The oldest extant English love song remains as faded ink on warped manuscript paper kept in the King’s College Archives.^[1] A 14th-century scribe jotted down all the lyrics but did not bother to complete the music, leaving an entire staff blank.^[2] He may have intended to compose this missing part of the music. Alternatively, he may have decided that this part did not need to be written, as it would be improvised; medieval musicians commonly improvised the accompaniment to notated monophonic lines.^[3] Either way, we readers of this work are left with the alluring task of composing this implied, unwritten part.

Varying interpretations of the manuscript

I enjoyed listening to two existing interpretations of this song on the Internet, while finding myself dissatisfied with the choices that informed their composition. The first is a series of three alternate compositions of the “missing part” by Ian Pittaway,^[5] who justified his compositional choices in a detailed blog.^[6] Pittaway’s foundational decision, for all his compositions, was to treat the melody written on the page as the cantus inferior, or accompaniment, to a missing cantus superior, or lead voice. He makes a compelling case for this choice. The resident of St. James’ Priory who wrote this down on the back of an old papal bull would most likely have followed ecclesiastical practices.^[7] In the church, the cantus inferior was written on the top line. Hence, the melody of “Bryd one brere,” written on the top line, should be understood as an accompaniment for a pre-existing melody that the scribe already knew by heart.

Pittaway further argues that the characteristics of the melody on the page are not what we would expect from a mode written according to the ecclesiastical practice of the time. It starts on a D and ends on a C, which is weird. The melody also lacks a stable “reciting note,” a lodestar that recurrently expresses the tonal center, in medieval (and especially ecclesiastical) music. Pittaway is not willing to accept that an ecclesiastical “lead melody” of the period could have this shifting, multimodal nature. Instead, he finds that it would make perfect sense as the accompaniment for a leading melody in G mixolydian.^[8]

The philosophical basis of Pittaway’s interpretations is Historically Informed Performance. Pittaway has consciously drawn, from medieval scholarship, a set of musical prescriptions that he applies to “Bryd one brere”:

This approach demonstrates an admirable quality of humility, in admitting that it is impossible to “reconstruct” the missing voice, but only to “construct” it. Pittaway, however, does not adequately explore the possibility that these “firm and established principles for medieval polyphony” are themselves a modern construction that are unable to account for the diversity of musical approaches in the Middle Ages.

What if the author of the “Bryd on brere” manuscript was intentionally composing a lead voice in a more flexible, multi-modal style? This would be completely possible, even for a monk. Maybe he was familiar with the principles of writing secular music and applied those to “Bryd one brere.” Pittaway does not consider that the very act of writing this secular love song on the back of an old papal bull was, if not sacrilegious, quirkily irreverent, suggesting that the writer may have been willing to break rules of both ecclesiastical and musical nature.

There is an even simpler possibility: that this writer copied down the pre-existing popular “secular” version of this song verbatim. After all, as Pittaway himself writes, “some secular tunes were clearly outside the modal system altogether.” The empty staff would have contained the accompaniment for this tune. And how do we account for the fact that the cantus superior staff was on the upper line? Well, maybe this (educated) ecclesiastical writer was simply aware of non-ecclesiastical conventions. Or maybe he followed his own convention; after all, this is the same delightfully flexible Middle Ages in which scribes changed language, seemingly on a whim, in the middle of a poem.

In my own version, I treat the given melody as the original melody of the song, in short, because it is a worthy possibility to try out.

In my own version, I treat the given melody as the original melody of the song, in short, because it is a worthy possibility to try out. There is precedent for this approach, as the other example from the Internet I listened to, arranged by singer Constance Fairfax and performed on the CD English Medieval Songs, treats the given melody as the lead.^[10] She does not, however, perform the song with any accompaniment. Citing Paul Hillier, she says that any accompaniment “should be done heterophonically,” an assumption that seems weak given the fact of the missing staff on the page.^[11] This staff heavily implies an intended contrapuntal or polyphonic texture, as heterophony would be, by contrast, improvised (heterophony concerns small ornamental deviations from one given melody).

I see Fairfax’s choice not to build an accompaniment as a missed opportunity. What is strong about her approach, though, is that she highlights and embraces the uniqueness of the tune, calling it “different from anything that was happening on the continent at the same time.”^[12] Rather than trying, like Pittaway, to mold the music to the prescriptive rules she believes all music of the period must follow, Fairfax allows the actual music on the page to inform her just as heavily as historical interpretation. For example, she wonders if the surprisingly large “jump of a sixth” in the melody “could be the result of clerical error,” as this was apparently not common in the melodic writing of the time, but she retains it, as she considers it “the most beautiful part of the song.”^[13] Thus Fairfax resists the rather modern impulse embedded in Historically Informed Practice to insist on standardizing everything.

My own approach combined Pittaway’s choice to compose a polyphonic arrangement from the monophonic information in the manuscript with Fairfax’s tolerance of the melody on the page as a viable lead melody. I wanted to bring out the bold melodic gestures—especially those large intervals Fairfax loves—in the melody, while also applying the principles of medieval polyphony, like Pittaway, to imagine what could have gone into the missing staff.

Notating my own interpretation of the manuscript

So, how to even begin to accurately interpret the staff that is filled in? I first revisited the Thomas Kelly reading for a needed reminder of the primacy of text in medieval notation. He writes that “at the foundation of our Western musical system….” is that “music is meant to go with words” and that “the basic unit of music-writing is not the note but the syllable.”^[14]

In Kelly’s excerpt above from Handel, it could not be more obvious to readers of this notation which syllable goes with which notes. The principle is that a syllable falls visually beneath its corresponding notes and is held until a new syllable. The principle is exactly the same in “Bryd one brere,” despite the added challenges of the distance between staves, lack of slurs, and poor visibility. My transcription of this excerpt should replicate the visual alignment of noteheads with syllables:

I went on to consult a few resources on deciphering neumes: a “Notation Manual” from by scholar Ross W. Duffin^[16] and “Medieval Music Introduction: Clefs and Staves,” by Nigel Home, an amateur trying to teach himself the basics.^[17] I carefully read these and cleared up a lot of things. Why do some staves have four lines? To save ink and space. Those strange C shapes at the start of each staff? Those are movable C clefs. How do we figure out the meter? We determine the rhythmic mode, which can be induced from the stress and length patterns of the music, which does not change. What could all the different square, diamond, and parallelogram shapes mean? They express rhythms and gestures relative to the presiding meter. According to Home, rhythmic mode is established by context,^[18] and in “Bryd one brere,” there is a clear succession of long-short (half-quarter) established:

But I found myself still confused about rhythm. Duffin and Home offered different interpretations of certain rhythmical figures; see for example their translations of a parallelogram-shaped ligature followed by a small square note-head.

Duffin: eighth, eighth, quarter:

Home: quarter, quarter, quarter:

It dawned on me that these differing interpretations of notation emblematized, yet again, the lack of standardization—and the exciting idiosyncrasy—in medieval writing of all kinds. I allowed my confusion to morph into tolerance of ambiguity. I decided that when it came to questions of rhythm, I would allow myself to treat ligatures with a particular amount of freeness, choosing to notate them as quarters, eighths, or even sixteenths based on what would make the most sense in the 2:1 metrical context.

What I would not do, though, is disrupt this metrical pattern. I recognize that this impulse partially came from my modern expectation of a consistent time signature. When I listened to the ensemble GALDRA’s performance of “Bryd one brere,” I found that its slowness and its use of duple bars that suspended the triple time evaporated the feeling of metrical pulse.^[21] The repetitive first words of the lyric demand, to my ears, a rhythmic pulse as strong as any nursery rhyme: “Bryd one brere brid brid one brere” (1).

I also feel supported by the way the manuscript’s layout compresses long clusters of notes so tightly that the eye scans them quickly. There were many instances where I simply could not imagine having a multi-measure-long melisma on a single word; there was no good textual reason to and no visual evidence for this. Later, in the revision stage, I did need, out of sheer pragmatism and kindness, to insert a few breath marks to allow the singer to breathe!

In translating rhythms, I also paid careful attention to the inherent scansion of the words themselves. The first “brere” is a great example, as the natural stress of “BRERe” (just as in modern “BRIar”) strengthens the interpretation of the corresponding notes above it as long-short or stressed-weak. In fact, the second occurrence of “brere” would have presented a trap if I had not been sensitive to speech rhythms and the established triple meter. Here, it looks like the there are two longs on the second line of the staff, directly above “brerre”:

But on closer inspection, that second note is actually a stemless breve that directly abuts a large vertical line that crosses the entire staff: in other words, a bar line?! I could see several possible reasons for such a marking. Maybe it meant to breathe before such a large leap of a minor sixth into the next note. Or maybe it reinforced the formal structure of the text: that after two repetitions of an apostrophe to this bird on a briar, we will now be making a new statement. The textual equivalent would be inserting a line break as follows:

On another note: what about the unusual minor sixth leap that occurs mid-melody? After having descended stepwise from D to C to B on “me thou,” we suddenly leap up to a G on the “rew” in “rewe.” This is the most dramatic moment in the melody and supports the visceral feeling of a word like “rewe.” Until this point, we have no hint of drama (besides maybe the dramatic word “crave”). In this scene, “kynd,” meaning man or humankind, gazes peacefully upon a “blithful” bird on a briar; the melody supports this reading of the mood with its lilting, soothing repetitions of “bryd one brere” (1-2).

The surprising shift at “rewe” in the music prepares the listener for a more, well, rueful interpretation as we go on. The text certainly becomes more rueful, as the speaker implores his beloved woman/bird to pity him before she might prepare his grave. The rhyme of “crave” (2) with “grave” (4) prompts us to try equating the speaker’s craving with his deathly downfall. So if the bird/woman does not heed the man’s craving for her attention, he will perish.

However, in the music, these words are painted slightly differently: “crave” on the note F with the expected long-short rhythm, but “grave” on C with a short-short. This actually makes a lot of sense. “Crave” and “grave” do not have to be equated, because it is not the man’s craving itself that inherently dooms him, but instead the woman/bird’s refusal that would. This is a good example of the music teaching us something about the text, rather than, as we more typically assume, a text elucidating its musical setting.

After the first “verse” of text (as lineated in Butterfield’s transcription), the visual correspondence between syllables and noteheads ceases, as the scribe continued chugging away at the text in a big block at the bottom of the manuscript. It feels safe to assume that this is because the melody repeats for the next two verses, and the scribe felt no need to be redundant. Indeed, this is also how Pittaway and Fairfax’s transcriptions treat the manuscript, and I agree with them.

The only challenges in setting the next two verses to this melody were in accommodating the different syllable patterns in certain lines. In particular, the refrain of “loveli trewe” warranted the composition of a new little phrase. Although I could fit the first instance of this refrain in line 7 into the same notes used in line 3 in the previous verse, I had more trouble with line 10. If I were to follow the same melody that I used for lines 2 or 6, I would only have two notes left to use for “loveli trewe.” Besides, I wanted to take the liberty of emphasizing this second interjection of this phrase, and felt that it could use some separation from the linear flow of the surrounding words. So here is my solution:

I know that I could be far more literal with my treatment of the verses. Fairfax’s transcription places all three verses, in the style of a hymn arrangement, below the single given melody, with minimal intervention. I intentionally avoided doing this, to try to evoke the improvisatory nature that I suspect this melody was sung with, and also to clearly paint the emotions in the text. I truly got to contend with such considerations when inventing the “unwritten accompaniment.”

Composing the “unwritten accompaniment”

After interpreting what was notated on the page, I set out to construct my own version of the accompanying voice that was not notated. I found myself very attracted to polyphonic partwriting principles of the period, as Pittaway described them, since they are very freeing. I, like many composers who pursue an undergraduate music education, am well-versed in Fux’s famous 1725 treatise on species counterpoint: Gradus ad Parnassum. Although I won’t deny that Fux’s principles lend themselves to very beautiful music, I have always found them arbitrary and aesthetically limiting. Pittaway’s summary of fourteenth-century accompaniment practice allows for several fun things Fux forbids: long strings of parallel thirds and sixths, fairly free uses of dissonant seconds and sevenths in passing, and resolutions on perfect fourths. I also observed, in Pittaway’s performance of “Bryd,” rather bizarre “crossings” between the two voices, where the higher voice becomes the lower one and vice versa.^[22] In my composition, I immediately take advantage of all these possibilities:

Composing using these principles produces a much more raucous, resonant, and unpredictable texture than the counterpoint that was largely standardized after Fux, which I would associate with a more refined, gentle, and sophisticated aesthetic. As I composed, I had to consciously force myself to adhere to the former aesthetic. For example, I often wanted to pretend the scribe had written B-flats where there were B-naturals, so I could avoid the yucky tritone between B-natural and F-natural that kept occurring in my writing between the two parts. But there was no getting around the B-natural!

There is a widespread myth that the tritone was considered “the devil’s interval” in medieval music and banned by the Catholic Church.^[23] This myth originated with a line in none other than Fux’s Gradus ad Parnassum:

Mi contra fa est diabolus in musica

Mi against fa is the Devil in music^[24]

Fux meant this as a practical word of caution, as tritones are hard to sing properly in a group. It was only later Romantic composers who interpreted this phrase literally. We have no evidence the Church ever banned tritones. This myth of the “devil’s interval” reveals much more about the narratives that Romantic and post-Romantic thought have projected onto the Middle Ages.^[25] In my own fear of tritones, I reflected a modern assumption about what music can be and had to purposefully let go of it to entertain the possibility of a tonally multilayered medieval manuscript.

This is a good moment to revisit Pittaway’s contention that the melody in the manuscript was unlikely to be a lead voice, as its construction does not have a consistent reciting note or a tonally stable construction. It starts on a D and ends on a C, and, I will add, lands on A and F at the end of some phrases in between.

But who is to say that we cannot imagine this piece as bitonal or even polytonal? We simply do not know how it was performed, and if we do treat the given melody as the lead, a level of polytonality is completely unavoidable, since it starts on D and resolves/ends on C. Perhaps this is why Paul Hillier, whom Fairfax cited, said that the only solution for accompaniment was heterophony—he may have dismissed any polyphonic option because it would necessitate polytonality.

If we embrace a feeling of polytonality, we can choose how to harmonically color the three moments where each verse ends on the note C:

1. mm. 15-17: The C of “grave” is harmonized with an A, creating a darker minor third between the voices that supports the unexpected darkness of the “grave” creeping into this love song (4):

2. mm. 33-4: The C of “alle,” by contrast, is harmonized with an E, as this major third feels bright, as the beloved is described as “fair and flower of all” (8):

3. mm. 51-2: Lastly, I choose to have a unison on C for the final word, “newe,” as this unison is one of the consonances that can end a song in the period, and sounds incongruous with the motion and harmony preceding it. I love this lack of resolution because we do not know whether the beloved will “save” the speaker of the lyric from his “sorrow” and make him “new” (12). I also find that this unison ends the piece on a somber note, balancing out the speaker’s excitement with the bittersweet longing that is so palpable in the lyrics.

We may never be certain if this approach reflects the practices of the era; regardless, in my opinion, it heightens the emotions in the text. This sentiment is also mirrored in a quotation Fairfax offers from Duffin:

Singers must “Develop one’s own version of medieval English…Only when there is a natural correspondence between the emotional tone of the text and its physical representation in sound…can the language actually be said to be authentic in any way.”^[13]

In other words, wherever true historical authenticity (whatever that means) is unattainable, authenticity to the “emotional tone of the text” should be the priority.

On a final note, I wanted my accompaniment to honor and enhance the gestural freedom that the melody takes in its multiple leaps of a sixth. After the lead dramatically leaps one last time—into the word “saven,” meaning “save”—I echo this gesture with a slightly less dramatic stepwise ascent of a sixth in the accompaniment (mm. 46-49):

This rising gesture to “blisse” highlights “blisse,” happiness, as the key thing the speaker is reaching for. By contrast, the given melody in the lead voice falls from a G on “saven” to a C and a B on “blisse” in my transcription. I like how this multilayered word painting allows multiple interpretative possibilities. Was bliss attained? Was it fleeting? Or does this question hardly matter, as the lyric is evoking the bittersweet taste of desire? After this moment, I have the singers linger melismatically on the pun “were were me newe,” which I interpret as a subjunctive or a counterfactual: “would clothe me anew.” Again, in my view, the lyric is all about wishing (mm. 50-52):

Comparing medieval and modern music notation

My foray into the manuscript of “Bryd one brere” showed, if anything, that it is very possible for someone nowadays to decipher medieval music notation (given some time, energy, and passion!). The principles of ledger lines, noteheads, horizontal time, vertical pitch, and the primacy of the syllable all exist in this manuscript from the 1300s, and these priorities would remain strong in its descendant, modern notation.

Rhythm was not the priority, but rather the kinetic gestures evoked by their various shapes.

What differentiates modern notation from its ancestor is its standardization and specialization of very specific visual choices for representing sound. As Kelly explains, medieval thought did not take it for granted that visual media could even represent music at all.^[26] Most scribes thought it sufficient that notation should show not “what the song is” but “how to sing the song.”^[27]

In the case of “Bryd one brere,” this means that ligatures lack consistent rhythmic values because rhythm was not the priority, but rather the kinetic gestures evoked by their various shapes. The author felt no need to set every word of the text to a melody because he knew he wrote enough down to jog his memory, or his readers’ memory, in the future. This use of notation as a highly gestural mnemonic is why the author of “Bryd one brere” would probably find it confusing if I told him I “transcribed” or “translated” his scribbles.

“You don’t knoweth the tune,” he would say, “and argh! That was just a doodle. Getteth out of my private stuffs!”

If I explained I did not know the song, so I was “improvising” or “composing” from what he provided, though, he would be less confused. The evolution of sheet music from performance aid to a metonym for a musical work itself has severely limited what music can be to what can be put on the page, at least until the 20th century. (Jazz began to remind the Western musical establishment that it’s possible to jam together in an intellectual way, but that’s a whole ‘nother paper.) This is why I confined the harmony, counterpoint, and meter of “Bryd one brere” to the largely standardized idioms of 21st-century classical notation.

I am very pleased, though, with how I challenged myself to allow the words and neumes in the manuscript to dictate my choices above all. I offer my own score and lyrics below; the more interpretations are made, the more we’ll make sense of the spirit of the original tune.

Score and Performance

Please contact me if you are interested in viewing the score!

Bibliography

Butterfield, Ardis, ed. “Bryd One Brere Brid Brid One Brer,” No. 9. In Medieval English Lyrics, 7–8. Norton Critical Editions (Norton: forthcoming), n.d.

Duffin, Ross. “Notation Manual.” Case Western Reserve University, 2000. https://artscimedia.case.edu/wp-content/uploads/sites/135/2020/04/20164113/NotationManual.pdf.

Fairfax, Constance. “Bryd One Brere: The First English Love Song,” April 14, 2010. https://web.archive.org/web/20100414072105/http://home.uchicago.edu/~atterlep/Music/Songs/brydonebrere.htm.

“Fragment, Words before 1250, Setting Possibly 14th Century.” GB-Ckc KCAR/6/2/137/01/1 SJP/50. King’s College Archives. https://www.diamm.ac.uk/sources/988/#/.

Horne, Nigel. “The Written Notation of Medieval Music,” n.d. https://www.dolmetsch.com/medieval.pdf.

Kelly, Thomas Forrest. Capturing Music: The Story of Notation. First edition. New York: W. W. Norton Company, 2015.

Jokinen, Anniina. “Middle English Lyrics: Bird on a Briar.” Luminarium: Anthology of English Literature. https://www.luminarium.org/medlit/medlyric/brere.php.

Pittaway, Ian. “Bird on a Briar (Bryd One Brere): Constructing the Missing Voice.” Early Music Muse (blog), March 12, 2019. https://earlymusicmuse.com/birdonabriar-constructingthemissingvoice/.

Pittaway, Ian. Bryd One Brere (Bird on a Briar) : Medieval Polyphonic Song – Constructing the Missing Voice, 2019. https://www.youtube.com/watch?v=9KwItjaathE.

Stephanie Zimmerman, Mara Yaffee, and Namgon Lee. Galdra Performs “Bryd One Brere,” 2019. https://www.youtube.com/watch?v=1z9csK5X0II.

Sword, JD. “Diabolus in Musica: Did the Catholic Church Ban the Tritone for Being ‘the Devil in Music’?” Skeptical Inquirer (blog), March 31, 2022. https://skepticalinquirer.org/exclusive/diabolus-in-musica-did-the-catholic-church-ban-the-tritone-for-being-the-devil-in-music/.

Lyrics

Ardis Butterfield’s transcription:

 My translation:

1. 
   

   Constance Fairfax, “Bryd One Brere: The First English Love Song,” April 14, 2010, https://web.archive.org/web/20100414072105/http://home.uchicago.edu/~atterlep/Music/Songs/brydonebrere.htm. ↑

   
   
2. 
   

   “Fragment, Words before 1250, Setting Possibly 14th Century,” GB-Ckc KCAR/6/2/137/01/1 SJP/50, King’s College Archives, accessed December 11, 2023, https://www.diamm.ac.uk/sources/988/#/. ↑

   
   
3. 
   

   Pittaway, Ian, “Bird on a Briar (Bryd One Brere): Constructing the Missing Voice,” Early Music Muse (blog), March 12, 2019, https://earlymusicmuse.com/birdonabriar-constructingthemissingvoice/. ↑

   
   
4. 
   

   “Fragment, Words before 1250, Setting Possibly 14th Century.” ↑

   
   
5. 
   

   Bryd One Brere (Bird on a Briar) : Medieval Polyphonic Song – Constructing the Missing Voice, 2019, https://www.youtube.com/watch?v=9KwItjaathE. ↑

   
   
6. 
   

   Pittaway, “Bird on a Briar (Bryd One Brere): Constructing the Missing Voice.” ↑

   
   
7. 
   

   Pittaway. ↑

   
   
8. 
   

   Pittaway. ↑

   
   
9. 
   

   Pittaway. ↑

   
   
10. 
    

    Luminarium: Anthology of English Literature, accessed December 11, 2023, https://www.luminarium.org/medlit/medlyric/brere.php. ↑

    
    
11. 
    

    Paul Hillier, quoted in Duffin, Ross W. A Performer’s Guide to Medieval Music. Indiana University Press: Bloomington, IL. 2000, quoted in Constance Fairfax, “Bryd One Brere: The First English Love Song,” April 14, 2010. ↑

    
    
12. 
    

    Fairfax, “Bryd One Brere: The First English Love Song.” ↑

    
    
13. 
    

    Fairfax. ↑

    
    
14. 
    

    Thomas Forrest Kelly, Capturing Music: The Story of Notation, First edition (New York: W. W. Norton Company, 2015), p. 12. ↑

    
    
15. 
    

    Image from Kelly, Capturing Music: The Story of Notation, p. 16. ↑

    
    
16. 
    

    Ross Duffin, “Notation Manual,” Case Western Reserve University, n.d., https://artscimedia.case.edu/wp-content/uploads/sites/135/2020/04/20164113/NotationManual.pdf. ↑

    
    
17. 
    

    Nigel Horne, “The Written Notation of Medieval Music,” n.d., https://www.dolmetsch.com/medieval.pdf. ↑

    
    
18. 
    

    Horne, “The Written Notation of Medieval Music,” p. 1. ↑

    
    
19. 
    

    See Duffin, “Notation Manual,” p. 5. ↑

    
    
20. 
    

    See Home, “The Written Notation of Medieval Music,” p. 5. ↑

    
    
21. 
    

    Stephanie Zimmerman, Mara Yaffee, and Namgon Lee, Galdra Performs “Bryd One Brere,” 2019, https://www.youtube.com/watch?v=1z9csK5X0II. ↑

    
    
22. 
    

    Ian Pittaway, Bryd One Brere (Bird on a Briar). ↑

    
    
23. 
    

    JD Sword, “Diabolus in Musica: Did the Catholic Church Ban the Tritone for Being ‘the Devil in Music’?,” Skeptical Inquirer (blog), March 31, 2022, https://skepticalinquirer.org/exclusive/diabolus-in-musica-did-the-catholic-church-ban-the-tritone-for-being-the-devil-in-music/. ↑

    
    
24. 
    

    Why “mi against fa,” if mi and fa form a second, not a tritone? What Fux meant, in context, was that the mi of the dominant scale conflicts with the fa of the tonic scale. For example, in C major, this would be the notes B (mi of G) and F (fa of C)—forming a tritone. ↑

    
    
25. 
    

    Sword, “Diabolus in Musica.” ↑

    
    
26. 
    

    Kelly, Thomas. Capturing Music: The Story of Notation, p. 16. ↑

    
    
27. 
    

    Kelly, p. 12. ↑