies and various embellishments (not to
mention the habit of QRS to use up old labels regardless of the contents
of the roll). The industry was focused on economy, entertainment value,
and plausible appearance rather than the needs of scholars to trace
origins.

That was then, but this is now. I absolutely believe that any deliberate
change of content in a recutting project should be strictly identified
in the labeling.

If I want rolls to be updated to modern tastes, I should make that
choice with my eyes open.

As one who is active in producing sheet music for piano ensembles,
I am painfully aware that I cannot get into the mindset of one who
lived a hundred years ago, one who worked full-time preparing thousands
of songs. Listen to any typical Imperial roll from 1928. The instincts
that produced it have vanished from the face of the earth. Modern
arrangements of period songs have their place. But I can hear the
difference in the first ten seconds.

Since it is the best rolls that attract attention for recutting
(not the marimba waltzes and tear-jerkers), the stakes are high.
There are various approaches to recutting them. What we need now
is not blind fetishism, but transparency.

Jim Neher

[ Here's a good 1928 Imperial roll example:
[ https://www.youtube.com/watch?v=x4rOzucweK8 
[ Who Wouldn't Be Jealous of You? - Imperial Songrecord 07170
[ -- Robbie