The medieval harp (3/3): performance practice

Psalter, England, c. 1225 (The Morgan Library and Museum, New York, MS G.25, folio 3v).

This is the third of three articles about the medieval harp. Having outlined harp history from the earliest evidence in Egypt to the end of the medieval period in the first article, and used medieval art and written witnesses to illustrate harp symbolism in the second, this final piece lays out the evidence for questions of harp performance.  

The basis of this article is a description by the author Thomas of the playing of a harper-hero named Horn, written c. 1170, combined with other sources to built up a picture of medieval harp practice. This includes: harp tuning as a performance; the training of musicians; the various ways in which medieval harps were tuned and the musical reasons for these tunings; harp repertoire; preludes and postludes; and medieval methods of polyphonic accompaniment.

Each of these three articles begins with a performance on medieval harp of a different French estampie from c. 1300, arranged to the historically attested performance principles set out in this article. This article begins with La quinte estampie RealThe fifth Royal estampie.   

Click the picture to play the video, which opens in a new window.
La quinte Estampie Real from Manuscrit du Roi, a manuscript of troubadour and trouvère songs
written c. 1250, with instrumental pieces such as this estampie added c. 1300.
The fifth Royal Estampie is arranged and played on medieval harp by Ian Pittaway
using medieval musical principles described in the article below.

Though there are many medieval images of small groups of musicians playing together, most written accounts indicate a solo performance, either a single instrumentalist playing a melody, or singing, or singing to their own instrumental accompaniment. The central question of this article is: how specifically did a solo medieval harper play?

A remarkably instructive passage is found in the Anglo-Norman Roman de Horn by Thomas, c. 1170, in which the hero of the story, Horn, sings and plays harp. First, the account of Horn’s performance.

“Then he took his harp to tune it. God! Whoever saw how well he handled it, touching the strings and making them vibrate, sometimes causing them to sing, sometimes harmonies, he would have been reminded of the heavenly harmony. This man, of all that are there, causes the most wonder. When he has played his notes he makes the harp go up so that the strings give out completely different notes. All those present marvel that he could play thus. And when he had done all this he begins to play the aforesaid lai of Baltof, in a loud and clear voice, just as the Bretons are versed in such performances. Afterwards he makes the strings of the instrument play exactly the same melody as he had just sung. He performed the whole lai, for he wished to omit nothing.”

As outlined in the second article on harp symbolism, there are clear elements in this account of a literary formula, which includes the hero-harper tuning to the needs of the piece, checking tuning with a prelude; the main performance; and the wonderment of the listening audience, so caught up in the music that they lose their senses. My working assumption is that, the hyperbole of the audience’s response aside, the literary formula is based on real musical practice. Considered this way, the details in the account of Horn’s playing tell us a great deal. What follows is that account, taken point by point, cross-referenced with other medieval witnesses to build up a picture of medieval harp practice.

Harp tuning as preliminary performance

“Then he took his harp to tune it. God! Whoever saw how well he handled it, touching the strings and making them vibrate, sometimes causing them to sing and at other times to join in harmonies, he would have been reminded of the heavenly harmony.”
From folio 6v of a copy of The Somme le Roi
(British Library Add MS 28162), c. 1290–1300,
a moral compendium compiled in 1279
by the Dominican Friar Laurent for
King Philip III of France (r. 1270-1285).
(As with all pictures, click to see larger in a
new window, click in the new window
to further enlarge.)

Tuning the harp is both symbolic and practical. As demonstrated in the second article, harp tuning in medieval iconography symbolises the creation of harmony between heaven and earth, soul and body. We see the same idea in this passage from Horn: when Horn tuned the harp, the listener “would have been reminded of the heavenly harmony.” It is also obviously practical because the harp needs to be in tune with itself: the standard medieval account of a harper’s performance begins with retuning to suit a particular piece of music (more of which below).

The account of Horn’s playing shows that tuning itself was an impressive part of the performance. The description of him playing strings melodically – “causing them to sing” – and harmonically – “at other times to join in harmonies” – fits the exact format of five pieces called Taster de cordeTesting the strings – in Italian lutenist Joanambrosio Dalza’s print of 1508, Intabulatura de Lauto Libro Quarto (Lute Intabulations Book Four), which consist of brief, fast-moving passages up and down to check the melodic tuning, with spread chords to check harmonic tuning, just as in Horn. Though more than 300 years apart, the function of Horn’s “touching the strings” and Dalza’s Testing the strings is the same: a preliminary performance to test the instrument’s tuning in readiness for the main piece that follows. In Horn, this preliminary tuning performance takes several sentences to describe, culminating in, “When he has played his notes he makes the harp go up so that the strings give out completely different notes”, i.e. he has retuned the harp and completed his prelude to test that it is in tune; and “All those present marvel that he could play thus.”   

Musical training

“This man, of all that are there, causes the most wonder.” 

This sentence not only fits the narrative of Horn the hero, that he could play so astonishingly well, it points to the reality of both courtly and professional musicians serving long apprenticeships to master their art.

In Tristan by Gottfried von Strassburg (died c. 1210), a 12th or early 13th century Middle High German version of the Tristan and Iseult legend, the titular hero is a knight whose skill is matched only by Lancelot. Since he studied with Welsh harp masters from the age of 7, Tristan is also an exceptional musician, a composer of lais (explained below), a harper and singer. When he was 14, Tristan visited King Mark of Cornwall’s castle and was handed a harp to play. First Tristan retuned the harp, then “He played such sweet tones and struck the harp so perfectly in the Breton manner that many who stood or sat nearby forgot their own names. Hearts and ears began to lose touch with reality, like mesmerised fools, and thoughts were awakened in many ways … With determination and agility his white fingers went into the strings, so that tones were created which filled the whole palace. And the eyes were not spared, either: many who were there intensely watched his hands.” 

Harp, fiddle and citole, the three instruments played
by Betó at 7 years old in Daurel e Betó, 12th century.
From folio 14v of Walter de Milemete, De Nobilitatibus,
Sapientiis, et Prudentiis Regum (On the Nobility,
Wisdom, and Prudence of Kings), Christ Church
MS 92, Christ Church, University of Oxford, 1326–27.

In the 12th century Provençal tale of Daurel e Betó, it is clear that being a skilled musician and singer was one of the expectations of an all-round skilled courtier. “When Betó was seven years old he knew how to fiddle well, and to play the citole and the harp in a noble fashion, and how to sing cansos [courtly love-songs], and how to compose by himself … When Betó was nine years old he was a squire of the king, he was personable and courteous and an eloquent speaker. He plays drafts and chess where money is bet, and goes hunting with dogs and greyhounds … The king loves him, and the queen adores him greatly, and her courteous daughter holds him very dear; ladies love him, young men and knights. He served at table during meals and stood graciously in the presence of the king … then he fiddles to them and sings willingly.”

Tristan and Betó were figures of legend, but the expectations, training and lifestyle were real, reflecting the contents of the didactic courtesy books popular in the 12th and 13th centuries. It was then that the word courteous entered the language, meaning behaving in a refined manner as at court. That meant being educated from childhood to a high degree as fitting for one of three roles: a knight who gave military service to a lord or monarch in return for titles and lands; a chatelaine, the mistress of a household; or an ecclesiastic, meaning a priest, monk or nun.

Being a courtly land-owning noble, in the upper strata of feudal society, meant having the best of everything from birth – land, property, food, clothes, art, servants, and masters and mistresses of education to teach children dancing, singing, music, the military arts, and chess to learn strategy. For boys and girls, 7 was a significant age when they learned reading, writing and Latin, with boys’ education being more intensive and thorough than for girls. In addition, at this age boys began to learn hunting, horsemanship and how to use weapons, training which intensified at the age of 12 to 14, when boys learned to wear armour, handle swords and lances, and joust, in preparation for knighthood at the age of 18. Rather than fighting, girls learned riding, housekeeping, sewing, weaving, spinning, and needlework (embroidery and tapestry-making).

Folio 61r of Walter de Milemete, De Nobilitatibus, Sapientiis, et Prudentiis Regum (On the Nobility, Wisdom, and Prudence of Kings), Christ Church MS 92, Christ Church, University of Oxford, 1326–27.

Such were the ways in which the medieval elite promulgated their values and affirmed their social superiority, and such was the context within which, from the age of 7, Tristan began a 7 year apprenticeship with harp masters, and Betó, by the age of 7, could play the fiddle, citole and harp. 

Minstrels, meaning professional entertainers such as those with whom Tristan had his harp apprenticeship, likewise had an extended period of training for their occupation. John of Gaunt (1340–99), Duke of Lancaster, was “King of the Minstrels” in the Minstrels’ Court, a professional body centred on Tutbury Castle, run as a trade guild. The ordinances of 1380 state that minstrel apprenticeships had to be at least 7 years in duration; apprentices could not perform for money; and before the training was deemed complete a musician had to be examined by the court and passed as fit to be a paid professional. Similarly, in the 14th and 15th centuries, the regulations of the minstrels’ guilds of London and Canterbury reflected those of Tutbury, insisting on a 7 year apprenticeship; that no member could teach music to anyone except his apprentice; and before an apprentice could become a professional he had to be approved by the guild and be made a freeman.

In addition to this extended initial training, minstrels had what today we would call continuing professional development. Between 1318 and 1447, Franco-Flemish and German towns held minstrel schools at which celebrated music teachers gave specialist instruction and there were instruments for sale. For example, in March 1386, the Duke of Burgundy paid 400 francs each to 7 minstrels in his service “for the schools of this year and for buying instruments”.

The minstrel schools performed three other functions. First, as a job market, where musicians could make useful contacts and try to gain more profitable or better-suited employment; and second, the dissemination of new repertoire. For example, a royal account of 1335 states that King Edward III of England sent his minstrels “across the seas, to learn new songs”; and, in 1436, a judicial officer in Lille, French Flanders, gave several municipal minstrels money “to help them afford the journey to the schools of Cambray [now Cambrai, France], to learn new songs.” Third, minstrels travelled to learn the styles of playing of individual influential musicians or regional approaches to music. For example, John of Aragon, in a letter of August 1377 to the minstrel Johani de Sent Luch, wrote of his shawm player, Jacomi Capeta: “we will allow Jacomi to play in the manner of Sist [presumably an influential minstrel] as he wishes … and he has also learned the manner of Flanders; thus, he plays in one manner or in the other, in whichever manner should please us most.”

This fits the pattern of both Tristan and Horn. The legendary Tristan was from Parmenie, a mythical territory from the south-western tip of Cornwall to the Isles of Scilly, but he “struck the harp so perfectly in the Breton manner”. Similarly, Horn’s musical apprenticeship was with Welsh harp masters, but he mastered other styles: “he begins to play the aforesaid lai of Baltof, in a loud and clear voice, just as the Bretons are versed in such performances.”

Scordatura: transposition and musica ficta  

“When he has played his notes he makes the harp go up so that the strings give out completely different notes. All those present marvel that he could play thus.”  

The next part of the account of Horn’s playing tells us that his preliminary playing was to test the tuning of a harp that had been retuned “so that the strings give out completely different notes.” This retuning of the harp – I emphasise, not checking a static tuning, but retuning – is standard in medieval accounts. Practically, what being in tune meant for a medieval player was particular to an individual piece of music. Before we explore that, we need to understand the basic tuning of the medieval harp. 

Guido of Arezzo, an 11th century Italian Benedictine monk, mapped all available notes in medieval music theory, known as ‘the whole gamut’, from the Latin gamma ut, being the Greek letter gamma, Γ, used for the lowest note, G, and the syllable ut, the lowest note of a hexachord (a pattern or scale of 6 rising notes with the names ut, re, mi, fa, sol, and la). The gamut is shown below.

As we see, the gamut consists entirely of natural notes, with the exception of Bb. These notes are the foundation of the eight medieval modes. We see four of these modes, called the authentic modes, below right. (Authentic and plagal modes are explained in this article.)

(The span of notes in a mode were thought of in terms of successions of 6 note hexachords rather than 8 note octaves, but for the sake of simplicity and to save exploring another piece of theory unnecessary at this point, the modes on the right are presented in octaves.)

The notes of the gamut were called musica recta or musica vera, meaning right or true music. This means that any changed notes – flattened or sharpened from the gamut – were deviations. These changed notes were called musica ficta or musica falsa, meaning feigned or false music, a term used in several 13th and 14th century treatises by writers such as Johannes de Garlandia (De mensurabili musica, France, c. 1240), Johannes de Grocheio (Ars musicae, Paris, 1270s–1300) and Walter Odington (De Speculatione Musices, Evesham, England, before 1316).

Medieval music treatises show that harps were tuned to musica recta, the diatonic natural notes of the gamut, but without Bb, so changing from B♮ to Bb or vice versa required retuning. We see an example on the right from the Berkeley theory manuscript, written in Paris before 1361, which gives 14th century tunings for several stringed instruments, including a harp with 11 strings tuned a b c’ d’ e’ f’ g’ a’ b’ c’’ d’’ (or possibly an octave down, depending on the size of the harp, which isn’t given). We see strings for b and an octave above, b’, but not for bb and bb’.

By the renaissance, the harp was still tuned in the diatonic natural notes of the gamut. For example, in the first printed western musical treatise, Musica getutscht by Sebastian Virdung, 1511, Bb and B♮ are alternative tunings for the same string in the diatonic series, and in Martin Agricola’s Musica instrumenalis deudsch, published in 1529 and 1545, we see the same, demonstrated by an illustration, below.

Left: “The Tablature Applied to the Harp” in Martin Agricola’s Musica instrumenalis deudsch, 1529.
The illustration shows 26 strings, with 4 strings each labelled both b and h for b♭ and b♮,
showing that the same string was used for both notes.
Right: On the reverse of the same page, “The Tablature Applied To The Psaltery” similarly shows
b♭ and b♮ as different tunings for the same string in a diatonic series.
(As with all pictures, click to see larger in a new window.)

Despite the harp being tuned, according to the theory books, in a diatonic series, harp retuning was standard practice in accounts of performance, as we see in the following four examples.

i. In the Old English Apollonius of Tyre, c. 1000, “Apollonius took his tuning key and then began to stir the harp strings with skill and to mingle the sound of the harp with joyful singing. And the king himself and all those who heard him called out with great cries and praised him.”

ii. In Tristan en prose, a prose version of the Tristan and Iseut story written c. 1230–1240s in France, Iseut plays a lai, overheard by a harper who then offers to perform another for her. She hands him her harp: “take this harp, and tune it according to the music for your lines … the harper … then began to tune it according to what he knew would be necessary for the music he was about to perform.”

iii. In the 13th century Breton romance, Galeran de Bretagne, Galeran taught his love, Fresne, how to play a lai of his composition. “‘Fresne,’ said Galeran, ‘I have tried out my skill with a new lai and I am very keen to teach it to you at once’ … ‘Begin,’ said Fresne, ‘then I will harp and learn the lai on my instrument.’ Then he began to play, and she listened, studying the way he cast his fingers on the strings. When he had listened to the notes he tuned them with his tuning key so that they were perfectly accorded. The words and the music were sweet, and he sang and played the lai until she knew both the words and the tune; then she tuned her silver-stringed harp to the lai.” This last clause shows that Fresne, before she played the lai, needed to retune her harp to the correct scordatura – alternative tuning – for that particular song.

iv. In the late 13th century romance, Sone de Nausay, a minstrel named Papegais travelled to the French royal court to perform for the king. “When [Papegais] had taken the harp she went straight back to the king and said to him, ‘Sire, you will hear one lai in which everything is true …’ First she made the harp sound and made all the strings agree. Then she began to sing the lai.”

Woman tuning a harp on folio 77r of MSR–05, Alexander Turnbull Library,
National Library of New Zealand. This is a manuscript dated c. 1120–50 of De institutione musica,
a treatise on music by Roman senator and philosopher, Boethius, c. 477–524 CE.
Creative Commons Attribution 3.0 New Zealand (CC BY 3.0 NZ).

Since the harp was a diatonic instrument in an age of diatonic music, why did Horn need to retune so that “the strings give out completely different notes”? Why do we find retuning as a standard and normal pre-requisite of performance in so many medieval accounts? Repeatedly we see that the basic diatonic tuning of the harp was modified according to the demands of a particular piece of music. There are two reasons this was necessary: pitch transposition and musica ficta.

In the renaissance, the harp still had a basic diatonic tuning. In his Scintille di musica, 1523, Giovanni Maria Lanfranco gives a vital clue about the role of transposition – moving the whole pitch of a piece of music up or down – on the medieval and renaissance harp. Lanfranco gives the harp a range of 2 octaves in 15 strings, starting on C, then states that if a harp starts from F then the Bs are flat, that is to say it is a transposed harp, with all the interval pitch relationships maintained a fourth up. We see exactly the same pitch transposition in a late 13th century Middle English song in praise of the Virgin Mary, Edi beo þu heuene quene. As we see from the modern transcription below, it has a finalis of f’ and a fixed bb’, unusual in written medieval music, making it a transposition a fourth up from c’, at which pitch the Bs would be natural.  

The reason for pitch transposition is to suit the range of a singer’s voice. In the case above, if a singer finds Edi beo þu heuene quene too high with a finalis of f’, the whole can be transposed down to c’, maintaining all relative pitch relationships by the raising of bb’ to b♮’. If c’ is too low for the singer, then the same relative pitch relationships can be maintained with a finalis of d’ by raising the F strings to F#s and the Cs to C#s, and so on with other possible transpositions, depending on the preferred pitch for the singer’s voice.    

Any doubt that pitch transposition was a reality in medieval music is dispelled by the organistrum or simfonie (the names were interchangeable) in the Cathedral of Santiago de Compostela, Galicia, completed in 1188, shown below. The player on the right has access to 12 sliders, raised to make contact with the strings, shortening the vibrating length to create different pitches. By shortening the vibrating length of a string by half, the pitch is raised by an octave, so we see that on this instrument there are 12 pitches to an octave, plus the unstopped string before the raising of the first slider, which would presumably be a pitch of one semitone or one tone down from the pitch at the first slider. 12 notes to an octave can only mean one thing: this organistrum is fully chromatic in an age when music was diatonic. There is only one way to make sense of this: the instrument is designed for transposition, to play all modes within a single octave. 

I have modified this 12th or early 13th century drawing by turning it and the note labels
180 degrees for easy comparison with the Santiago de Compostela organistrum.

Above is a drawing labelled “organistrum” from a late 12th or early 13th century manuscript from the monastery of Saint Blaise, Germany. The original document perished in a fire in 1768. Fortunately, the manuscript had been copied by German theologian, historian and musicologist Martin Gerbert, who published it in his De cantu et musica sacra, 1774. We see that this organistrum has one diatonic octave from C to c, and has the separate notes and bb and b♮, shown as b and h. On this organistrum, none of the four diatonic authentic modes below left have the range of even one octave. In the mixolydian mode, for example, only the first four notes are available before the player has to change to the lower octave, if we assume the strings were in unison. It may have been that this instrument was strung in parallel octaves, removing the problem of missing notes, as the lower sliders would then have played notes at the lower and the upper octaves, but this is hypothetical and not shown. 

Whereas the Saint Blaise organistrum has 8 sliders in a diatonic octave, the Compostela organistrum has 12 sliders in a chromatic octave. If those 12 sliders play C to c, then the open leading note must presumably be B, or perhaps Bb (as that makes more musical sense than 13 notes from C to c#).  

The four medieval authentic modes. The tenor or reciting note is the characteristic note of the mode, around which the melody is typically formulated. (For more on modes, click here.)

The fully chromatic Santiago organistrum can only have been a transposing instrument, an ingenious solution to playing modal chant with a short note range. Above right we observe that not only are all four authentic modes now available by the use of semitones and transposition, all the available chromatic notes are in use across the four modes. This would not be thought of as an instrument employing an unusual amount of musica ficta, but an instrument playing musica recta in a transposed form.

Moving the same transposition principle to the diatonic harp, this seems to be precisely what Pierre of Peckham had in mind in his Lumiere as Lais (Light to the Laity), 1267: “One may change the settings by tuning different notes, and by different arrangements of variously placed semitones. By this means there is diverse tuning on the harp.” For example, as we see above, if a song in the dorian mode (above top left) is just a little high for a singer’s voice, it can be transposed down a tone by flattening the E and B strings (above top right); and if a song in the lydian mode (above left second up) is out of range for a singer’s voice, it may be transposed down a fourth or up a fifth by the retuning of the F strings to F# (above right second up); and so on.  

Thus the transposition of the harp starting on C to the harp starting on F described in Lanfranco’s Scintille di musica in 1523, which suggests the pitch-shifting principle stated in Pierre of Peckham’s Lumiere as Lais in 1267, is confirmed in the medieval period by the obvious meaning of the Santiago de Compostela organistrum in 1188.

The other reason for non-standard tuning or scordatura on the medieval harp is the application of musica ficta, to include notes outside of the diatonic gamut. By the early 14th century, general rules had developed to move further away from strictly diatonic music, described by James of Hesbaye (also known as Iacobus de Ispania, Jacques de Liège or Iacobus Leodiensis) in his treatise, Speculum musicae (The Mirror of Music), c. 1325. The rule was that when a stable interval between two voices – a unison, an octave, or a fifth – was preceded by any other unstable interval, then the closest approach from an unstable to a stable interval would be taken. 

Thus, as we see above, the movement between two voices of thirds to fifths or sixths to octaves would be major, and thirds to unisons or thirds to lower fifths would be minor. James of Hesbaye stated that if the closest approach from an unstable to a stable harmony was not written in the music, the singer would make it so in the performance – and therefore, we must conclude, so would an instrumentalist. This means that polyphonic music from the 14th and 15th century requires not only B and Bb, but also the ficta notes F#, C#, G#, and Eb. How could a diatonic harper play them?

The psaltery illustration in the Berkeley theory manuscript. (As with all pictures, click to see larger in a new window.)

There is clear evidence in the Berkeley theory manuscript, before 1361, that the author knew of additional strings being added to the diatonic psaltery. The psaltery illustration (right) and its commentary give a range of an octave and a major third, A to c#’. Longer double courses are arranged in twos, so we have pitches A B then a gap, c d then a gap, and so on. Within each gap is a shorter single course coming from the right. The pitches for the first octave are as follows, with the shorter strings in square brackets:

A Bb [B] c d [eb] e f [f#] g

The next octave is similar, but not identical:

a [bb] b c’ [c#’]

The strings on a psaltery sit above the soundboard, making it possible for a pin to be inserted for a shorter string, thereby marking it out visually as a ficta addition. On a single row harp, with strings suspended in air between the tuning pin and soundboard, such an arrangement of shorter additional strings is impossible, making the addition of extra strings visually indistinguishable from the basic diatonic notes. This probably explains why the medieval harp remained diatonic, and other solutions were sought for playing ficta. Even on the psaltery, the Berkley arrangement appears nowhere else in iconography, suggesting the idea was either short-lived, localised, or both.

As we have seen with James of Hesbaye’s principles for applying musica ficta, sharpening or flattening a note is contextual, dependent on the interval between the first and second voice. Therefore when, for example, both F and F# are needed in the same piece of music, the retuning of all F strings (for example) on a harp to ficta notes isn’t a satisfactory answer. We have to move into the renaissance for evidence of solutions for playing contextual ficta on a diatonic harp.

One solution is in the treatise, Declaracion de instrumentos musicales, 1555, by Franciscan friar Juan Bermudo. Bermudo describes the playing of Ludovico, royal harper to Ferdinand II, King of Aragon (r. 1479–1516). In the following passage, Bermudo refers to a cadence, the closing notes of a musical phrase, in which a string is raised by a semitone compared to the same note in other parts of the music. For example, in the dorian mode, the 7th note is C and the 8th note or tonic is D, so at a resolving cadence the leading note C is raised to C# for the closest approach. Bermudo wrote: “I was told that Ludovico, whenever he played a cadence, he placed a finger under the string and thus made it sound as a semitone. Great ability was required to do that”. This involves reaching up with one hand to the part of the string that lies vertically across the wood below the tuning pin, and pressing the string against the wood, thus effectively shortening the vibrating string length and raising the note by a semitone. This fretting technique can be seen in the video below.

Click the picture to play the video of Quhat mightie motione by Alexander Montgomery
(1540?-1610?), played on bray harp by Ian Pittaway. Fretting technique is demonstrated
at 28–31 seconds, and again at 1.27–1.31, 2.27–2.30, and 3.25–3.30. 

Another solution to needing the same note either flat and natural or natural and sharp in the same piece is to tune the string ficta in one octave and recta in the other, and play the phrase which includes the needed note in the available octave. This is present in Alonso Mudarra’s piece for vihuela, imitating Ludovico’s harping: Fantasía X que contrahaze la harpa en la manera de LudovicoFantasía 10 that imitates the harp in the style of Ludovico. This piece can be heard played by Ema Stein on vihuela in the video below, in which the final passage has lower f played against higher f#’. (For more on the vihuela, click here.)

Click the picture to play the video.

The solution of tuning musica ficta in one octave and musica recta in another is seen and heard in the following two videos of French estampies from c. 1300 in Manuscrit du Roi. In La quarte Estampie Royal below, octave swapping to accommodate musica ficta in a differentially-tuned octave is heard in puncta (section) 4, beginning 1.09, and puncta 5, beginning 1.29.

Click the picture to play the video, which opens in a new window.

In La Seste estampie Real below, puncta 4 requires both B natural and B flat, so the higher octave is tuned to B natural, the lower to B flat. The octave-swapping in puncta 4 begins at 1.20 the first time through, 2.58 second time through.

Click the picture to play the video, which opens in a new window.

Other solutions for playing ficta on a harp were found in the renaissance, such as the Spanish arpa de dos órdenes, harp of two rows, with a diatonic row of strings and an additional row including chromatic ficta notes. Such later developments are outside the scope of this present article on the medieval harp.

The harp’s repertoire and status

“And when he had done all this he begins to play the aforesaid lai of Baltof, in a loud and clear voice, just as the Bretons are versed in such performances.”

A lai or lay is a medieval form of sung love poetry which seems to have originated in Brittany in the 9th century, spreading to Wales and Cornwall and then to the rest of Europe by the 12th century, and especially associated with the Tristan and Iseut legend.

Over time, there emerged three distinct types. The 12th century lais of Marie de France are narrative. For these, the music has not survived, if they ever did have music. The lais in 13th and 14th century troubadour manuscripts are lyrical, complex and ambitious, each subsection of text with its own metre and its own musical setting. Then there are the simple lais of Arthurian romances, such as Trisan en prose, c. 1230, which have stanzas with equal-length lines and, when the tune is given, a strophic melody (the same in each verse). These Arthurian lais seem to have been written specially for the story.

Furtmeyr-Bibel, Bayerische Staatsbibliothek, Munich, BSB Cgm 8010 a,
folio 374r, 1465–70.

In all stories where lais are sung, they are performed solo by a singer-harper, as the following examples illustrate.

Guiron le Courtois, an Arthurian legend written c. 1240, describes the first ever lai being written by King Meliadus of Leonis, father of Tristan. In suffering secretly in his love for the Queen of Scotland, Meliadus “composed a poem about his love which was more wondrous and subtle than anyone had ever composed before, and he set this poem to music such that it might be sung to the harp … Meliadus called this poem which he composed for the love of his lady lays, as a sign that he wished to leave all other music. And you may be sure, all and some, that this was the first lai which was ever sung to the harp.”

In Thomas’ Tristan, c. 1170, Iseut “sings very sweetly, attuning her voice to the” harp, singing a lai with a voice that is “sweet and the music soft”.

In the early 13th century Breton romance by Jean Renaut, Galeran de Bretagne, Galeran is keen to teach his love, Fresne, a lai he has just composed. “Begin, says Fresne, then I will harp and learn the lai on my instrument.”

In the 13th century romance, Sone de Nausay, Papegais goes to the king with her harp to sing a lai. “First she made the harp sound and made all the strings agree. Then she began to sing the lai.”

The harp clearly had a special relationship with the lai, but it was not an exclusive association, as we read in the anonymous Provençal romance, Flamenca, c. 1250. A meal was eaten and then, “After the minstrels arose, each one wished to make himself heard. There you would have heard strings resounding in many tunings. He who knew some new fiddle tune or a canso [courtly love song] or descort [irregular song expressing disagreement] or lais pressed forward as best he could. One plays the lai of Cabrefoil on the viola [fiddle], and the other the lai of Tintagoil, one sang the lai of the Fins Amanz, and another the lai which Yvain composed. One brought the arpa [harp], another the viula [fiddle], one plays the flaütella [flute], and another whistles, one brings the giga [a different type of fiddle to the viola, or possibly a simfony/organistrum] and another the rota [two row harp with a sound-box between the strings, described here]”.

We know from the examples of lais above that the harp was an instrument for love songs. The harp’s religious symbolism and association with the biblical King David (an example of which is shown below – see the second article for explanations) meant it was also the favoured instrument of high-ranking clergy, who regularly had personal minstrel harpers, and the harp was used to accompany the ceremonial consecrations of bishops, abbots and priors (described in this article about minstrels).

King David playing harp on folio 10r of The Ormesby Psalter (Bodleian Library MS. Douce 366),
England, 1250–1330. © Bodleian Libraries, University of Oxford. CC-BY-NC 4.0.

Illustrations such as those below show that the harp was also associated with one of the troubadours and with dance music. On the left is Guilhem (de) Montanhagol, fl. 1233–1268, a Provençal troubadour of humble birth, depicted in Troubadour MS O, 13th century (Bibliothèque nationale de France, Français 854, folio 124r). Other troubadours are illustrated playing pipe and tabor, gittern, and fiddle. On the right is an illustration of three dancers accompanied by a single harper in Guglielmo Ebreo’s dance treatise of 1463, De pratica seu arte tripudiiOn the practice or art of dancing (Bibliothèque nationale de France, ital. 973).

At the bottom end of the social scale, in Geoffrey Chaucer’s Pardoner’s Tale of The Canterbury Tales, written 1387–1400, Chaucer associates “Singers with harps” with

baudes, wafereres,
Whiche been the verray develes officeres
To kindle and blowe the fyr of lecherye,
That is annexed un-to glotonye

prostitutes, cake sellers,
Who are the very devil’s servants
To kindle and blow the fire of lechery,
That is joined to gluttony

So while the harp had symbolic associations with holiness and had a special relationship with lais, that is of course not the totality of harp repertoire. There is no reason to think that then, as now, any instrument had a repertoire restricted by genre: any type of music can be played on any instrument, if the player so wishes. While some instruments were clearly more suited to some occasions – the shawm, for example, was especially associated with outdoor and ceremonial occasions – in the medieval period it is virtually unknown for any written music to indicate the particular instrument to be used. 

Prelude – song – postlude

“Afterwards he makes the strings of the instrument play exactly the same melody as he had just sung. He performed the whole lai, for he wished to omit nothing.”
Rutland Psalter (British Library Add MS 62925), folio 50r, c. 1260.

In his late 13th century treatise, Ars Musica, Parisian music theorist Johannes de Grocheio (Jean de Gruchy) stated that most non-devotional music is introduced with a prelude, and that he and others preferred it on the vielle (medieval fiddle): “amongst these [musical instruments], instruments with strings have primacy … And further, amongst all stringed instruments seen by us, the vielle is seen to prevail … The good artist generally introduces … every musical form on the vielle.” This is probably the same as Horn’s harp prelude to test tuning, which continued in renaissance lute performance, as discussed above; and it is certainly the same as described in Dominican Friar Laurent’s moral compendium, compiled in 1279 for King Philip II of France, in which Laurent compares his own prologue to the Lord’s Prayer to a prelude played on the fiddle.

After the prelude, Horn sings “the whole lai, for he wished to omit nothing.” How long this took depends on which type of lai we think he played – the unknown original Breton form; the long and complex troubadour form; or the shorter, simpler lais of romance manuscripts. The statement that he sang “the whole lai, for he wished to omit nothing” suggests this was a song of substantial length which he did not edit down. Some surviving lais, such as those of Marie de France, in which she openly cites Breton inspiration, were 600–1000 lines long, which I estimate would take between 45 and 70 minutes to perform, not including the instrumental tuning-checking prelude and the postlude using the same melodic material: “Afterwards [Horn] makes the strings of the instrument play exactly the same melody as he had just sung.” The clear significance is that a lai performed this way was an absorbing cultural event, taking the audience on a dramatic emotional journey that took time to unfold. This was not a performance similar to a 3 minute pop song, but more akin to a full stage play, a novel or a feature film. 

Musical arrangements

Finally we turn to the medieval harper making two-handed polyphony from a single-line melody, as the present-day harper must so often do from medieval sources. Thomas describes Horn’s harping as “asquantes feiz chanter, asquantes organer”, “sometimes causing [the strings] to sing, sometimes organer [harmonies]”. It seems so obvious that two hands on a harp would be used for melody and harmony that it hardly needs stating, but what was harmony in a medieval context?

The word used in medieval sources is organum, meaning a second polyphonic line accompanying the leading melody. Some types of organum are described or implied in sources specifically for the harp, while other polyphonic techniques used in vocal music would logically have been applied to the harp, as follows.

The vocal drone above or below the melody was a form of organum described as common in Rome by Guido d’Arezzo in his Micrologus de Musica, 1026. In the anonymous chant treatise, Summa Musice, c. 1200, the vocal drone is called “diaphona basilica” – basilica usually refers to the Roman ecclesiastical tradition. Similarly, on a range of medieval instruments, the bourdon refers to a drone note. The bourdon is the name for the continuous drone pipe of a portative organbordonus organorum; the continuous drone pipe of a bagpipe; and the plucked or bowed drone string of a vielle (medieval fiddle). In Marin Mersenne’s Harmonie universelle, Paris, 1636, the bourdon is the plucked drone course of a psaltery, tuned to G, a fourth below the c which begins the diatonic series, as we see below left. Though this is a 17th century source, the pictorial evidence from several centuries before suggests this was a medieval practice, as we see below right: the freely moving hand coupled with the hand plucking on the edge of the psaltery strongly suggests a melody-over-drone playing style. 

Left: The psaltery with its bourdon in Marin Mersenne’s Harmonie universelle, Paris, 1636.
Right: One of several clear depictions of melody-over-bourdon-drone technique.
These two psalteries are in the Códice de los músicos manuscript of the Cantigas de Santa María,
1257–83 (Real Biblioteca del Monasterio de el Escorial, RBME Cat b-I-2, folio 89r).
(As with all pictures, click to see larger in a new window, click in the new window to enlarge again.)

In 13th century Anglo-Norman, the bottom string of the harp was likewise called the bourdon, suggesting the same technique for the harp as for the psaltery. In a 12th or 13th century unpublished commentary on the Psalms in the Bodleian Library, Oxford (referenced by Christopher Page, 1987, p. 120), the playing style of the cithara (harp) of the Psalms is described: “two hands are used in playing the cithara; one hand continually plucks the lower strings; the other hand plucks the higher strings, not continually, but at intervals and in turn.” Again, the obvious meaning is a melody over a drone. That the drone is played on “the lower strings”, plural, indicates more than one bourdon or drone string, and this is confirmed by Pierre of Peckham in his Lumiere as Lais, 1267.

The medieval harp technique of melody of over drone is demonstrated the first time through the tune of La quinte Estampie Real from Manuscrit du Roi, c. 1300, in the video below.

Click the picture to play the video, which opens in a new window.
This is the same video that heads this article. 

Other forms of organum or medieval harmony may be summarised as follows. Each technique may be used on its own, or as part of the menu of options within a single piece of music. The medieval options were:  

• heterophony: follow the melody line with some variations, such as adding in some extra notes or holding notes, e.g. in modern terms, where the melody has three crotchets, hold a dotted minim. This is heard in the above video of La quinte Estampie Real in the lower octave, second time through.

• parallelisms: meaning parallel fifths or octaves. This is heard in La quinte Estampie Real second time through.

• fifthing: each note of the organum voice is a fifth or an octave above the cantus superior (leading voice), with no more than two adjacent notes in consecutive parallel octaves or fifths. This is demonstrated in this video of Cantiga de Santa Maria 344 at 1.08.

• contrary motion: the organum voice moves contrary-wise to the cantus superior, i.e. when the lead voice moves up, the second voice moves down. Heard in La quinte Estampie Real above, second time through.

• “minstrelish” organum: the second voice follows the melody with runs of additional notes, similar to what was known as divisions or diminutions in the renaissance. Shown in La quinte Estampie Real video, third time through.

• gymel, in English or Scottish medieval music only: the organum voice is in parallel thirds or sixths, or is constantly reaching for parallel thirds or sixths with passing notes between. This is used, for example, in the song Foweles in þe frith.  

All of these techniques are described in detail, with medieval references, musical examples, soundfiles and videos, in the article, Performing medieval music. Part 2/3: Turning monophony into polyphony.

There remains one question, which is whether medieval instrumentalist-singers always sang and played at the same time. It may seem like an odd question, but there is evidence of a playing style known as alternatim, clearly stated in a 13th century gloss on a 5th century work by Martianus Minneus Felix Capella, a writer whose prose, poetry and descriptions of the liberal arts were hugely influential in the medieval period. The work is Capella’s De nuptiis Philologiae et Mercurii (On the Marriage of Philology and Mercury), and the gloss states that “while they [minstrels] play their instrument, they stay silent; and while their instrument is silent, they sing.” This may mean these minstrels only sang a cappella and only played instrumentals, or it may be a description of a practice mentioned by Gerald of Wales in his Itinerarium Cambriae (Itinerary Through Wales), written in 1191 about events in 1188. Gerald describes travelling with a band of horsemen, preceded by a singer/fiddler “who replied to the notes of the song on his fiddle in an alternatim fashion.” This is the same performance style described in Claris et Laris, a 13th century French romance, in which a minstrel sang a song with a refrain and played his fiddle for the refrains only: “There they listened attentively to a minstrel, who sang them a song and performed the refrains on his fiddle, which was both good and beautiful.” 

Conclusion

Rutland Psalter (British Library Add MS 62925), folio 55r, c. 1260.

We have seen that, because of the feudal social context, the land-owning elite had the resources and financial opportunities to train as skilled musicians, and indeed such was expected as part of the courtly code. The standard of professional harp playing, with 7 year apprenticeships for minstrels, must have been very high indeed, with further opportunities for national and international travel to maintain top-rate skills and learn the most up-to-date repertoire.   

The medieval harp was quite unlike the modern instrument: smaller, no levers, no pedals, diatonic not chromatic, with manual retuning a standard part of performance to make it versatile and adaptable, capable of transposition and musica ficta.

Performance was also quite unlike the present day: none of the chordal arpeggiaton used by modern harpers (as there would be no chords in the modern sense until the 15th–16th century), but a range of polyphonic techniques based on two (or more) related voices, and some performances more akin to a musical drama in three parts than a modern 3 minute song.

 

© Ian Pittaway. Not to be reproduced in any form without permission. All rights reserved.

 

Bibliography

Bagby, Benjamin (2000) Imagining the Early Medieval Harp. In: Ross W. Duffin (ed.) A Performer’s Guide to Medieval Music. Bloomington: Indiana University Press. 

Fulton, Cheryl Ann (2000) Playing the Late Medieval Harp. In: Ross W. Duffin (ed.) A Performer’s Guide to Medieval Music. Bloomington: Indiana University Press. 

Gómez, Maricarmen (1990) Minstrel Schools in the Late Middle Ages. Translated by Barbara Haggh. Early Music, Vol. 18, No. 2 (May 1990), pp. 212-216. Available online by clicking here.

Grocheio, Johannes de (1270s–1300, modern publication 2011) Ars musice. Edited and translated by Constant J. Mews, John N. Crossley, Catherine Jeffreys, Leigh McKinnon, and Carol J. Williams. Kalamazoo: Medieval Institute Publications.

Haines, John (2005) Lai layout in the Paris Prose Tristant Manuscript. Scriptorium, Vol. 59, No. 1, pp. 3-28. Available online by clicking here.

Lyngby, Thomas (2008) Aristocratic Education in Europe. Available online by clicking here.

Page, Christopher (1980) Fourteenth-Century Instruments and Tunings: A Treatise by Jean Vaillant? (Berkeley, MS 744). The Galpin Society Journal 33, March 1980. Available online by clicking here.

Page, Christopher (1987) Voices & Instruments of the Middle Ages. Instrumental practice and songs in France 1100-1330. London: J. M. Dent & Sons.

Page, Christopher (1991) Summa Musice. A thirteenth-century manual for singers. Cambridge: Cambridge University Press.

Pittaway, Ian (2015) The Psilvery Psound of the Psaltery: a brief history. Available online by clicking here.

Pittaway, Ian (2018) Performing medieval music. Part 2/3: Turning monophony into polyphony. Available online by clicking here.

Pittaway, Ian (2021) “the verray develes officeres”: minstrels and the medieval church. Available online by clicking here.

Rastall, Richard (1968) Secular Musicians in Late Medieval England. A thesis presented to the Victoria University of Manchester for the degree of Doctor of Philosophy by George Richard Rastall. Available online by clicking here.

Schulter, Margo (2000) Hexachords, solmization, and musica ficta. Available online by clicking here.

Southworth, John (1989) The English Medieval Minstrel. Woodbridge: The Boydell Press. 

Leave a Reply

Your email address will not be published. Required fields are marked *

CAPTCHA ImageChange Image