PolyphonyDatabase.com is a detailed catalogue of early music sources designed to help musicians perform, academics study, and enthusiasts explore a vast and glorious repertoire quickly and easily. It aims to combine the practicality of CPDL with the academic rigour and ambition of the RISM census, to make use of similar projects where possible, and to directly combat the frustrations performing musicians have with all existing resources.
Its three main goals are:
The database was founded by Francis Bevan in 2014 as an outlet for his editing hobby and is regularly updated by him and a small team of enthusiasts. If you would like to contribute some cataloguing time, recordings or editions, submit corrections, commission an edition or just donate some cash, please get in touch with Francis via email: polyphonydatabase@gmail.com.
The best way to help fund the project is to commission performing editions. New editions can be made quickly for as little as £10.
2025 update: Francis has rewritten everything in Node.js with new advanced search capabilities and improved cataloguing tools. Please get in touch if you spot any bugs!
The text search will match case insensitive exact strings in titles only. You can use the percentage sign % as a wildcard to match any number of characters. e.g. "missa % marc" will return "Missa Papae Marcelli".
i/j and u/v are also treated as interchangeable so results for "Jesu" will include titles containing 'Iesu' and 'jesv' etc.
Titles are being matched to liturgical feasts and seasons semi-automatically: each incipit is compared against the complete propers of the Mass and Office (using the corpus of the excellent Divinum Officium project, 1960 rubrics), and every proposed match is reviewed by a human before it appears here. This is very much a work in progress.
A function listed against a piece means its text genuinely appears somewhere in that day's propers — but many texts (especially psalm verses) serve on several days, and historical or local uses (Sarum, pre-Tridentine, votive) go beyond the 1960 books, so a piece's use is not necessarily limited to the days listed, nor is the list of pieces for any feast yet complete.
A red clef means the voice is missing from this source. Where there are no concordances to fill in the gaps, this means the piece will require reconstruction.
A green clef means the voice is incomplete in this source, e.g. a fragment of a larger work.
A yellow clef denotes a canonic voice — one derived from another part following a canon's rule — whether or not it is written out in the source. Where the voice isn't notated, the clef shown may be known from the canon's rule or our best guess. The piece is still complete and performable. Use the "Has canonic voice" filter to find all pieces involving canon.
A faded clef means much of the piece can be performed without this voice. For example, an extra 6th voice in the Agnus Dei of an otherwise 5vv Mass setting, or a short gimel in a big votive antiphon. Filtering by number of voices will exclude these clefs, while filtering by a specific voice combination will include them.
A blue clef denotes a voice that uses more than one clef in this source - this is particularly prevalent in earlier printed sources. We've estimated the larger clef on the left to be the most used to give an idea of voice distribution at a glance.
x clefs are used as placeholders for voices we know are necessary, but haven't yet worked out which, normally because we don't have access to a full facsimile of the source.
| Title | Function(s) | Composer | # | Sources |
|---|
