I think the taxonomy of transcriptions of great works is very clear when we realize that some transcriptions are available that are not by great/famous musicians, ... but then there's the Brahms LH version of the Chaconne and some of the Liszt transcriptions that have greater life (and merit?) than the orginal and we're back to square one . I think that from the issue of dependence (the original has life of it's own, the transcription is like a virus borrowing the DNA of the host's cells in order to survive) it should first and always go under the original composer, despite the stature of the transcriber. Sure am glad I don't have to make these decisions .

Spoken like a true MD



Good point, also the Rachmaninoff transcriptions, for instance, of the Kreisler Liebesfreud/Liebesleid, which IMO are metamorphoses that are greater than the original. But I was speaking merely of the convention (what usually seems to be done, whether I agree with it or not). There's also the case for this site of transcriptions of non-keyboard works by composers who didn't write anything for the keyboard (for instance, the Vivaldi-Bach keyboard concertos) of which I play one on the site), and the Wagner-Liszt transcriptions, the former having written very few piano pieces and none of them great. That throws a bit of a wrench into the idea of linking to both composers. I think it makes sense to go with transcriber just for convenience, since it seems to cover all of these possible cases. The transcriber is, after all, the one who put the actual ink to paper and made the decision to imitate the work at all, regardless of how good that imitation is and how great the transcriber is.

Great

Hehe yes, I meant the second, obviously. In fact I meant 'dispute' rather than 'argue'.
Language is not my strong point

Yes here we have it again. It would be weird to have Kreisler or Vivaldi listed as composers on the site, because they wrote nothing for the piano. OTOH it is even weirder to have Hess, Kempff, and Siloti, who did not compose, listed as composers, just because they happened to transcribe the odd Bach piece.
It is a conundrum we'll never get out of. I think we should simply forbid transcriptions. But some here might not agree

Namely Andrew

I do think you're jesting, but just in case, we do need to have transcriptions here.

Yes, he's kidding. But I sort of wish it were true. Except let's wait until I'm done working on my Rachmaninov-transcribed piece, then the rule can take effect.

This discussion is deja vu, as we've examined it before. Anyway, all of the many examples I've ever encountered have always appeared as composer followed by transcriber even when the transcriber is also famous. But there are exceptions to every rule, of course.

David

Just to join in the fun. If one can only file a piece one way, it seems to me that by composer is best. Particularly in instances such as Bach, and I wouldn't be surprised if it occured some with Buxtehude, some pieces have been transcribed by more than one composer/arranger. Filing them under the original composer allows one to see that there are different versions to compare.

Scott

True, but I don't think the conversation was about that. It was about whether transcriptions should be categorized by transcriber or by original composer (e.g., on the site).