Why the “Josquin Research Project”?
Formed by Jesse Rodin in 2010 in collaboration with Craig Sapp
at CCARH, the JRP was provided
funding to confront a well-known scholarly problem: of roughly
340 works attributed to Josquin des Prez, only a fraction are
securely attributed. As an initial goal, we sought to digitize
all these works to enable comparative analysis. Drawing on a foundational essay by
Joshua Rifkin, we chose the name "Josquin Research Project" in emulation
of the Rembrandt Research
Project, which is similarly interested in questions of attribution.
Shortly after beginning our work, we came to see how much more
this project could accomplish. While we continue to be interested in
attributive research, we now pose a range of questions aimed at discerning
relationships among pieces, composers, genres, even whole periods. By
way of example: How does style change over time? What separates a song
from a mass? What really makes one composer different from another?
What is your process for putting music online?
- Starting with a reference score (usually a published edition),
we enter music using the notation software Finale, then tag items such
as mensuration signs and section headings.
- Rodin reviews each file by listening to it twice: once while
viewing it in Finale, another time while reading along with the reference
score. (During this process he adjusts editorial accidentals.) He then
exports the file into Music XML.
- Sapp translates the MusicXML data into
Humdrum syntax, which is
used to create search indexes, analyze the music, and generate graphical
scores for display on the JRP website.
- Rodin checks the online version for errors.
If you find a mistake, please visit the relevant Work page and submit an
How good is your data?
We pride ourselves on presenting extremely high-quality data. Not only
do we choose our reference scores with care, but we also go out of our
way to remove what we consider to be unhelpful features of some modern
editions (e.g., quixotic barring). When we transcribe, we enter the
music voice-by-voice in real time; listening to how each newly added
voice fits with the others helps prevent errors. Equally important is the
threefold review process undertaken by the project director (see above).
Why do JRP editions use original note values and mensuration signs?
We believe modern editions should facilitate comparison with original
sources. We also want users to be able to compare works across composers,
genres, and time periods without having to adjust for different editorial
Why doesn’t the JRP encode variant readings or text underlay?
In an ideal world we would. Unfortunately doing so would take so much time
that new works would be added at a snail’s pace. We therefore use modern
editions as reference points for our transcriptions and, except in rare
cases, encode only textual incipits. This points up an important aspect
of the project: the JRP is above all a powerful finding aid and analytical
tool. It is not intended to replace published critical editions.
Why don’t all JRP editions include editorial accidentals?
Rifkin has argued that editorial accidentals, while implicit in the
sources that have come down to us, are not strictly speaking part of
the musical texts. They belong instead to the realm of realization:
it is up to the singer to decide between what we today would call f and
f#. For all of this, we recognize that many people find editorial
accidentals helpful. The JRP therefore offers two versions of each score:
one free of editorial accidentals, another that displays accidentals
that have been added by the project director. (The latter underpins our
MIDI files; for practical reasons, decisions about which accidentals to
include cannot be weighed with great care.) All JRP analyses depend on
the version of the score without editorial accidentals.
Why does the JRP use unconventional Latin spellings (e.g., "celi")?
For the sake of transparency, we generally adopt the spellings used in the
surviving musical sources.
How do I know what edition the JRP has used as a reference for a given work?
We plan to implement a display of this information soon on each Work
page. While we generally rely on Complete Works editions and other
scholarly editions, we routinely make changes in order to remove barriers
to a transparent representation of the original notation. For the same
reason, we sometimes also consult original manuscript and print sources.
I’m working on repertoire that I want to see on the JRP site. How can I help make this happen?
Volunteer to transcribe for us! We’ll need you to "tag" your
Finale files for use by the JRP; to that end we can make available a
series of straightforward instructional videos. (By way of example,
the first section of a Kyrie must be labeled: "Section: Kyrie I.")
What if this repertoire falls outside the JRP’s chronological range (ca. 1420–1520)?
We would still be happy to consider it. (We already host some
mid-sixteenth-century music in the form of motets posthumously attributed
What are the project’s funding sources?
The JRP has received funding from the American Council of Learned
Societies and from Stanford University, with in-kind support from the
Center for Computer Assisted Research in the Humanities
How can I contribute?
The JRP is constantly developing new tools and expanding our
database of scores. To make a tax-free donation through Stanford
University, enter "Josquin Research Project" in the "Special
instructions/Other designation" field here.
We also welcome donations of data and time. Occasionally we are able
to hire paid employees.
Something isn’t working/could work better. What can I do?
Let us know. We are committed to addressing problems and answering questions.