Abstract
It seems natural to us to search for the authentic form of Gregorian chant, as it is thought to have become corrupted over the centuries. But can restoring the melodies of this corpus be reduced to finding its original form free of all error? Indeed, what can we know about melodies that have been written only after several centuries of practice? How can we detect the original version when we know that there was no "composition" in the modern sense? A set of traditional vocal practices constitutes the cradle of Gregorian chant, and a return to these vocal practices may well be the key to rediscovering this chant in all its original purity and strength.