Mssr. Griffin is correct - there is no authoritative version of Allegri's Miserere. The piece, from its creation, was always being ornamented and re-ornamented as it was sung -- and sometimes even rewritten -- to make it fresh and compelling, since it was performed every year during Holy Week for several centuries. (This was part of the magic and allure of the Sistine Chapel Choir.) For slightly more background, here.

The version we typically hear (with the high C) is a 19th-century attempt to notate definitively the way the piece was sung (which doesn't exactly make sense, since the piece was supposed to vary each year, thus making a definitive version impossible). It is ironic that in this "definitive" version, the high C results from a mis-transposition of the last few measures of the soli part. It was part of neither the original composition nor its first two centuries of use.

There have been many attempts in the last few decades to try to unearth Allegri's original score and to resuscitate more historically accurate ornamental techniques. Some time ago musicologist Hugh Keyte prepared an edition in which the music for each choral section was drawn from a manuscript from a different time period, by which he could show the variation of ornamentation techniques the Sistine Chapel Choir employed in this piece. A performance of that (quite slow, but quite beautiful) can be found here, and comparing it with the score, you find
- 0:00 Allegri's original polyphony, c.1668 (interspersed with recto tono chant)
- 2:00 Sistine Chapel anon. manuscript, c.1791.
- 3:17 Hugh Keyte's projected ornamentation, c.1987, based on Allegri (earlier) and Alfieri (to follow) - we get a high B-flat here, but we don't change keys!
- 4:25 Charles Burney's edition, 1771.
- 5:41 Rev. Pietro Alfieri's edition, 1840 - return of the high B-flat!
- 7:00 Wroclaw manuscript, early 19th-cent., pretty dramatically ornamented, and yet again the high B-flat.
- 9:31 Keyte, this time based on Burney et al.
- 10:51 Paris Conservatory manuscript (date not indicated).
- 12:33 Sistine Chapel anon. manuscript, c.1791 (again), concluding nine-voice tutti.

The research continues. In 2014, a fellow published a book solely about the Miserere and its historical development. The book includes two different realizations of the score, and (of course) the author will happily sell you performing copies of either (or both) editions. You can hear the ensemble he conducts singing an excerpt from the Miserere here (scroll down a bit). It's pitched quite high, and some of the ornaments are wild.

Finally, if you've made it this far and you're still interested, you can see me conducting the Keyte score in a 2010 concert here (except that instead of recto tono chant, our editor employed the Tonus Peregrinus, to lovely effect).

Interestingly, the Tallis scholars' CD of the piece includes two versions, presumably for this reason: one in the standard arrangement, and one with vocal improvisation added by a soprano. It was quite interesting to listen to.

I've heard many different versions of the Allegri recently and have been awed by the variations in the soprano line. We're attempting to sing the Miserere this year and my soprano has a wonderful high C. When she's not a church musician, she sings R&B so her improvisation skills are very good (though they may have to be tweaked for style!). I plan on giving her artistic license.... I'm looking forward to some fun.