Introduction to French Baroque Dance: Muffat on ‘Vrai Mouvement’

This article was written for a course on HIP for Harps taught for the Guildhall School of Music & Drama. It offers a very brief introduction based on Muffat (1698) and focussed on the typical movements of baroque suites.

Period discourse about music around the year 1700 was much concerned with contrasts in National Style, specifically Italian and French. Italian style (imitated also outside Italy, of course) favoured drama and virtuoso display in such genres as Opera, Toccata, Sonata & Concerto. French style (also imitated abroad) preferred descriptive character pieces to abstract sonatas and celebrated above all the noble art of Dance.

French theatrical music (especially Ballets), chamber music (in particular, Suites) and social activities were unified by the elegance and energy of dance, and depended for variety on the contrasting characters of distinctive dance-types. As modern-day performers of these repertoires, our understanding of the music is enormously increased if we know something of the dancing that inspired it.

I strongly recommend every student of Historically Informed Performance to go to class and learn some dances from the appropriate period of music. It isn’t necessary to become a great dancer: right from the beginning you will start to notice from the inside what it feels like to dance the music you love. No amount of teaching or demonstration can replace this personal, embodied experience.

At the very least, watch as much baroque dancing as you can, so that you have a clear visual inspiration to guide your playing. Play for dance rehearsals, in order to learn what their art requires of your delivery. The ideal in this period was that the music should appear to be produced by the action of the dancers’ feet striking the floor. Strong moments in the dance move upwards, preparatory energy is gathered by sinking in order to expand and rise again. And watching good baroque dancers, we can imagine that our sustained notes are similarly suspended in the air as if weightless, like a elegantly poised dancer, balanced and seeming to float almost off the ground.

Watch here 1: Introduction

Watch here 2: Suite

Watch here 3: Minuet etc

For a more substantial audio/visual introduction, I warmly recommed Paige Whitley-Baugess’ Introduction to Baroque D ance videos, available here.  

French period sources suggest that many subtleties of le bon gout – Good Taste – can only be acquired by studying with a fine teacher, born into the culture of Louis XIV’s France. As foreigners from the 21st century, we can all be thankful for Georg Muffat’s (1698) systematic analysis of French style, describing le bon gout in terms of a coherent set of principles, just as grammar-books describe the use of language. Indeed, this concept of a collection of rules is precisely how Art itself was defined, in this period. Read more about period philosophy of art here.

Muffat’s First Observations on the French style of playing dance-tunes according to the method of Monsieur Lully are presented in four languages (Latin, German, Italian, French) as the introduction to his second Florilegium collection, available free online here. The four versions are not identical, and it is worth studying fine points of detail across all four texts. David Wilson’s English translation is here. My summary below follows the French text.

For an alternative path through Muffat & French Baroque Dance, see my 2020 article here.


“Here you can discover the principal secrets in a few words”


“Two functions admirably well linked together:

  • “To charm the ear
  • “Simultaneously, to mark so well the movements of the dance, that one recognises immediately which type each tune represents, and one feels irresistibly inspired to dance.”

This is Muffat’s reworking of the classic Three Aims of Rhetoric: to delight, to explain and to move the passions. The musician’s purpose is literally to move listeners’ feet, and thereby to affect their emotions.

The word mouvement has a wide semantic field that includes the physical movements of dancing, contrasting formal sections (e.g. the movements of a suite), the speed of the music, the emotional Affekt of the music and the dancing, and the rhythmic structure of a particular dance-type. All these elements are interdependent.

“Five requirements:

  1. “To play in tune”
  2. “Bowing”
  3. “To keep constantly/constant the True Movement of each piece”
  4. “To observe certain usages of repeats, notations, style and dancing”
  5. “Ornamentation”

Muffat’s insistence on le vrai Mouvement  – True Movement – goes further than simply keeping the beat and maintaining constant tempo. This mouvement is also what a jazz musician would call the ‘groove’ of the dance, a characteristic rhythmic pattern, not necessarily strictly mathematical (often the first beat of the bar needs to be long), but established from the beginning and maintained until the end, and strongly linked to the particular physical movements and emotional Affekts associated with each dance-type.

For example, the Chaconne is usually a celebratory, festive, theatrical ‘party’ dance often marking the happy ending of a music-drama, or associated with the comedy clown, Harlequin. It is usually constructed in double-units of four-bar phrases featuring a descending bass-line, with hemiola at significant cadences, and a groove running across the bar-lines: 2 3 1, 2 3 1. The first beat is long, giving space either for a breath between mini-phrases, or for an expressive dissonance on the first beat resolved on the second.

The Minuet is a formal social dance, often marking the presentation of a couple to the assembled company. It is usually constructed with a great deal of symmetry: four-bar and eight-bar phrases; eight-bar or sixteen-bar repeated sections etc. The basic unit is two bars, which corresponds to one minuet-step. The groove mixes, often alternately,  rhythmic patterns of crotchet-minim [short-long] and minim-crotchet [long short].

Both these dances are usually notated in 3/4, and could plausibly be played within a similar range of tempi according to circumstances. In this, they might appear very alike. But once you’ve played a few of each type, and (ideally) learnt to dance them too, you will be able to distinguish them from the very first few notes, just as Muffat writes. This is the significance of vrai mouvement, much more than just ‘constant speed’.

“Play in tune”

Muffat singles out the diatonic semitone mi-fa as the usual source of problems for inexperienced players. At an elementary level, violinists have to learn to position their fingers to create a narrower spacing for the semitone than for the tones. Failure here is a serious assault on the listener’s ears.

At a higher level of sophistication, Muffat’s hint to raise the mi may be linked to the ongoing transition from the pure thirds of Quarter-comma Meantone towards the slightly wider thirds of Sixth-comma Meantone, as the accepted practice for ‘being in tune’. Most 18th-century ‘circulating temperaments’ (for keyboard instruments) were derived from Sixth-comma Meantone, so it is highly plausible that slightly wider thirds became generally accepted.

Muffat also mentions that ornaments should not be false. Sometimes ornaments require chromatic alteration to fit within the local harmonies, and whichever notes one chooses to play, they must be in tune, of course. Playing an ornament in the wrong place also offends the ear. Squeaks and noises are also to be avoided.

In contrast to the lengthy debates amongst today’s Early Musicians on the subject of Temperament, Muffat writes that there is only one accepted way of being in tune. He deals with the whole subject in 14 lines.


Muffat devotes about 100 lines – more than two pages, plus two pages of musical examples to this crucial topic. Bowing for string instruments corresponds to tonguing syllables for wind-players and fingering for keyboards, harp, lutes and guitars. Strict rules of style create characteristic patterns of articulation: Good and Bad notes, legato or separation between one note and the next, contrasting qualities of onset-attack for individual notes.  

Muffat states that unanimity of bowing is essential. This translates for harpists and others into a requirement for intense scrutiny of note-by-note articulation patterns.

In this French style, the first beat is always given a down-bow, even if the previous note was also down-bow. This creates silences of articulation before some down-beats. But Muffat marvels how, “in spite of so many down-bows and retakes” (lifting the bow up again, to facilitate two successive down-bows with the very short French Baroque bow), “one never hears anything disagreeable or coarse, but rather a wonderful combination of speed and the length of the bow-strokes; of admirable equality of measure and diversity of phrasings; of tender sweetness and vivacity of playing”

What I have translated here as ‘phrasings’ is yet another appearance of the word mouvemens, here suggesting the movement of the bow, as well as of the notes and of the dancers’ feet, and of the emotions that all these work together to produce.

This rule of “first-beat = down-bow” takes precedence. After this, Good and Bad notes get down- and up-bows respectively, as far as possible. In triple metre, three crotchets to the bar (for example), the last note could be taken down-bow (in slow tempo) or up-bow (in fast tempo). Two successive up-bows can be divided – craquer – to articulate the final note clearly. In very fast tempo, a group of notes can be played ‘upside-down’ if necessary.  In a passage of dotted notes alternating with short notes, one should not slur short-long, but might slur long-short.

If you have any skill at all on the violin, it’s worth playing through Muffat’s examples to see how they feel and sound. If not, you can create a similar effect by singing Frank Sinatra style with dooby-doo. Use ‘doo’ for a Good note, down-bow. Use ‘bee’ for a Bad note, up-bow. Advancing in sophistication, you can imitate craquer with the syllables ‘beeper’, making more or less of a seperation between ‘beep’ and ‘per’ as you judge appropriate.

Muffat avoids down-bow on the second beat, so the combination crotchet and two quavers at the beginning of a bar forces you to craquer the two quavers. Thus the famous Minuet from the Anna Magdalena Bach Notenbuch would not go “Doo dooby dooby / doo dooby” but (more elegantly) “Doo beeper dooby / doo beeper”, when played in the French style. 



Muffat gives a few examples of how Italian violinists played Minuets, often starting with an upbow on the first note. The Anna Magdalena Minuet comes out very nicely with alternate bows, starting up-bow, but sounding very different in that Italian style: “Bee dooby dooby / doo-bee-doo”


The mesure (a bar, yes, but also the time-span measured by the regular down-up movement of the Tactus hand-beat) can have different mouvements. A jazz-musician might express this by saying “a steady count can have all kinds of different grooves”. 

Muffat gives “three requirements:

  1. “Understand well the vrai mouvement  – the groove – of each piece
  2. “Once you’ve understood it, be able to keep it for as long as you play the same piece, always with the same regularity, without changing, slowing or rushing it.
  3. “Give certain notes some swing, to make it sound more cool.”

“To understand better the groove of each piece… knowing how to dance is a great help. Most of the best violinists in France are very good dancers, so it’s not surprising that they are so well able to find and maintain the groove of the beat.”

“Having understood and started the beat, not everyone is able to keep it precisely constant for the entire duration of the piece.”

Muffat does not accept playing the whole piece slower or faster one time than another [his next paragraph suggests that this refers to playing a dance several times through consecutively, rather than to separate performances on different occasions] He also disapproves of alterations to the groove bar by bar or note by note. 

  1. “Reject the abuse of playing whatever kind of piece the first time very gently, then gradually faster and faster, and the last time very fast and rushing”
  2. “Don’t wait at the cadence more or less than the note-values indicate”
  3. “Don’t rush the ending”
  4. “Don’t panic when you see short note-values”
  5. “Don’t shorten the last note of the bar”

Playing for dancers is an excellent way to learn how to ‘phrase-off’ and ‘breathe’ at cadences, without disturbing the vrai mouvement. Muffat’s 5th rule is equivalent to ‘Don’t crowd the downbeat’. 

Muffat defines precisely – “diminutions of the first order” – which note-values should be ‘swung’, with examples for various metres. A succession of short notes written as equal are performed long-short, approximately as if the first, third, fifth note etc were dotted, and the following notes shortened accordingly. We should keep in mind that a Baroque Dot is itself a variable quality, according to context we might over-dot or under-dot. The appropriate amount of swing varies with the dance-type: more vigorous for a fast dance with popular origins, more subtle for a slow, courtly dance. 

I consider that Muffat’s insistence on conserving the vrai mouvement implies maintaining the same swing for the duration of a particular piece, as jazz-musicians tend to do nowadays. Many of my illustrious colleages disagree with me on this, but it must be said that most of them choose not to maintain vrai mouvement at all. Muffat makes it abundantly clear that vrai mouvement must be maintained: but there is room for legitimate debate as to whether the ‘swing’ of notes inégales comes under this rule or not.

The complete rhythmic identity of a given dance – its characteristic vrai mouvement – is thus constructed on several levels. The slow count of Tactus, the mesure, is steady (as in all Baroque music, with the exception of préludes non mesurées and plainchant). The principal division of the bar (into two or three) also carries the groove. So a Gavotte typically has two minim beats per bar, and the principal division structures the groove as short-short-long (crotchet crotchet minim). If you tap your feet and clap to this groove, you’ll probably find yourself wanting to sing We will, we will rock you! The quavers are swung – waving your banner all over the place. The emotional power comes from the dance-energy, which is stoked by maintaining the count, groove and swing steady from beginning to end. Temps di Gavotte Anglais (1977) here.

Because they experienced Lully’s airs as dance-music, violinists of Muffat’s time were more likely to rush towards the end. Modern-day early musicians regard Lully as art-music, and are more in danger of applying inappropriate 19th-century rallentando. Muffat is crystal-clear: keep the vrai mouvement from beginning to end. And just as I do here, Muffat repeats this point many times (otherwise, he rarely repeats any of his remarks).


Good Delivery

  1. “Finish tuning before the audience arrrive.”
  2. “Dont make noise” nor practise your party pieces before the show starts
  3. “French pitch is a tone, or for opera a minor third, lower than German pitch”
  4.  “Balance up the band”, “don’t have everyone play first violin!”
  5. There are usually two viola parts: “viola 1 is better on a small viola than on a violin”. Viola 2 is played by a large viola. Muffat approves adding a double-bass, but the French were not yet using double-basses in dance-music in 1698.
  6. “Observe the repeats” (notice the French habit of a short repeat – petite reprise – at the end of the last section)
  7. “It is very useful for keeping the precision of the mesure to give each [downbeat] with a small movement of the foot, as the Lullists do.”


It is interesting to notice how difficult modern-day players find it, to tap their feet on the down-beat (and only on the down-beat). I recommend it to students, and frequently request it from my ensembles, just as Muffat does. 


Instrumental ornaments for dance-music are mostly derived from vocal ornamentation. There are many more than one would imagine, and Muffat gives only a brief introduction. Nevertheless, this is the largest chapter of his essay, occupying three pages of text and another three pages of music examples.

Pincement – lower mordent, starts and ends on the written note, usually descending by a semitone, usually short, usually without additional repercussions.

Tremblement  – short trill from above, starts from the upper auxiliary, can be simple, or turned, may end early or continue into the next written note

Both these are played on the beat.

Muffat describes many more ornaments and how to execute them. He then addresses the question of where each ornament-type can be applied. His ten detailed rules depend on whether the note is Good or Bad, ascending or descending, moving by step or leaping, with exceptions for a mi and special conditions for the first note of a piece, of a significant section, of an ascent or descent. At cadences certain notes require a tremblement, others refuse it.

He gives some examples of diminutions (improvised variations), and warns that two tremblements are generally not used in succession, though he lists specific exceptions to this rule.

Muffat asserts that the whole secret of French ornamentation is codified in his 10 rules. These ornaments bring the “sweetness, vigour and beauty” of the Lullian method. 

“The melody suffers if ornaments are omitted, inappropriate, excessive or badly executed. Omission leaves the melody and harmony naked and undecorated; inappropriate playing is rough and barbaric; excessive ornamentation sounds confused and ridiculous; poor execution sounds heavy and constrained.” 

“The slightest failure in ornamentation betrays the would-be Lullist as inexperienced in this style.”

Muffat’s approximately ten ornament types (it depends how you count the sub-types) and ten rules are an amazingly concise encapsulation of the bon gout of the subtle and elegant French Baroque style. And if you apply his rules to Lully’s (sparsely marked) orchestral scores, the result is strongly consistent with (very detailed) ornament-markings in D’Anglebert’s harpsichord transcriptions of those same scores. 

See also Quantz on ornamentation, here.

The Suite

Many Suites are not intended to be danced. It is acceptable to take a different speed (a complex piece of chamber music may need to be played slower than the corresponding movement would be danced), but whatever the chosen tempo might be, it is maintained throughout. In late 17th-century England, Mace details a practice of making pauses and then an a tempo conclusion to (fast) Sarabandes, but otherwise there is no period evidence to support the application of tempo rubato to Baroque dance-music.

The Allemande that begins many Baroque suites was not danced. I speculate that the title refers to a German way of playing with arpeggios (the modern term is style brisé) dance-music of the type that the French used for Entrées. Characteristic figures include the short upbeat “Ta-dah!” found as a head-motive in many dance-types, alternations of dotted and short notes, and a three-note upbeat figure.

The Courante was an old-fashioned, noble dance – ‘the dance of Kings’. It has the most complex rhythms of all, contrasting, combining and creating ambiguity between 3/2 and 6/4, with a “Ta-Dah” opening; groups of crotchet, dotted crotchet, quaver ambiguously accented on first or second note; and a decisive shift to 6/4 at significant cadences. 

The Italian Corrente is different, with continuous running notes in the melody, and without the complex cross-rhythms of the French type.

The Sarabande was a fast dance that slowed down over the decades. Choreographies are characterised by held balances and sudden spins or leaps into a new pose. Often the music has a similar sharp contrast in note-values and amount of activity. The groove has a strong and/or sustained second beat of three.

Gigues vary in speed and groove – see Quantz and others for details. French Gigues often begin with imitation between treble and bass, and have a strong sense of the upbeat. The Italian Giga tends to flow more continuously, and without marking the upbeats. 

The Loure is a slow-motion Gigue.

The Passpied is a high-speed Minuet.

Bourée and Rigaudon might have had their origins in popular, rural traditions, but had become a highly sophisticated, courtly protrayal of Pastoral. Looking at the musical notation, it is impossible for us to distinguish between the two types, but in the period they were sharply differentiated: we don’t know how. Two quavers on the upbeat, groove (often in the bass) with three crotchets and a rest, final bars with crotchet, two quavers, crotchet (or four quavers & crotchet) over that groove; all of this with strong duple (minim) count and vigorous swing on the quavers.

Period writers disagreed as to whether Passacaille and Chaconne could be distinguished, and if so how. You are in good company if you consider Chaconnes to be major mode, Passacailles minor, but perhaps the most famous Chaconne of all is from Bach’s D minor Partita. 

The Musette is a courtly imitation of a pastoral bagpipe tune, usually in 6/8. The Tambourin imitates a tambourine. 

The Sicilienne is a slow 6/8 with groups of dotted quaver, semiquaver, quaver. The Canarie is a fast 6/8 with the same rhythmic grouping.


If you are studying a dance-movement, I strongly recommend that before starting to “interpret” the particular piece at hand, you first become familiar with the general characteristics of that dance-type. So before going too deeply into Bach’s famous violin Chaconne, first play lots of (simpler) Chaconnes (and Passacailles), watch Chaconnes being danced, learn to dance one yourself, and generally make yourself at home with the identity of the Chaconne as a dance-type. 

Work through Muffat’s 10 ornament rules and apply them to your particular piece. 

You will now have a much clearer idea of how Bach’s composition resembles all Chaconnes, and where its particular individuality lies. Above all remember Muffat’s two essential functions: the listeners have to recognise the dance from the very first notes, and they have to feel inspired to dance themselves. Even in such complex and profound music as Bach’s, this spirit of the dance must live, energised by the constant flow of vrai mouvement.

It’s some 20 years since I recorded dance-music from Feuillet’s (1700) Chorégraphie. CD here. My research for that recording started me on the paths that I have followed since, of Rhythm & Rhetoric; Tactus, Text, Gesture and Ornamentation. And in the intervening years I’ve had the opportunity to play this repertoire with fine orchestras (both modern and early), and see it danced by experts. Nowadays, I would play some of the movements a little faster, and most of them with more dance energy, a little less chamber-music reticence, and with – I hope – a stronger and truer sense of mouvement.  

Writing this article also gave me the opportunity to re-read Muffat, and glean a little more detail of his bowing rules, resulting in ‘Doo beeper dooby’ above and a rewrite of the discussion of the same Minuet in my 2020 article, here. It’s always worth re-reading a source that you think you know already. What you have discovered since the previous reading will have changed your viewpoint, and you may well notice something that you previously overlooked. 



Baroque FAQs for Modern Musicians

This is the last in a series of articles following up classes on Early Music on Modern Harps that I taught this semester for the Guildhall School of Music & Drama, London. Although our case-studies come from harp repertoire, the principles we explored are relevant for any Historically Informed performer. This article could make a useful introduction for any modern instrumentalist or singer.

Previous articles in the series discuss Historical Principles & Online ResourcesPrinciples & Practice, Ornamentation and Dance Music. Our focus was on the 18th century (specific works by J. S. Bach, Handel, C.P.E. Bach, Mozart) and the principal sources consulted were the three Versuch publications around the middle of the century (Quantz, CPE Bach, Leopold Mozart), the Essai for harp by Meyer, and (back in 1698) Muffat’s remarks on French dance-style in Florilegium Secundum. Links to all of these sources and more, in the previous posts.

The questions below were asked by students in the final class, and/or arose from work-in-progress recordings of their baroque pieces that they sent me for private comments. Whereas in previous articles, the agenda was set by the historical priorities of period sources, in this post the questions were posed by today’s students. This is a significant distinction: what we today think is a high priority may not have been so important back then. It’s always good to assess from historical sources how significant your question was, in the dialetic of the period.


What are Good & Bad notes, are they just loud & soft?

The concept of Good/Bad notes is fundamental to renaissance and baroque music, and is given a lot of attention in historical sources. The underlying principle is that instrumental music imitates the human voice, playing as if the music had a text. In vocal music, the sung text is of paramount importance. Caccini (1601) writes that Music is “text & rhythm, with sound last of all. And not the other way around”. The structure of each mid-18th-century Versuch is a short introduction to musical fundamentals, followed by a large section on what Early Musicians call “articulation”: how to start a note, how to join or separate notes into short groups. For flute, this articulation is done with tonguing syllables; string instruments do it with bow-strokes; keyboard and harp do it with fingering patterns. This is a high priority question for period writers. See Principles & Practice.


Good & Bad Syllables

Good/Bad notes in music correspond to Good/Bad syllables in speech. In music and in poetics, these syllables are sometimes called Long/Short: Good is Long, Bad is Short. In modern terms, we would say accented and unaccented syllables. In the mediterranean languages, the accented syllable is not hit suddenly on the intial consonant, but gets its accent from a sustained weight on the vowel: this corresponds to Leopold Mozart’s description of a slow start to the bow-stroke, even on a loud note.

Thus, baroque violin teachers will often coach modern string-players to use a slow bow-stroke where an “accent” is needed. Similarly on the low-tension strings of early harps, a Good note can have a slow finger-movement. This is not so easy to apply to modern harp, where the heavy strings need a certain amount of snap in the finger-action. But imagining that the note has a slow bloom, rather than a percussive attack is already very helpful.

Comparing Good/Bad to language gives us the clue that it does not have to be exaggerated: it just has to be the right way around. When we say the word “around”, we do not make a large, or conscious accent on the second syllable. But we would notice immediately if someone accented the first syllable instead. This is what is needed for our Good/Bad notes too.

Good & Bad Beats

During the 18th century, the idea developed of an intrinsic heirarchy of the bar. Today, we learn this in our elementary music education. In common time, beat 1 is strong, beat 2 is weak. Beat 3 is medium-strong, but less than beat 1. Beat 4 is weak, or could be energised as an upbeat. This is the basic shape of Time, although particular pieces will make artistic variations around this underlying structure. The principle extends to sub-divisions of the beats: ONE + two + THREE + four + And to the next level of subdivision:  ONE a + a two a + a THREE a + a four a + a. In 3/4 time: ONE a + a two a + a THREE a + a.

Good/Bad is definitely not forte/piano. But there is something of Long/Short about it, in two inter-related ways: how long is the note, and how long is the time-space it can occupy.

If we think about the repeated quavers in the Left Hand of CPE Bach’s Sonata, we could beat Tactus as quaver-down, quaver-up. These gives a pair-wise groove of Good-Bad. Time itself has this groove, so that ONE is imperceptibly longer than +. This intrinsic hierarchy of the bar gradually becomes the main focus of 18th-century discussions of Good/Bad, for example in Marpurg (1755).

Good & Bad Notes

Meanwhile, the notes we play into this grooved Time have a patterning of their own, the ONE is definitely a long note and the + is a short note. This relationship between notes was the focus of 17th-century discussion of Good/Bad, for example in Muffat  (1698).

These two effects combine so that ONE is a long note fully occupying a long space; whilst + is a short note only partially occupying what is anyway a shorter space.

Quantz gives two ways of counting a slow 3/4, in quavers or in crotchets. If we count in crotchets, the groove is ONE two THREE, or Long Passive Short. So the downbeat quaver is a long note in the longest space; beat two has a long passivity; beat three is a long note in a short, actively upbeat space. All the offbeat quavers are short/bad. We could pronounce as a mantra something like the words “PLAYer, Silence, BEATer” to get the feeling of the combination of pairwise quavers with triple-metre crotchets.

And we need to practise the Left Hand, with any continuo realisation we might add, until this fundamental rhythm is absolutely correct.

Whilst it’s easy to grasp the intellectual idea of Good/Bad, it needs lots of practice to acheive it effortlessly and without exaggeration. That practice is training the ears to listen for Good/Bad and to spot any wrong-way-around relationships; and training the fingers to execute the phrasing as if automatically, and at a very subtle level. Ears and fingers must be trained in partnership.

A particular case of Good/Bad, and similarly linked to the scansion of poetry, is the idea that the last Good note in a phrase has the Principal Accent. Usually, this is not the very last note of the phrase, one or more Bad notes follow. A useful general rule therefore, is that for almost every phrase, the Last note is short and un-accented.


How to create ‘mini-phrases’?

In Baroque music, long passages of semiquavers are not ‘moto perpetuo‘, but are built-up from many short phrases. CPE Bach calls these Figuren (figures) the most short-term units (say 3 to 5 notes), and Gedanken (thoughts, ideas), perhaps linking two or three Figuren. One passage of semiquavers may contain several Gedanken, each containing several Figuren. Just as in Rhetorical Speech, we need to join together what belongs together, and separate each group of notes from the next group. These words occur very frequently in the Versuch, this is an important concept in this period.

Typically, this joining and separating creates rhythmic patterns that are maintained until there is a clear change. But from one unit to the next, even whilst the basic pattern is maintained (i.e. the same number of notes starting with the same relation to the Tactus, on-beat, after the beat, or before the beat) the sequence often continues by contrasts. A legato group is followed by an arpeggio group, a staccato group etc. See Principles & Practice.

Useful guide-lines are: “Last note short“, “Breathe after the one“, “Stepwise motion ~ legato, jumps ~ staccato“. A jump can also show the place for a mini-break. The mini-phrases are defined by mini-breaks, often between two successive semiquavers: the Tactus beat in crotchets or minims continues without faltering.

If the notes are not whizzing by too quickly, it may be possible to shorten the last note of a mini-phrase by damping, create an actual silence, and start the next mini-phrase with the appropriate Bad or Good articulation.

In allegro semiquavers, there will not be time for this. But the separation between one mini-phrase and the next can be communicated with an unaccented last note of the old phrase, a sliver of time for a mini-breath (but without disturbing the Tactus), an energised re-start of the new phrase, and a clear sense of repeating a unit, and of any contrast between the previous unit and the new one.


What about historical fingering?

This is another crucial concept for this period. After a short introduction, CPE Bach’s Versuch devotes almost a third of the book, pages 15-50, to fingering.

For harps and keyboards, 18th-century fingerings often clarify join/separate: the principle is to move the hand only in the mini-breaks, and keep each mini-phrase ‘in the hand’. This principle is utterly different from the modern concept of fingering, which seeks to make a passage as safe and efficient as possible. On the contrary, historical fingerings introduce deliberate ‘inefficiencies’, in order to discourage smooth joining of what is supposed to be separate.

In the following examples of harp-fingerings, I apply the principles of historical fingerings (from Meyer 1763 – see Online Resources – and  – specially recommended, and now available free online – Cousineau 1784)  to examples from CPE Bach’s harp Sonata and Mozart’s flute & harp Concerto.

[My Cousineau link takes you to the second ‘imperial’ edition, c1803. The Fuzeau facsimile publication states 1784 for the first edition, the US Library of Congress (who have online images of each page) says ‘1786?’ The title pages are undated. At the time of writing, an original second edition was being sold for €1,000]

An efficient modern fingering for CPE Bach second movement facilitates joining the third note D to the next G, with a hand-movement before the B [as shown by the square brackets].



The guideline “breathe after the one”  would suggest a separation after the D, making three upbeats to the middle of the bar. This is supported by the Figur in the LH, which has three upbeats at the end of the bar. So my historically informed fingering moves the hand “after the one”.

EXAMPLE 2 CPE Bach/Cousineau


In red, I show the Abzug (phrase-off, forte/piano, see below) in the Appoggiatura, recommended by many sources. CPE himself says that it is the most important element. Quantz gives detailed dynamic contrasts for each note within ornaments. Leopold Mozart instructs violinists to hineinschleifen (sneak into, slide into) the main note (piano).

After the second appoggiatura, we should also observe the good/bad relationship of F#-G, especially because the ornamented F# is the Principal Accent of the phrase, after which the guideline applies: “last note short, no accent”.

For a scale, Meyer gives two alternative fingerings. If there is nothing else afterwards, the standard fingering jumps the thumb to make the long note different from the run of short notes. Notice that within the scale, the hand jumps “after the one”. This is his default fingering. The alternative, more familiar to modern eyes, can be applied when the notes are very fast, but it lacks the detailed phrasing of the default option.

EXAMPLE 3 after Meyer & Cousineau


But the alternative becomes preferable, if the top note is not to be distinguished as ‘different’, but joined into the scale, with a break “after the one”. See Example 4.

In this passage from the first movement of the Mozart, the first note of the scale (treble C) is on the beat, so it is a Good. The next note D is also a good. For flautists (after Quantz): “Di diddle”, for – old fashioned – violinists (after Muffat): Down, down-up. (Leopold Mozart would probably apply some interesting slurred bowing). For harp, perhaps 4 4321321 encouraging a separation after the first note; rather than the ‘more efficient’  4 3214321, which would join irrevocably after the first note.

EXAMPLE 4 Mozart




I have adjusted the beaming. The fingering follows the smallest units of Figuren, and // marks the caesura between one Gedanke and the next.

The pattern of “breathe after the one” continues with a caesura after the high a, facilitated by fingering, and similarly after the g in the third bar. But the music imposes a new pattern, also clarified by my historically informed fingering,  at the beginning of the last bar. Red f_p shows two more examples of Abzug.

Between the 1760s and the 1780s, the standard Good/Bad descending fingering for harp 12323232 (familiar also from 17th-century Spanish harp technique) is gradually superseded by Join/Separate fingerings using all four fingers. You start with the thumb, and the last, lowest four notes get 1234. In between, you use as many fingers as needed for the number of notes you have. So a seven-note descent would be 123 1234.

The adjustment takes place at the upper end of the scale, so that the last, lowest notes use all four fingers 1234. This results in a distinctive fingering for a five-note descent, in which you hop the thumb: 1 1234. [Fully-fingered sources feature a LOT of repeated thumb-strokes in this period.]


EXAMPLE 5 Mozart/Cousineau


In Example 5, I apply Cousineau’s (1784) fingering principles to Mozart’s (1778) descending scales in parallel tenths: This fingering encourages “breathe after the one” between the two Figuren of the first bar, shown by my changes to the beaming. The octave leap indicates a stronger “breathe after the one” between two Gedanken, shown by my // caesura mark.

It would not be inappropriate to use ‘old-fashioned’ 32 descending fingerings. These would ensure correct Good/Bad relationships, but would leave the player to create Join/Separate between Figuren.

EXAMPLE 6 Mozart/Meyer

Contrariwise, ‘fashionable’ Cousineau-type fingerings (mentioned as an alternative by Meyer 20 years earlier, so certainly not excluded from Mozart’s Concerto) prioritise Join/Separate, and leave the player to take care of Good/Bad. As Leopold Mozart makes clear in his detailed instructions for varying the pressure from note to note, within a single bow-stroke, 18th-century music requires both Good/Bad and Join/Separate.

What about the Bass?

Period sources pay great atttention to the continuo bass. The second edition of CPE Bach’s Versuch has an additional and longer book, 355 pages entirely devoted to Generalbass, including a final chapter which extends realisation of a continuo-bass towards improvisation of a free Fantasia.

Modern harpists tend to focus on the right-hand melody, viewing the music from the top down. Baroque music is constructed from the bottom upwards: the bass is no mere accompaniment, but rather provides the fundamental framework of rhythm and harmony that defines the structure for the ornamental melody. The heritage of Renaissance polyphony is that music is woven from the strands of individual ‘voices’; each strand has its own integrity, character and logic. The typical texture of Baroque music is the polarisation of treble and bass, i.e. 2-voice polyphony with a continuo-realisation filling-in the mid-range.

From the beginning of the Baroque period (Agazzari 1607) to the transition into the Classical (Leopold Mozart 1756), period sources assign to the bass the role of maintaining Tactus.  The continuo does not follow the soloist, rather the bass creates a dependable rhythmic structure – like the rhythm section of a jazz-band. As with a jazz-band, it is acceptable for a baroque soloist not to be together with the bass, for the sake of elegant expressiveness around the steady groove: it is not acceptable for the groove to falter. See Monteverdi & Jazz. This is of course the opposite of today’s standard practice, even amongst most Early Music ensembles.

Harpists, lutenists and keyboard players must combine the roles of soloist and bass-section in one person. Modern players might need reminding to play the bass more strongly (as an equal partner), and to maintain the bass rhythm in Tactus (whatever technical challenges, complex ornaments, or expressive moments the melody might have).


My research in Consciousness Studies suggests that the optimal strategy could be to place one’s conscious attention on the bass, focussing on tight connection to the steady Tactus. Assuming sufficient advance practice, the melody can be better left to the unconscious mind, letting the fingers ‘do it for themselves’. Trills, for example, go better when you don’t think about them.  Like a hypnotist’s swinging pocket-watch, or a meditation mantra, the constant down-up of Tactus (physically enacted in rehearsal, or imagined in solo performance) entrains the mind into Flow.

The paradoxical instruction to “Listen more than you play” can help the mind find that state of consciousness where mindful Observing facilitates ‘personal best’ performance, without a conscious sense of Doing. In baroque music, you can achieve this by “being the continuo-player”, creating the rhythm whilst listening to the solo (even though, you are actually playing that solo yourself).

Imagining, or even physically beating, a complete Tactus (down-up) to start yourself off (i.e. give yourself “a bar for nothing”) is an excellent way to connect yourself to the power of Tactus, to the Music of the Spheres, as you start to play.


What to do with Long Trills?

In a word, practise. Long trills are described in detail in all the mid-18th-century sources under discussion here. Harp sources admit that they are difficult, and they are more difficult still on modern harp.

So practise. Practise trills non-metrically, with a long appoggiatura, and then repercussions accelerating from slow to fast and all the way into the final turn and last note.

Then practise this beautifully shaped trill, whilst playing a simple bass in crotchets. The bass maintains Tactus, the trill is not aligned note-for-note with the bass, but you find the last note simultaneously. If the trill is long enough, combine it with messa di voce. But don’t try to be super-loud whilst trilling, and don’t try for too many reiterations. Shapeliness in the trill, and Tactus in the bass, are the priorities.


Frederick the Great plays a flute concerto in Sans Souci Palace. CPE Bach accompanies at the harpsichord, Quantz looks on at his pupil’s performance.



What is Abzug?

This is another central concept in period discourse about ornamentation. Literally “pulling off”, Abzug is the forte/piano contrast between an appoggiatura and its main note.

Leopold Mozart describes it as sliding into, sneaking into the main note (see the music examples above). Quantz describes a slight swelling of the sound on the ornamental note (so not an aggressive attack, but a slow-blooming sound; for violin a slow bow-stroke), with a smooth, soft transition into the main note.

On lute, one could literally “pull-off” from the fingerboard with a left-hand finger, in order to play the main note without any plucking action of the right hand. On harp, we can imitate this with a slow but firm finger-movement on the ornamental note, and a very passive action on the main (second) note, avoiding any articulate start-noise whatsoever.

Practise it.

The same forte/piano effect is needed every time from dissonance to resolution, as well as for any melodic moment with a pair of notes that function like a written-out appoggiatura. In the first music example above, as well as the Abzüge marked in red for explicit appoggiaturas, a subtle version of the effect is needed in the second bar on the high c-b, a-g, and (especially) f#-e pairs, and on the b-a pair at the end of the previous bar.

You need Abzug again and again. CPE Bach considers this the most important element of ornamentation. Indeed, the entire repertoire of the Empfindsamkeit period is characterised by the sensitive gesture of Abzug: every piece is full of opportunities to apply it. A missed Abzug is like marching into San Souci Palace in your muddy boots – you have just trampled on what should have been an occasion for the most elegant sophistication.


Don’t forget to pull them off!


Appoggiatura onto a Triplet?

The standard rule is that the appoggiatura takes half of the value of the written note (two-thirds, if the written note is dotted). So the realisation of an appoggiatura onto a triplet divides the first note in half. But the more important element is – all together now: the Abzug. The appoggiatura itself needs a slow bloom, and the written note is soft; the remaining two notes of the triplet should be light, since they are Bad notes.  It should sound like “Play-a Trip-let”, not “Da doo-ron-ron”!




The (appropriate) tendency to lengthen the appoggiatura results in a rhythm that approaches the sound of a quadruplet, though still with the first note louder and slurred to the second. Some sources recommend this quadruplet realisation, others condemn it. Best practice is probably to keep some semblance of a triplet, but with a nice long appoggiatura and plenty of Abzug.


How to play a Short Trill?

There are lots of short trills in this repertoire, and longer or turned trills can legimately be simplified into short trills. So it’s a significant element of the style and a most useful skill to acquire.

The historical fingering is 2311, and the Abzug requires a decrescendo from first note to last. CPE Bach recommends you to schnellern (quicken, enliven) the first note, to make the ornament crisp and light. It should sound like “Tickle my toes!” and not “before the beat“.

EXAMPLE 8 Short Trill


Short Trills in Mozart

Practise this until you can fire off a whole chain of ‘flying short trills’ as Genlis (1802) teaches and Mozart requires. [The link is to the second edition of 1811].


Genlis’ second example (above) is not a realisation of the first example, but a preliminary exercise for those ‘flying trills’, at half speed and with extra time between each Figur.

As Genlis explains: ‘the two slurred notes are done by sliding the thumb on these two strings’. What I deduce from the third thumb stroke that follows each time (where one might have expected finger 2), is that after the two slurred notes, the sliding thumb comes to rest against the next string (continuing the movement onto the next string helps the slide flow nicely). At this point the exercise takes extra time, to teach you to apply a caesura here, before starting the next Figur. When you do restart, your thumb is already placed on the string you are going to need.

For the real thing, the full speed ‘flying trills’, each Figur starts with an upbeat, continuing the pattern of the first two notes. As one would expect from Muffat and others, the trills are on the Good notes – this is confirmed at the end of the sequence.



Mozart introduces his flying trills with a preliminary longer trill, turned so that its Figur ends on the second (crotchet) beat of the bar. The autograph staccato on this d indicates “Last note short”, allowing you to “Breathe after the one”. The staccato on the following c indicates it is an upbeat, and the Gedanke is now Genlis-style flying trills, each Figur having an upbeat to a Good-note trill.

EXAMPLE 10 Mozart/Genlis

This upbeat pattern continues into the next bar, which has rapid Alberti 64 harmonies in the left hand and bold downward leaps in the right (first you must “breathe after the one”), leading to a whole bar Long Trill over the same rapid Alberti pattern, now on the dominant seventh.

All these fireworks signal the end of the movement. After this comes the improvised cadenza (Quantz’s Easy and Fundamental Instructions show how 2 players can improvise together)  and final tutti.


Short Trills in Handel

There is a tricky short trill on a dotted note in the Handel Concerto. Although it is difficult to execute this correctly in the time available, it should start with upper auxiliary (not the main note) on the beat (not before), so that the complete Figur has the crisp sound of a demanding publisher: “Prrrint today!”  [the rolled r represents the repercussions of the trill] and not a lazy: “What about next week?”.


EXAMPLE 11 Handel


How should I damp?

This is a harp-specific question, and is discussed in several period Harp treatises, but with insufficient detail. The suggestions below are based on my personal experience.

For modern harpists, you might first consider threading a strip of felt through the very lowest strings – you don’t actually play these in Baroque pieces, and it might be better to lose the excessive resonance that they add.

Second, learn the Baroque way to damp by having your finger (and/or thumb) return to the string after playing (same finger, same string). This allows you to damp specific notes really quickly, rather than moving both hands to cuddle the strings and damp the whole instrument, which is very slow. You can damp individual notes or entire chords, in either hand.

Sometimes you can add rhythmic energy by damping where a rest is written on the beat. Damp crisply, precisely on the beat, even get some percussive noise from your fingers contacting the sounding strings.


EXAMPLE 12 Handel


Sometimes you need to damp to control bass resonance. If you damp between each note and the next, you produce a staccato effect: this would not be the optimum phrasing for movement by step.

But if you play, play the next note and then quickly damp the previous one, you produce a strong effect of legato.

You can mix these two ways to damp [legato, staccato] in order to create legato pairs, each pair separated from the next. This long-short sound is appropriate for Good/Bad.

EXAMPLE 13 Handel

The last note of any phrase could be damped, to make it short. If you play it without accent (as you nearly always should), the damping will be less abrupt, and might not even be necessary.

Any upbeat could be damped to create a “silence of articulation”, this throws the accent onto the next note.

Often you will need to damp to clarify a rising melody in the bass. This frequently applies at perfect cadences, if the dominant rises to the tonic; but it can also occur at the beginning of the phrase.

The bass cadence with an octave leap on the dominant implies staccati for that octave leap.

In every instance, you can adjust the damping [legato or staccato, and how much] to produce the most appropriate phrasing.

EXAMPLE 14 Handel

Combining all these techniques results in a LOT of damping, subtly adjusted, for various desired results. Such frequent damping is supported by the (limited) historical information available. The greater resonance of the modern instrument makes damping even more necessary than on baroque harp.

Damping with the left hand can establish the “groove” of a dance, or a dance-like movement. In the third movement of the Handel Concerto, the groove is the reverse triple metre, short-long, quaver-crotchet. You can make this energetic and clear by playing the downbeat strong and damping crisply, to produce a repeating groove effect that sounds like the words “Short Phrases”.

Notice how the semiquavers create a Figur across the bar-line, “breathe after the one”: both hands have a short note in the long space of the downbeat, but for different reasons.

EXAMPLE 15 Handel

All this takes practice. You need to train your ears and hands simultaneously, to hear the need for, and effect of damping, and to create the effects you want.


How to simplify Ornaments?

Period sources recognise that it is harder to play trills on harp, than on harpsichord. It’s even harder on modern harp than on baroque instruments. So it can be a great help to simplify ornaments. Certainly, it is better to simplify the composer’s ornament, than to omit it, to play it wrongly, to play the wrong type of ornament, or (heaven forbid!) to play an ornament without Abzug.

In place of a long trill with initial appoggiatura and final turn, you can make things easier for yourself with these three steps (in this order of application):

  1. Reduce the number of reiterations of the trill.
  2. Omit the final turn
  3. Omit the initial appoggiatura

If you needed to apply all three three steps, you will be left with a Short Trill, and you should have practised this sufficiently to be confident in it for any eventuality.

If you are really under pressure, you can convert a turned Trill into a simple Turn (upper auxilary, main-note, lower-auxiliary, main-note). Make the first (upper) note long and remember the Abzug.

It’s not so good to change a Short Trill into a simple Appoggiatura, because the Short Trill is meant to sound lively and brilliant, whereas the Appoggiatura should melt, languishing. A Turn could be a better solution: there are still four notes to play, but the fingers can manage them faster. For a fast Turn, try 1231, which should come out crisper than 1232.


How does Continuo-experience help one’s Solo-playing?

The great harpsichordists and composers of the baroque were also expert continuo-players: JS and CPE Bach lead the way!

The best way to progress rapidly as a harpist or keyboard-player studying baroque repertoire is first to acquire basic continuo skills. Playing in ensembles will inform your ears and mind, with the opportunity to hear the same fundamental principles applied in subtly different ways by different instruments and voices. Ensemble-playing also provides an energetic group dynamic and a supportive social group, and gives access to exciting large-scale projects. Don’t miss the chance to play in a baroque opera or orchestra.

As a continuo-player, you can adjust realisation to your (gradually increasing) level of skill, contributing something useful right from the start, without needing to be exposed as a soloist until you are ready.

For harpists, a single-action harp is likely to be accepted by HIP training-ensembles, even in 17th-century repertoire, and for a modern player presents less of a barrier to immediate gratification: double and triple harps are more challenging. It is to be hoped that an open-minded training ensemble would admit a keen student even on modern harp, either as a stepping stone towards baroque harp, or as a way to gather experience for solo-playing on the modern instrument.

The experience of playing continuo will transform your view of the role of your left hand. And the continuo-player’s view of ensemble music, from the bottom upwards, is the best approach to baroque solo-playing.

Familiarity with figured and un-figured basses will consolidate your understanding of baroque harmony, and help you recognise the character of dissonances and sequences: the excitement of rising 5 6, the subtleties 6 5 and 5b dissonances, the sweet melancholy of chains of 7s.

EXAMPLE 16 Handel


How can I give my performance more clarity and more character?

See above: Tactus, Good/Bad, Join/Separate.

For harpists: damping. For modern harpists, a basic position somewhat près de la table: for baroque harps, this position is standard.

For anyone: “Long notes long, short notes short”, and “Last note short, no-accent”. Ornaments on the beat. Contrast one Figur with the next.


How can I make my performance more expressive?

See above. Sensitise yourself to the flavour of each dissonance, and show the tension-release of each dissonance-resolution.

For harpists, move your fingers down, even more près de la table, for a dissonance, and up (higher than normal) for resolution. A basic position somewhat près de la table results in small changes down or up making a big difference to tone-colour.

Apply Abzug to appoggiaturas. Search for the particular character of each Figur.


Should I play marked Repeats?



Should I add Rallentando?


Muffat and Leopold Mozart clearly state that the same tempo should be maintained from beginning to end. There is historical evidence for rallentando, but not in dance-music, and perhaps only when it is specifically notated. It tends to occur where the note values get smaller and smaller at the end of a section; or where there is a final cadence after a silence (e.g. Hallelujah Chorus). Meanwhile, Leopold says simply, keep exactly the same tempo from beginning to end.

Remember, “what everyone does today” and “my favourite CD” are NOT historical evidence. Leopold Mozart is.

If you are keen to add rallentando, find a source to support your wish. [Student challenge!] But… also beware of the temptation to look into the sources to support a decision you have already taken. A better strategy is to read the sources with an open-mind, and then decide. If you read the whole of Leopold Mozart, you will have plenty to think about and apply, before you need to go looking for another source in order to explore exceptional cases and outlier opinions.



18th-century style calls for a enormous amount of short-term detail, many contrasted Figuren, many presentations of dissonance-resolution, and many, many Abzüge. All the while, you maintain the groove of steady Tactus in the bass.

Harpists: see my article on Empfindsamkeit and Single Action harp.

Historically Informed Performance is not what I say, not what Early Musicians do today, not what you hear on CDs, but performance based on historical information. Use IMSLP to get original scores, and use the mighty Versuch publications as reference books to answer your performance practice questions. Harpists: read Meyer, Cousineau and (for elite soloist-level skills) Genlis.

Try to establish a habit of checking what you are told (including what you have read here!), and checking your own assumptions. The state of knowlege advances when someone has the courage to question the status quo.

Dare to be different!




Vrai mouvement – Introduction to French Baroque dance-music

This is another in a series of posts following up a course on Early Music for Modern Harpists that I am teaching for the Guildhall School of Music & Drama, but it should serve any performer as a first introduction to French baroque dance-music and the movements of the Suite. See also Introduction to 18th-century Ornamentation, Principles & Practice, and Online Resources.


To a great extent, Baroque dance-music is French, and French Baroque music is dances. French style is also associated with the delicate subtlety of ornamentation, so that the energy and physicality of the dance co-exists with the intricate sophistication of precise control. As Muffat writes for violinists in Florilegium Secundum (1698),

‘In spite of so many retakes and down-bows, one never hears anything harsh or crude, but on the contrary one finds a marvellous combination of great speed and long bow-strokes; of an admirable consistency of Tactus and a diversity of movements; and of a tender sweetness and vivacity of Play.’

Muffat’s essay, originally printed in French, Latin, Italian and German, is probably the best period Introduction to this style, which he associates with Lully’s violinists. In spite of the preference of modern opera houses and baroque orchestras for Rameau (1683-1764)  and Rebel (1666-1747), Lully’s music remained the reference in 18th-century France. And the instrument most associated with the noble style of French baroque dance was the violin: the dancing-master’s minature pochette or the 24-strong violin band (with all sizes of violin-family instruments).

Even if you are not a string player, consideration of the implications of Muffat’s rules for Lulliste bowing is a fast-track to creating appropriate short-term phrasing (what Early Music players call ‘articulation’) for French dance-music on any instrument.

Each dance-type has its own characteristics, and in performing this repertoire, getting a feel for the family resemblance between all Menuets (for example) is more important than trying to ‘interpret’ the particular minuet at hand. Muffat again:

Concerning the different dance-movements, three things are required. 1: To know well the true movement of each piece. 2: Having recognised it, knowing how to keep it as long as one plays the same piece, always with the same consistency, without change of slowing or accelerating. 3: To adjust and compensate for the value of certain notes, for greater beauty.

Muffat’s vrai mouvement is much more than just the speed, though finding a suitable speed is important. Quantz (see Online Resources) gives tempi based on a notional MM 80 ‘pulse’ for various dance-types in Versuch (1752) from page 268 , and Saint-Lambert (Lully’s father-in-law) calibrates his indications to an average walking pace, see Les Principes du Clavecin (1702).

The French term mouvement also implies the Affekt, the emotional character, and (as Muffat’s requirements indicate) this depends on finding rhythmic subtleties and maintaining them consistently all the way through each piece. So in addition to the regularity of Tactus, in dance-music we have additional consistency of patterning within the Tactus. And this patterning is subtle – every note is not the same, smaller-note values may be unequal within the beat – but it is maintained consistently from bar to bar. We can think of this as the rhythmic “groove” of each dance-type: the pattern is distinctive, possibly assymmetrical, often subtle, and this pattern is established from the outset and kept strongly throughout.

There are four levels of rhythmic patterning. Often the whole bar corresponds to the early 17th-century concept of Tactus, and you can beat time one bar down, one bar up. This beat is equal and regular, though with the subtlety of arsis/thesis, see The Practice of Tactus.

Phrases are nearly always symmetrically organised in 4-, 8-, 16-bar groups, with repeats of each section. Don’t omit repeats, and don’t vary them either. Rather play the whole dance a second time, with repeats again, but in a varied version – French sources call this a Double.

Within the bar, the individual beats (often crotchets) have a characteristic organisation of good/bad and join/separate. So in a Sarabande beats 1 and 2 are Good; in a Chaconne one links together beats  2-3-1. These beats usually correspond to dance-steps, and the connection between feet and beat in French music led to a concentration on this level of rhythmic organisation. So the Menuet can also be beaten with an unequal (but reguarly maintained beat), 1-2 down, 3 up.

At the next subsidiary level of rhythmic organisation (often quavers), equally-written note-values are performed unequally, pair-wise, usually long-short. The amount of swing in this inégalité is crucial for establishing (and maintaining) the character of each dance: robust country-dances get a vigorous swing, sad noble dances get a very subtle swing. The bible of baroque swing is Betty Bang Mather Dance Rhythms of the French Baroque (1997).

Muffat’s word mouvement also reminds us that Baroque dances were not just music: there was dancing, too! The best way to understand any dance-type is to learn to dance it, even if you think you have two left legs! I would regard an introduction to historical dancing as an essential element to any HIP musician’s training – and as great fun, too! The standard introduction to the physical embodiment of this music is Hilton Dance and Music of Court and Theatre (1997).

Court and Theatre were the principal milieux for the noble style of dancing, but many of these dances had their origins in the street or the countryside. Mattheson describes the contrasting characters of various dance-types. You can develop your own feeling for the area of emotions associated with each dance-type by reading song-texts set to particular dance-metres, and simply by playing many examples of the type you wish to study.

For each dance-type, you need to have a feeling for tempo, metre (duple or triple), groove, social milieu, area of Affekt and typical dance-steps. Some dances are essentially stylised walking, others are mostly leaps, others mix leaps, spins and held balances. The New Grove Dictionary entry on a particular dance-type can be a good jumping-off point for further reading.

Dance-music was often published and performed as chamber-music in Suites, linked by a common tonality. The core of the baroque suite is the AllemandeCouranteSarabande group, often with a Gigue afterwards. A Chaconne might be added at the end; a Prélude or Ouverture at the beginning; Bourée, Rigaudon and other country-dances towards the end; and theatrical or programmatic pieces were introduced for variety. For social dancing, long sets of a single dance-type (especially minuets) were often needed.

Handel’s first opera, Almira (1705), listen here has a ball-room scene, set at a French-style Assemblée, in which a sequence of dances is interspersed with conversational recitatives and arias, a theatrical presentation of social dancing at court.

Case-study: the Menuet

The Menuet was a court dance, each couple would have to dance their formal minuet in front of the judgemental gaze of their aristocratic superiors, as they entered the hall of an Assemblée. The step is a stylised walk, and the dancers’ paths trace out geometrical patterns on the floor. There are also many theatrical minuets, and many pieces that feel minuet-like, even though they are not actually dances: the slow movement of Handel’s Harp Concerto would be an example.



Muffat’s rules for violin-bowing can help us find the vrai mouvement, the ‘groove’ of this dance, and I take Christian Petzold’s well-known menuet copied into the Anna Magdalena Bach Notenbuch as a case-study.



Baroque violins have lower string-tension than modern instruments. And French baroque violins had even lower string-tension. French violins were significantly smaller, but had lighter strings and were tuned a tone or a minor-third lower than in Italy. All these differences combine to produce very low string tension: it’s like playing on rubber-bands!  And to coax these slack strings into sound, they had very short bows.

Long & Short notes

At this point, you can experiment for yourself, by using a pencil as an imaginary, short French-style violin bow. To sustain a long note, you will have to be very sparing with the bow, and the string will take some time to ‘speak’. The result is a very drawn-out messa di voce, with a lot of intensity and a sensation of tension waiting to be released as you hope that you can get through such a long note with such a short bow.

For a short note, you’ll have to move the bow with a sprightly action, to get the floppy string to speak promptly – it’s almost like a bowed pizzicato. So the first result is that long and short notes are utterly different from one another: a long note is not just a short note sustained, it’s a completely different animal!

Bowing and inégalité

Muffat’s detailed bowing rules can be summarised as

1. Down-bow on the down-beat;

2. Down/Up bows for Good/Bad notes, respectively.

So French violinists would take the first note of Petzold’s minuet with a down-bow (Italians would play it Up). The next note is a Good, so it might seem also to require a Down-bow.  With a short bow, two successive Downs will require lifting the bow back Up again in-between (what violinists call a Retake), and this necessarily shortens the first note, creating a staccato effect. Nevertheless, this is acheived with elegant lightness, like a dancer leaping high but landing lightly.

However, in the fine detail of Muffat’s bowing rules, he gives precisely this rhythmic pattern (crotchet quaver-quaver) at the beginning of the bar, marked ‘down up push’. The preference for up-bow on the second note of the bar outweighs the desirability of down-bow on a Good note. Nevertheless, the downward leap of the fifth d’-g supports a detached first note.

The quavers that follow would be played pair-wise long/short, good/bad and down/up, quite legato within each pair, but with a small separation between one pair and the next. Within each pair, the second note is unaccented – the swing is gentle and elegant, not spiky!

Groove: le vrai mouvement

We  can beat Tactus bar by bar, down/up. This gives us the first level of equal movement, corresponding to the dotted minims that we find in the bass from bar 2 onwards. In general, we expect to find the fundamental rhythmic structure in the bass, and subdivisions in the treble.

We can also beat Tactus in crotchets, 1 2 down, 3 up. This gives us minim-crotchet unequal movement, that we see in the bass of the first bar and elsewhere.

The harmonic rhythm of bar 15 is the reverse of this: crotchet-minim. The mixture of these two patterns, long-short and short-long, is characteristic of the Minuet.

Baroque theorists linked these structural patterns, often heard in the bass-line of dance-music, with the metrical “feet” of poetic scansion. Long-short is Trochaic, and short-long is Iambic: the combination of these two feet creates the essential structure of the minuet’s vrai mouvement.

In the melody, the initial leap followed by stepwise movement produces a crotchet-minim Iambic structure for the first bar, with the minim sub-divided into swung quavers. So in this bar, the minim-crotchet, Trochaic bass has one of the Minuet’s two typical structures, whilst the melody has the other.

In bar 2, the bowing would be Muffat’s standard: down-up push. This might tend to create a joining between beats 1-2, and a separation before beat 3. But the downward leap again suggests a detached first note. Although I’m accustomed to hearing this bar structured Trochaically minim-crotchet, perhaps the downward leap should encourage us towards Iambic crotchet-minim again.

Bars 3-4 have the same structure in the melody as the first two bars. The ornament on the second note of bar 3 confirms the Iambic crotchet-minim structure of this bar, so similar to bar 1.

The next three bars have the note-values of the first bar, but without the initial leap. This suggests more legato between first and second note, whilst the harmonic shift on the second note of each bar implies a crotchet-minim Iambic structure.

In bar 8, Quantz’s rule for Appoggituras tells us to make the ornamental note two thirds of the length of the written note, and to resolve quietly and smoothly into the written note. The structure is therefore Trochaic minim-crotchet, breaking the pattern of the previous 3 bars. There are couple of bars with swung quavers all the way through in the melody and a Trochaic minim-crotchet structure in the bass, and the harmonies show the structure of the penultimate bar also to be Trochaic minim-crotchet.

There are no other patterns in this minuet. Muffat’s strictly maintained mouvement can be understood by superimposing all the allowed patterns, and ‘weighting’ them according to how often each is heard. You can listen to the result here.

As you listen, imagine yourself dancing with elegant steps and graceful balances along the floor, in smoothly curved patterns, wearing 18th-century courtly dress, and with the assembled aristocracy looking on, and subdued conversation in polite French, with period pronunciation of course. By now you are well on the path towards developing a feel for the vrai mouvement of the menuet.


Beyond Versailles


We find French dances in English, German and even Italian music, and of course in the music of Johann Sebastian Bach, see Jenne & Little Dance and the Music of J. S. Bach (1991/2009). Their first publication addressed works by Bach that bore the names of dances―a considerable corpus. In the second, expanded version they study also a great number of his works that use identifiable dance rhythms but do not bear dance-specific titles.

There is a glossy online Introduction to French Baroque Music presented in English by the Centre de Musique Baroque de Versailles. Their view appropriately contrasts French and Italian approaches, but they seem unaware of the richness of Spanish dance-culture, which brought together Old and New World, even African music, popular and courtly styles.

Hispanic culture contributed one of the most famous dances of Baroque France, Les Folies d’Espagne as well as the Canaries dance-type. As in France, so in Spain, Portugal and the New World, standard dance-types and (more than in France) the ground basses associated with them defined the territory for much chamber, theatrical and (also more than in France) even sacred music. Ribayaz’s 1677 book Luz y Norte offers a ‘guiding light and North star by which to explore all Spanish music’ – listen here.

English late-17th-century Country Dances became well-known in the 20th-century folk music revival. With simple steps and formulaic group choreographies, they were much, much easier for amateur dancers than the technically demanding solo dances in which French aristocrats emulated professonal theatre dancers. Country dances became popular in France as contredansesLes manches vertes is Greensleeves.

This article can only be a brief Introduction. The next step is to become familiar with various dance-types, by reading more about them, and – even better – by playing and dancing them.

For a slightly different take on Muffat and French Baroque Dance, see the 2021 approach to this subject here.

Introduction to mid-18th-century Ornamentation

This is another post related to a series of classes I am teaching at the Guildhall School of Music & Drama on Early Music for Modern Harpists: see also Principles & Practice and Online Resources.

I hope this article will be useful for any student approaching high Baroque and early Classical music. And before anyone even starts to think of exceptions to the simple guidelines I offer, let me emphasise that this is only an Introduction. Quantz has 12 pages on ornamentation for beginners (starting from p77), Leopold Mozart 59 pages of detail (from p193), CPE Bach 68 pages (p51 onwards), and even Meyer’s harp Method has two pages of text and four pages of music examples (including lots of arpeggios, of course). Links all these primary sources are in my Online Resources post.

So this short summary is necessarily simplified, but it is soundly based on these four mid-18th-century Essays. These mighty historical documents are pretty heavy going, if one tries to read them all the way through. Even a thorough survey of a general topic, such as Ornamentation, is a daunting project. But you can well use them as reference works, looking up the particular Ornament at hand and getting a quick answer to a specific question.

My focus here is on the mid-18th century, and the particular application is to modern harp. Fingerings, and some of my comments, are specific to harp, even to modern harp. But realisations and most of my comments should be useful as a starting-point for any performer.

Irish traditional music preserves a lively practice of ornamentation, which derives in part from local 18th-century styles. During and after the time of Carolan, the native tradition continued to flourish (even as it adapted to adversity), and available sources are fairly close (in time and milieu) to that tradition. [Inevitably, the information becomes more sketchy, as one goes further back in time]. So 18th-century Celtic repertoires (Scots and Welsh too) are ripe for exploration by today’s historical harpers, and I include some remarks on Ornamentation for Irish harp. Don’t apply these to European music!

Nomenclature is a challenge – the same ornament is given different names in different languages, and by different writers. And composers and printers use the same signs sometimes for quite different ornaments. So in this Introduction I use the simplest possible English names: if you have mastered Associated Board Grade V Theory, you will manage just fine.


Variations & Graces


There are two broad categories of ornamentation. Free variation, in which the player (spontaneously or with preparation in advance) changes the composer’s melody, usually by playing many short notes in the place of one long note. Such variations were called Divisions or Diminutions in the 17th-century, and in her 1802/1811 Method the Comptesse de Genlis calls them broderies (embroidery).

Improvised variations are beyond the scope of this Introduction, but Quantz’s Easy and Fundamental Instructions whereby either vocal or instrumental Performers … may learn how to introduce Extempore Embellishments or Variations as also Ornamental Cadences with Propriety, Taste and regularity were translated from his Versuch into English in 1780 – free download here.

According to Quantz and his translators, those Embellishments are the Productions of a momentary Invention or Fancy of the Performer, and in this Respect are different from those common Graces that are distinguish’d by particular Marks, such as Shakes [trills] and Beats [mordents] etc.

This article is concerned with ‘those common Graces‘ that might be marked in the score with signs, or should be added by the performer where necessary. Quantz and CPE Bach call them Manieren. These are what we normally think of today as Ornaments, applying to a particular note, rather than Variations that change the whole melody into different notes.


Ornament signs

German (and Austrian) 18th-century music explored a mixed style, influenced by earlier Italian and/or French aesthetics. Ornaments on a certain note, whether indicated by a sign or supplied by the performer, were regarded as part of the French heritage within the overall style. This fits neatly with the period characterisation of French style as subtle, tender, delicate, elegant, fashionable and balletic; as opposed to the directness, strength, passion, raw energy and drama of the Italian style. Thus there survives JS Bach’s handwritten copy of D’Anglebert’s table of ornaments from 17th-century France.



If your piece has ornament signs, you should not assume that they have the same meaning as modern signs, not even that they have the same meaning as signs from other historical sources. Many original publications included a specific table of ornaments, and you should look for a list of signs that is as close as possible to the piece you are studying.

Purcell’s 1696 table gives period English names for ‘Graces’: these names differ from modern terminology, and there are subtle differences in vocabulary between different sources even in the same language. Again, you should look for an explanatory source that is as close as possible to the piece you are studying.



JS Bach left a simplified table of ornaments for his 9-year-old son, Wilhelm Friedemann. This can be a good starting point for modern players.



For a particular piece or repertoire, it is well worth creating an ornament table of your own, using signs that give a visual representation of the ornament you have decided to apply. Write your signs into your score, and keep the table handy as a reminder, not only of the notes implied by each sign, but how to play them: fast/slow, loud/soft and fingering etc.

Jane Weidensaul’s edition of the CPE Bach Sonata for Harp applies information from his Versuch to suggest realisations of each ornament. This is a fine work of applied research, but it is only a first step. It fails to take into account differences between keyboard (the subject of the Versuch) and harp (the instrument for which the Sonata was written), or between baroque and modern harps.  Many of the suggested realisations are unplayable in an appropriate tempo. And the next step would be to apply CPE Bach’s and Quantz’s recommendations for subtle dynamic and timing contrasts within each ornament (see below).

This 2014 article by Colin Booth discusses ornamentation in JS Bach’s Goldberg Variations and is certainly helpful for the composer’s entire output, and as a discussion of the aesthetics of ornamentation for the whole period.


Beyond this Introduction


Amongst specialist performers and researchers, there is debate about changes in musical taste from one generation to another, from Johann Sebastian’s ornaments to Carl Philipp Emmanuel’s. That debate is beyond the scope of this introduction, and beyond the needs of most mainstream players. Indeed, one of the problems of today’s Early Music is that experts discuss abstruse corners of the field so passionately, examining exceptional cases and outlier opinions, with the result that historically informed (but non-specialist) musicians and mainstream performers can easily lose sight of standard period practice and the overwhelming historical consensus.

For Ornamentation, students will find a broad consensus between the four Essays discussed here, and need not worry – not yet, at least! – about subtle differences between CPE and JSB, or between Leopold and Wolfgang Amadeus, in their approach to Graces.

18-century Ornamentation for Irish harp has many similarities to European practices, and also some notable differences. There is a most interesting ornament table, supposedly based on 18th-century traditions, published in Bunting The Ancient Music of Ireland (1840). The two sections excerpted here resemble European Appoggiaturas and Trills, which Bunting categorises according to the period English names of fall and shake.


Ideals and Practicalities


There is a modern tendency to regard the harpsichord as the ideal of baroque music, to be emulated by other instruments. This is not unreasonable, for 20th-century harpsichordists and harpsichord-playing directors have been very influential in today’s Early Music, and we have the inspiring historical examples of JS and CPE Bach. But the sound of the harpsichord is certainly not a baroque ideal, for it is very far from the sound of the human voice (the philosophical ideal of all Baroque playing), and its mechanical nature limits the subtlety of its ornamentations. Probably the best modern-day examples of stylish ornamentation come from baroque flautists, applying all the subtleties of Quantz’s Versuch.

Listen here: CDs are not primary sources, but nevertheless I recommend listening to Laurence Dean’s flute-playing in mid-18th century repertoire, for example the Andantino from this trio Sonata by Georg Benda.

It is harder to play ornaments on baroque harp than on harpsichord, and 18th-century sources advise that harpists don’t have to play all the ornaments that a keyboard-player would execute. It’s even harder on modern harp, where thicker strings, higher string-tension and large-lever finger-movements work against speed and lightness in ornamentation. My advice is to reduce the number of ornaments where necessary, and to reduce the number of iterations in trills. In short: not too many twiddles!

But even modern harpists should add ornaments to the score, where they are essentially needed, for example at cadences (see below).

Amongst plucked-string instruments, lute-family and baroque guitar are able to realise the most elegant trills.

Listen here: I recommend Xavier Diaz-Latorre’s playing, for example this Chaconne by De Visée.  Notice that the resolutions of appoggiaturas and the iterations of trills are not  re-struck by the plucking fingers of the right hand, but are made by the left hand only. This is a subtle effect that harpists can only attempt to emulate.

The lower string-tension of baroque harps (French 18th-century ‘single-action’ pedal harps had especially low pitch and low string-tension) facilitates the speed, lightness and subtlety of ornamentation.

Listen here: Here is a Chaconne by Lully, with D’Anglebert’s principles of ornamentation applied, on 17th-century triple harp.

See also Single Action Harp: making Sensibility of the Méthodes.

Where to play What?


We might regard all these little twiddles as somewhat inessential. But some of them are part of the ‘grammar’ of Baroque music, and cannot be omitted. And if they are missing, they must be supplied.

We English speakers might regard the acute accent in the word café as a piece of French-style decoration, harmless enough, but not really essential. If we see cafe, we are neither confused nor offended. But for any Francophone, the é is essential: if it is missing, the word is wrong!  And so it is with French-style ornaments in Baroque music. Don’t go around saying “Kayf”!

The best historical Introduction to the French style of ornamentation in Lully’s time is in Muffat’s Florilegium Secundum (1698), as part of a general introduction to French baroque dance music in four languages: German, French, Latin, and Italian. Writing for ‘foreigners’ (i.e. not French), Muffat’s approach is very useful for us today, as ‘foreigners’ to this historical period. He gives detailed rules of which ornament to apply where.

The rules are indeed detailed. “It is uncouth to give a tremblement to an ascending good note… unless it is a mi or a note sharpened with #, which is almost always ornamented with a tremblement“. But in just 10 paragraphs, Muffat summarises “all the secrets of ornaments played a la francoise“. Highly recommended reading.

Some situations, in particular cadences, demand that the player supply an ornament, even if the composer has not notated it. Muffat: “At cadences, there are certain notes that demand a tremblement and others that refuse it”.

At a Perfect Cadence, with V-I harmonies, typical melodies require some kind of trill from the upper auxilary: Soprano Cadence (tonic, leading-note, tonic: trill on the leading-note); or Tenor Cadence (supertonic, tonic: trill on the supertonic). The Alto and Bass Cadences should not be given a trill.



See here for Cadential Shakes in Irish music.


Quantz and CPE Bach show instances where Appoggiaturas should be added, for example to melodies descending in thirds. We see such Appoggiaturas written, for example in the second bar of CPE Bach’s harp Sonata. It has not yet become standard practice amongst today’s Early Musicians to add these, but the historical evidence for them is clear. Read more in the Easy and Fundamental Instructions. 


CPE Bach’s melody descends in thirds through the principal notes G E C. Each is given an Appoggiatura, and the (longer) last note has a more complex Appoggiatura with Short Trill.


Repeated or varied ornaments?

Although Appoggiaturas are often repeated, as in CPE Bach’s example above, one element of subtlety can be the avoidance of an immediate repeat of precisely the same ornament. Muffat: “One certainly does not approve of two tremblements in a row”.

Instead, you can use a slightly different version of the same basic ornament type, a more elaborate or simpler trill for example. Usually, the basic type is defined by the situation and the degree of elaboration is up to you – see Muffat’s rules for details. Thus CPE Bach elaborates his third Appoggiatura, above.

Reluctance to repeat the same ornament seems not to be a feature of Irish 18th-century harp-playing. This transcription, based on the Forde MS, 154, shows the ornament that Bunting calls Striking Upwards applied three times in succession to the second strain of Ta me mo cholad, seen also in other sources for this tune.



In this context, the ornament seems to function as an Appoggiatura (perhaps slow) plus a Mordent (fast). Indeed, it looks like the mirror image, ascending, of CPE Bach’s elaborated descending Appoggiatura.

But Bunting’s description of Striking Upwards seems to indicate a brisk execution of the whole ornament. We might conclude that there can be subtleties of timing, even when an ornament is realised with the same pitches.



There are two, inter-related, questions of timing. How should we time the ornament within the note-value it is attached to? And how should we time individual notes within the ornament itself? Period sources gives us detailed answers.

Many sources emphasise that it is important to adapt your ornament to the note-value of the written note, and according to the tempo of the music. In general, ornaments should be longer and slower, if the note-value is longer; shorter and faster if the note-value is short.

For clavichord, with relatively little sustain, CPE Bach likes ornaments to fill up all the available space within the written note. Other sources leave the end of the written note plain: this works well on the harp with its long sustain (even more so for modern harp and historical Irish harp, with even longer sustain). On a dotted note, you can finish the ornament on the dot.

The timing of individual notes within the ornament is beyond the scope of this Introduction. But if the first note is an Appoggiatura, or functions like an Appoggiatura, it can be longer. The detailed information in Quantz and CPE Bach perhaps suggests a tendency to move from slow to fast within ornaments, which we can trace back to Caccini’s trillo in 1601. See this Introduction to ornamentation for Monteverdi’s period.

The most important timing rule is to begin the ornament on the beat, not before. You can practise this by playing a bass note, or tapping your foot, simultaneously with the start of the ornament.

There are some special case exceptions to this rule, and some outlier opinions in period sources and amongst 20th-century commentators. For today’s specialists, this is an area for debate and sophisticated subtlety, applied only in very particular circumstances. Read all 80 pages of Quantz’s and CPE’s remarks on ornaments, before you venture into this fascinating quagmire.

Long Trill


Begin on the beat, with the upper auxiliary (not the written note). Add a concluding turn if there is enough time. Add an initial appoggitura if there is enough time- hold the appoggitura as long as you can. A very long trill can start very slowly and gradually speed up.

Harpists – don’t try for too many reiterations!

Harpists, lutenists, keyboard-players – practise your ornaments with a bass accompaniment, to make sure that you start the ornament on the beat (as defined by the bass), not before the beat. Others can tap their foot with the first note of the ornament.

The alternative harp-fingering comes from Cousineau (1784).


We see something similar in 18th-century Irish Harp ornamentation, but using fingers 2324, without thumb; and beginning on the main note, rather than the upper auxiliary. See Irish Long Shake.

Lower mordent

Begin on the beat, with the written note. Play a slower ornament and/or add reiterations if there is time, and to have a gentler effect. Play fast and snappy, to make it bite.

Practise with a bass note or a foot-tap to define the beat. It should sound like the word “ORnament”, and not like “This is WRONG“!

The alternative execution from 18th-century Irish harp playing relies on the sustaining power of historical brass strings, or indeed of the thick strings of a modern harp. Two plucking actions and one damping movement create the illusion of three notes being played. Damp actively, a bit of string noise helps the illusion.



Short Trill

This has to start on the upper  auxiliary, so the shortest acceptable version has four notes.

Begin on the beat, with the upper auxiliary.  If there is more time, play a more gentle trill with more reiterations.

Practise with a bass note or a foot-tap to define the beat. It should sound like “RIGHT on the beat”, and not like “BeFORE the beat” nor “Before the BEAT“.

The harp fingering is from 18th-century French Methods. Slide the thumb from upper auxiliary to main note, moving the thumb itself, not the whole hand (too slow, too heavy).

The alternative execution is based on Irish techniques, but adapted (the Irish style for this ornament starts on the main note). It works surprisingly well, done fast and actively.




Many 18th-century sources define the Appoggiatura as the most important ornament of all. Luckily it is easy to play. As the Italian name suggests, “lean” on the auxiliary note, and ooze gently into the resolution, which is played softer.

Take the Appoggiatura on a long note, typically after shorter notes, and in the same direction (from below or above) as the approach to that long note.

Start on the beat. Sustain the appoggiatura for half the length of the written note (if it’s a dotted note, for two thirds of the length).



The most important thing about Ornaments


Quantz and CPE Bach concur that the most important element is the Abzug (literally, pulling away), diminuendo. An Appoggiatura is played with a little swelling on the auxiliary (louder still, if it makes a strong dissonance), and then gently and softly into the resolution.

In general, the use of loud/soft within an ornament gives lots of character. Often, ornaments go from loud to soft. The details are in CPE Bach and Quantz, and are most easily accessible for English-speakers in the period translation of Quantz, Easy and Fundamental Instructions (see above).

Subtle use of fast/slow within an ornament is also a vital expressive resource. The general rule is to go from slow to fast. The details are in CPE Bach and Quantz, and are most easily accessible for English-speakers in Easy and Fundamental Instructions.

Quantz gives a sample slow movement, Adagio with ornaments applied and links to his rules for realising them. It’s in the Versuch and included in the Easy and Fundamental Instructions too.


This is a very basic summary of a very short Introduction.  Handel with care! (sic)

  • Adjust to the tempo and note-value.
  • Start with the upper note.
  • On the beat.
  • From Loud to Soft [most important].
  • From Slow to Fast.


  • Cadences need trills


If you apply this summary you have made a brave start. Hurray! Now go and read Easy and Fundamental, because it is easy.

It is also Fundamental. So read it!






Ornamenting Monteverdi – Add, Alter or Divide?

How far do you and your ensemble go, when ornamenting music by Monteverdi and his contemporaries?


1) Go on, do some ornaments like on the CDs.

2) Not too much ornamention, [insert name here]!

3) Not THAT ornament!

4) Let’s study examples of ornamentation from period ornamentation manuals.

5) Let’s study how to apply those ornaments, by looking at scores and treatises.

I thought I was somewhere between steps 4 and 5, but in researching for this article, I began to realise that the typical approach of today’s Early Music is not just slightly off-target, it’s diametrically opposed to historical evidence. Even for such well-known works as Orfeo (1607) and the (1610) Vespers, our understanding of ornamentation needs a complete reset.


How should we ornament Cadences?

Often, the problem is expressed as a well-intended question:

What to do with all those cadences?


The Bass Cadence usually appears in the lowest voice. Tenor and Soprano cadences can be in any voice. The names Tenor and Soprano are used to identify the melodic shape, not the particular voice.



Cadence in La Musica

“Tenor Cadence” in the soprano voice. “Bass Cadence” in the continuo


In particular, in dramatic, sacred or courtly monody (let’s not muddy the waters by calling it Recitative  more here), how should we ornament what seem to be over-frequent cadences in long notes (minims or semibreves), especially the descending whole tone of the Tenor Cadence?

Diminution Manuals

When we look to the sources, there is an easy answer to this question. There are several historical treatises on the Art of Diminution, showing how any long note can be divided into shorter notes (hence the period English term for this practice, Division), with many examples of passaggi to be taken as models for prepared or spontaneous ornamentating. To ensure that the Diminution flows smoothly to the next note, these examples are categorised by the interval between the two long notes: up or down; unison, second, third, fourth etc. So all we have to do is select a treatise from the early 1600s (or a little earlier, representing the style that Monteverdi’s singers would have learnt from their teachers), and turn to the section on descending a second, and we can see a dozen or more historical solutions.

If a modern-day singer prefers to learn by ear (a reasonable and historical preference), then instead of studying recent CDs, it’s easy to record a selection of historical examples, or to transcribe period dimutions into Sibelius and export a sound-file. There are links at the end of this article for Divisions of the Descending Second by Virgiliano (1600) and Ortiz (1553). And this is a good moment to mention Helen Roberts’ excellent Passaggi app for improvisation and ornamentation, which offers a 21st-century learning approach using period sources. Practice your divisions here


Making divisions on a descending second is an attractive and source-based answer for ornamenting Monteverdi’s monody. Unfortunately, it’s the wrong answer! And that’s because we have been asking the wrong question. Instead of asking “how should we ornament cadences?”, we should ask “how should we apply ornamentation?”.

How should we apply ornamentation?

The difference between these two questions becomes clear if we look at Caccini’s Le Nuove Musiche (1601).  [Translation of the Preface and link to the original  print here.] For his ‘noble manner of singing’ Caccini seeks to update the 16th-century practice of diminutions by ‘avoiding the old-fashioned manner of passaggi‘, which is ‘more suitable for wind and string instruments than for the voice’. Nevertheless, his didactic examples and composed songs have plenty of diminutions, fitting into the continuing tradition of Arie Passaggiati [ornamented arias]. So in preparing this article, I thought I would be able to extract from Le Nuove Musiche a useful selection of models for ornamenting cadences, to compare and contrast with Ortiz and Virgiliano.

I was wrong. Firstly, there are few Tenor Cadences in Caccini’s songs: he uses many Imperfect Cadences (7 6, in continuo-speak), and at Perfect Cadences (4 3) he prefers to give the voice the Soprano Cadence. But even more significantly, when he does write a Tenor Cadence, he almost always leaves it plain. There might be plenty of diminutions earlier in the phrase, but at the cadence itself there is a plain long note (in a couple of instances, he adds a simple trillo).


Dolcissimo sospiro (Le Nuove Musiche, 1601) passaggi and plain cadence

Somewhat rattled by this, I turned back to Monteverdi’s most famous examples of written-out diminutions: Orfeo’s magnificent ornamented aria Possente Spirto , the famous Echo piece from the Vespers, Audi Caelum and Monteverdi’s take on the Three Tenors: Duo Seraphim. Here as in Caccini’s teaching examples and chamber-songs, it is clearly seen that the passaggi finish just before each cadence.


Possente Spirto (Orfeo, 1607) passaggi and plain cadence


Audi Coelum (1610 Vespers) passaggi and plain cadences


Duo Seraphim (1610 Vespers) passaggi and plain cadence

Standard modern practice – leaving a phrase plain and then dividing the long note at the cadence – is the opposite of what Caccini and Monteverdi notate.


Prefaces & Treatises

In addition to Diminution Manuals and composed Arie Passeggiati, there are written commentaries on ornamentation practices in several early seicento Prefaces and Treatises.

In the Preface to the earliest surviving baroque music-drama, Rappresentatione di Anima e di Corpo (1600), Cavalieri’s instruction for singers is ‘senza passaggi‘ (and for the continuo, ‘senza diminutioni‘. In sacred music for solo voices and continuo, Cento Concerti Ecclesiastici (1602), Viadana warns singers not to add to the few melismas he writes. Caccini (1601) similarly remarks that he has notated all that is necessary for his chamber-songs.

Peri’s Preface to Euridice (1600) links, translation and commentary here emphasises the speech-like quality of his dramatic monody, in which intermediate syllables are sung so lightly that their pitch is almost indiscernible. He mentions that the famous soprano Vittoria Archilei has added diminutions to some of his previous compositions (‘more to obey the practice of our times, rather than because she thinks that therein lies the beauty and force of our singing’), and hints at ‘those beauties and delicacies which cannot be written, or if written cannot be learnt from notation’. Caccini also warns that the most exquisite touches – squisitezze – are beyond notation, but that they can be learnt from written examples combined with practice.


Caccini describes and shows how to apply a new type of ’emotional ornaments’, vocal effetti (effects) that produce affetti (emotions), instead of old-fashioned passaggi. Most frequently mentioned are crescendi/diminuendi on a single note, especially such on exclamatory words as Ahi! Deh!  etc. Phrases can be started with an exclamatione (forte-subito piano -crescendo, all on one note), with piano-crescendo on the first note, or rising from a third below. In the middle of the phrase, rhythms are systematically altered to make long notes longer, short notes shorter. For cadences, Caccini gives two simple ornaments: the one-note trillo (for the Tenor Cadence) and the two-note gruppo (for the Soprano Cadence).


Both examples (contrary to some modern-day recordings and performances) accelerate from slow to fast.

Similar effetti were introduced by Cavalieri (1600) in his Preface, and are indicated in the score itself by the intial letter of each effect. G = groppolo, T = Trillo etc.


Listen to and learn by ear Cavalieri’s & Caccini’s effetti here.

Cavalieri applies these effetti infrequently, perhaps just three or four times for the largest role (Anima), and less often for each supporting role. There are just two indications of gruppo in ensemble music.

This way of ornamenting may not be as ‘new’ as Caccini suggests. Giustiani (1628) describes differing ornamentation practices for sacred polyphony (passaggi) and monody (effetti) amongst performers in Rome as early as 1570. See Timothy McGee’s “How one Learned to Ornament in Late-Sixteenth Century Italy”. 

Rhythmic Alteration


Monteverdi notates rhythmic alterations in which the voice anticipates or lags behind the continuo accompaniment.

In bar 2, the singer’s “sed” is immediately after the Tactus-beat (just a quaver rest), ahead of the continuo-bass (which comes a crotchet after the beat)

In bar 4, the singer’s “Ni…” again enters  unexpectedly early, anticipating the continuo-bass which is played on the next Tactus-beat

In bar 5, the singer’s “sed” is immediately after the Tactus-beat, and the continuo-bass has a minim, played on the Tactus-beat. Although the singer’s syncopated rhythm is similar to bar 2, the effect here is of a delay, rather than an anticipation, because the structure-defining continuo-bass plays first. 

At the beginning of bar 4, notice also the effetto ornament on the final note of the Imperfect Cadence: Monteverdi uses this effetto also in Orfeo, but only a couple of times in the whole opera.



My esteemed colleague, Xavier Diaz-Latorre, makes the excellent suggestion that one can better understand these jazzy syncopations by first trying the “square” version, with voice and continuo together in the obvious way, in order to appreciate the subtle effect of the altered rhythm. Perhaps singers might introduce this effect spontaneously, even where it is not notated, whilst continuo-players maintain the steady groove of Tactus. Monteverdi, Caccini and Jazz here.

Caccini gives many examples of rhythmic alteration. The common feature is that long notes are extended to become extra long, and short notes are correspondingly adjusted to be even shorter, within the same Tactus-duration.


Rhythmic Alteration


Speech-like or Song-like?

In the Preface to Comboattimento (published in 1638, but first performed in 1624), Monteverdi writes that Testo, the Narrator, should avoid gorghe and trilli, except in the one Aria.

This is supported by the anonymous Il Corago (c1630),  who notes [at the end of Chapter X] that ‘there are only a few cadences appropriate to each voice, and these occur frequently’. Nevertheless, dramatic monody ‘lacks those ornaments and beauties which greatly embellish singing: I mean [a lack] of passaggi, trilli, gorghiggiamenti, because these are too far removed from the normal manner of speaking and work against moving the passions… and for this reason singers are forbidden to use these ornaments and adorments when they act [or perform, declaim] in this style.’ 

In Chapter XI section III, Il Corago explains that ornamentation can be applied to diegetic songs (representing on stage the act of singing). ‘To give the musician an opportunity to use all the artistry of music, such as gorge in passaggi and sweetly drawn-out melodies, the poet can have some of them represent singing… giving an opportunity to both composer and singer to make those passaggi and beauties which are absent from the current style recitativo‘.

For the whole genre of dramatic monody, Doni [Annotationi  pages 60-62] makes a similar broad distinction between speech-like and song-like music, and categorises speech-like monody into three types: Narrative, Expressive and Special. For Doni, (as for McGee and me), the word recitativo is problematic, but what he calls Special Recitative is closer to song (although he does not approve of that!) It is found in many pastoral scenes (Doni’s examples are from the opening scenes of Peri’s Euridice) as well as in Prologues. It is tempting to assume that Special Recitative might therefore allow some passaggi, but the evidence of seicento music-drama scores suggests that although there is plenty of music that Doni would categorise as Special Recitative, hardly any passaggi are notated.

Nevertheless, the Prologue to the 1589 Florentine Intermedi has wonderful passaggi, sung by Vittoria Archilei. These passaggi stop before each cadence.



There is also an Aria Passeggiata by Caccini, in which passaggi in the middle of the phrase lead seamlessly into gruppi at cadences.  The character-role is a Sorceress who will tear down the moon from the skies.


Most probably, the Intermedi reflect earlier practice, as yet less influenced by the northward spread of Neapolitan and Roman influences.

In 1607, Orfeo’s famous aria Possente Spirto shows the seicento application of the old-fashioned aria passeggiata genre , as the protagonist sings a diegetic song at the central moment of the drama. The divisions are spectacular and charm Caronte’s ears, even though they do not move his emotions. The passaggi end before each cadence, though Monteverdi adds effetti at some cadences.


Possente Spirto (Orfeo 1609) passaggi before the cadence, effetti at the cadence



Echo scenes were a special case, where the poet cleverly answers his character’s questions with a responding Echo, whilst the composer provides short bursts of passaggi for the Echo to imitate in reply. There is nearly always a moment where the audience are led to expect an answer, but the Echo remains silent. Cavalieri’s Anima sets a musical challenge for the imitating Echo with ornamentation that aptly supports the meaning of each key word, and the poet triumphs at the end of the scene by creating a complete sentence from the Echo’s replies.


In Orfeo, Striggio’s Echo is less witty, but more expressive, and Monteverdi does not notate passaggi for the scene, perhaps because the key words are so emotionally loaded.



Audi Coelum brings the theatrical device of the Echo into the sacred domain, and enlivens the syllabic style of theatrical music with thrilling passaggi.



Genre distinctions – Phrases & Cadences – Passaggi Effetti

Putting all this information together, we can begin to understand in which genres, and at which moments within the phrase,  different types of ornamentation might be applied. We have to distinguish between song-like music and speech-like dramatic monody; between final cadences and the preceding phrase; between dividing long notes into elaborate passaggi and adding restrained effetti. 

Passaggi are associated with the main body of the phrase, with song-like music, and with ear-charming delight. Effetti are associated with cadences, with speech-like music, and with ‘moving the passions’. All kinds of musicial complexity, including passaggi, were considered inappropriate in dramatic monody, because they diminished the speech-like quality of the musical declamation, and therefore worked against emotional communication. But if an actor represents a character singing, then ornamentation becomes ‘realistic’ and appropriate.

Perhaps there was a slight tendency for song-like writing (Doni’s Special Recitative) to encourage small doses of passaggi even in theatrical music, but there is scant evidence for this. The reverse is evident: certainly there was a strong tendency for theatrical restraint (from ear-tickling divisions) and passion (in emotional effects) to be applied to chamber and sacred monody, especially where the text suggested a first-person embodiment of a character-role. In this sense, the soloist of Audi Coelum represents the character-role of a soul crying out to heaven for counsel; the tenor who sings Nigra sum sed formosa speaks the words of the Shulamite “I am a black girl, but beautiful”! 

Where passaggi were employed, they end before the cadence itself. Cadences can have trilli, gruppi etc, but only infrequently.

How to deal with cadences?

This leaves us wondering, what we should do with the cadences, if we don’t ornament them. Doni describes [Annotatione page 362] the standard practice of early seicento theatrical monody (even though as a theorist, he disapproves of it). He is shocked at the contrast between the fully sung penultimate syllable and the almost unpronounced final syllable.

The penultimate note (the accented syllable of the word, and the Principal Accent of the poetic verse) is nearly always a long note (minim or semibreve) and is really sung (in contrast to the speech-like delivery of the rest of the line, clearly described by Caccini and Peri as ‘something less than singing’). The final note (the unaccented final syllable) is short and unaccented, barely pronounced. Il Corago warns that final syllables should not be dropped entirely, but Cavalieri indicates a silence at the end of almost every phrase. Gagliano’s stage directions have the singer starting to move on the penultimate syllable of a strophe, so that they have already turned away from the audience for the final syllable. 

In sharp contrast to today’s standard practice, the period recipe is: almost speak in dramatic styles, or add passaggi to song-like phrases, but don’t ornament the cadence. At the cadence, really sing the penultimate note nice and long, and then the last note is short and unaccented.

Meanwhile, if you actually count the cadences, they are usually no more frequent in monody than in polyphony. Of course, prima prattica polyphony disguises cadences by avoiding simultaneous cadences in all the voices, whereas in monody voice and bass usually make the cadence together. But the percieved ‘problem’ of cadences may be one that modern performers have inflicted on themselves, by inappropriately sustaining last notes and by attempting to ornament in the one place they should not!

Ornaments or emotions?

Doni (from his viewpoint as a theorist) and the utterly practical Il Corago both distinguish between song-like music with passaggi that delights the ear, and speech-like music that focusses on the pronunciation and emotions of the text in order to move the listener’s passions. In Orfeo, La Musica can do both:

Io su cetera d’or cantando soglio

Mortal orrechio lusinghar talhora

E in questo guisa a l’armonia sonora

De la lira del ciel, piu l’alme involglio

Singing to the golden lyre as always,

I can beguile mortal ears for a while.

And in this way, with the sonorous harmony

of the lyre of heaven, I can even influence souls.

But significantly, Caronte is not moved by the passaggi of Possente Spirto. And Il Corago explains why [Chapter X]:  ornamentation ‘distracts from the material that is sung about, and transfers your attention to the simple aural delight in masterly singing’. Doni describes this same dichotomy in terms of the rhetorical purposes of music: to delight, or to move the emotions? Caccini similarly contrasts the old-fashioned delight in passaggi with a new focus on moving the passions by means of crescendo/diminuendo, esclamationi and other effetti.

In this context, our modern focus on ornamentation of Monteverdi’s Recitative misses the point entirely. Instead of trying to apply passaggi to the cadences of dramatic monody, we should be focussed on delivering text to and inducing emotional response amongst audience members. And that’s why I’ve utterly lost patience with that ornament: it’s the wrong answer to the wrong question in the wrong situation. 

Finally, as promised, here are links to Diminutions of the Descending Second by Ortiz (1553) transcription and soundfile and Vigiliano (1600) transcription and soundfile. But these files come with a HIP health-warning: the entire argument of this article is that it is not historical to apply these passaggi to the cadences of Monteverdi’s dramatic monody. 

Orlando Orlando: Drama and dance-rhythms

1st November 2019:
To celebrate Orlando Orlando‘s being nominated for Russia’s highest theatrical award, the Golden Mask, in 6 categories – best production Georgij Isaakyan, best design Hartmut Schörghofer, best musical direction Andrew Lawrence-King, best lighting design Alexey Nikolaev , best female soloist Maria Mashulia, best male soloist Kiril Novakhatko – this article has been updated with additional commentary on Handel’s techniques of Drama & Dance-rhythms.

This article was first posted in connection with the premiere of Handel’s Orlando at the Helikon Theatre in Moscow, 27th March 2019, entitled Orlando, Orlando: Handel’s Orlando (1733) in memory of the victims of the shooting at the Pulse nightclub in Orlando, Florida (2016). Music by George Frideric Handel, Libretto adapted from Carlo Sigismondo Capece L’Orlando (1711) after Ludovico Ariosto Orlando Furioso (1516/1532). Concept & Adaptation by Georgij Isaakyan (Director), Edition by Andrew Lawrence-King (Musical Director), Techno episodes by Gabriel Prokofiev, Design by Hartmut Schörghofer.

Synopsis of Georgy Isaakyan’s version (read online and/or download pdf)

Orlando Orlando libretto (includes English translation: read online and/or download pdf)

This production is not an ‘authentic’ reconstruction of baroque opera, but a new work of music-theatrical creativity in which 18th-century music tells a 21st-century story, bringing together Gabriel Prokofiev’s specially composed electronic music and the most modern understanding of how George Frideric’s score would have sounded at the King’s Theatre, London in 1733.

For Orlando, Handel assembled an unusually large orchestra with a powerful bass-section, and the dance-rhythm of the fashionable Gavotte is heard several times, representing Orlando’s fury.

In his madness, Orlando identifies Angelica as the mythological godess Persephone: “Beautiful eyes, no, do not weep, no”

In his madness, Orlando mistakes Dorinda for the goddess Venus, or an enemy warrior: “Already, I wrestle him; already I embrace him  with the force of my arm”

In the extraordinary mad-scene created for the famous Italian castrato Senesino, bass instruments play alone as the protagonist descends into a hell of jealous rage.

“I am my own spirit, cut off from myself. I am a ghost, and like a ghost I want to make the journey down there to the kingdom of sorrow!”

And the full orchestra lurches into 5/8 metre as Orlando imagines himself rowing Charon’s boat into the underworld.

“There is boat across the river Styx! In spite of Charon, already I’m rowing over the waves”

Handel freely borrowed from other composers’ (and his own) work, and the previous season he re-wrote two earlier dramas, expanding the chamber-opera Acis & Galatea and transforming a one-act staged masque into the first English oratorio, Esther, performed as a three-act concert with the addition of solo harp, trumpets, drums and a chorus. For Orlando, Handel adapted Carlo Sigismondo Capece’s (1711) story of mad jealousy, itself a re-working of episodes from Ariosto’s 16th-century classic, Orlando furioso. Bernard Picart’s (1710) engraving of the giant Atlas, republished in 1733 as Le Temple des Muses, was re-interpreted as the stage set for the opening scene with the magician Zoroastro.

Perhaps this 1731 print inspired the first scene of Handel’s ‘Orlando’

Handel’s audience were thrilled by several spectacular stage transformations, utilising the full resources of period stage machinery and dramatically presented as the result of Zoroastro’s magic, assisted by his demons. In our production, Schörghofer’s design employs modern stage technology to offer the audience surprise and spectacle, whilst clarifying the subtly interwoven stories as characters from medieval romances (Chanson de Roland, 11th cent) are re-drawn by Boiardo (Orlando innamorato, 1495) Capece, Handel and Isaakyan.

“Orlando Orlando” at Helikon Theatre

A German musician producing Italian opera in England, Handel writes a conventional French-style overture, but surprises the audience with up-to-date dance-music, a fast Italian giga.

This Italian giga has characteristically continuous movement in the melody line, with a driving bass.

Listen to how Gabriel Prokofiev transforms Handel’s giga, the height of fashion in 1733, into 21st-century electronic dance-music.


The rhythmic drive of the giga is disrupted with broken phrases to depict Dorinda’s misplaced faith in ‘sweet little lies’.

In spite of trills and rests, this Aria still shows the characteristics of an Italian giga: “Oh dear little words, sweet glances; even if you are lies, how I will believe you!”

The step-and-jump rhythms of a French gigue are heard in Medoro’s second Act aria;

The restrained movement of a French gigue characterises Medoro’s hesitation: “I would like to be able to love you, but…”

the slow swing of an Italian folk-dance, the  siciliano characterises Dorinda’s wistful longing;

More gentle than a giga, the tender siciliano characterises Dorinda’s nostalgia for a love that never was: “If I return to the meadow, I am made to see my Medoro in every flower”

Orlando’s lament in hell is sung to a French passacaille.

In French operas the descending bass of the minor-mode passacaille suggests tragic passions and creates opportunities for expressive dissonances and chromatic variations: “For from tears even in the kingdom of Hell, pity can be aroused in everyone”. The audience come to realise that this text is ironic: in his madness, Orlando shows no pity for Angelica, and changes his Gavotte-refrain to “Yes, eyes, weep, yes, yes!”

The composer’s bold strokes of dramaturgical re-designing and contrasting musical styles were further transformed by unwritten baroque performance practices. Continuo-players spontaneously realised the written bass-line with rich harmonies and strong rhythms; singers added their own variations to the repeated section of a da capo aria; sometimes time would stop whilst singers  or instrumentalists improvised a final cadenza. Handel did not conduct, but directed by playing the harpsichord, alongside the theorbo (bass lute). The expression of the vocal line was not indicated with markings of piano and forte, but follows from the accentuation and emotions of the words.

This Recitative is not just rapid patter, look at all the rhetorical detail: A long note and glorious high notes for “As custodian of your glory…”. Strong dissonance for “I stimulate you to follow it”. Another long note for “Urge.. ” and the highest notes and thrilling contrasts of short notes for “…your heart to great works!”. A long sigh “Ah!” with an intake of breath afterwards, dissonance and Orlando’s voice dropping “love takes it all away from me”. Zoroastro’s voice rises with long notes and an unexpected sharp in the melody-line for “It will be given back to you by valour!”. Orlando’s falling phrase (which would be given the conventional drooping appoggiatura) “It languishes in my breast”. Zoroastro’s strong retort with high notes “Scorned…”, snappily broken phrases “is that what you want to be…” and a suitably horrible melodic tritone “by a vile little boy?”. The “little boy” is Cupid as the flute’s flapping wings show in the following bars.

Instrumentalists similarly have few written phrasing-marks, but imitate the crisp articulation of the Italian language with a great variety of bow-strokes.

What might appear to be just a series of equal quavers acquires subtle rhythmic patterning from the long/short, accented/un-accented syllables of the Italian text, imitated in this English-language metrical paraphrase: “Respond to it for me; your heart might tell you that.. I discard all your love”. Today’s performers might usefully channel a jazz-singer’s approach to text and rhythm, rather than classical training.

For the eerie calm of Orlando’s final aria we added baroque harp, which in Handel’s dramatic works suggests a vision of heavenly peace. Trumpets and drums represent royal authority and military power; horns and oboes a pastoral idyll; the flute an amorous nightingale or Cupid’s fluttering wings. Modern scholarship has revealed the subtle structure of Handel’s recitatives, which imitate the pitch contours and speech rhythms of a great actor in the baroque theatre.

Studying the text as dramatic speech in the grandiose style of baroque spoken theatre reveals how accurately Handel notates [what Il Corago first described c1630 as] ‘the declamation of a fine actor’, in the generation between Thomas Betterton and David Garrick. As shown in my English-language metrical paraphrase: Zoroastro barks out his anger with the urgency of poetic anapests followed by the characteristic contrast of short and long notes “To what risks you’re exposed now, you reckless lovers, by blinded love!”. Angelica’s reply is a languid drawl “We only have to get free from Orlando.” Zoroastro barks again with the upward intonation of an abrupt question “And if he comes here?” – singers can appropriately add an upward appoggiatura. Medoro tries to assert himself, but Handel’s downward inflections betray the character’s weakness “My heart is also valiant!” and Angelica interrupts with powerful rhythm and a strong upward leap “P’haps for my sake, he would not be so cruel” – the conventional appoggiatura makes a harsh dissonance here. Zoroastro mimics her phrase with the slow tempo of bitter sarcasm “And he’ll be nice… to his unfaithful lover?”. With a wonderfully dramatic contrast, he switches back to fast anapests “Hurry up and get running, fly away from his anger…”. The notated rhythms of Handel’s music work perfectly as dramatic speech.

See my previous article on tempo and rhythm for Handel, here.

We added a chorus, whose members (in the manner of Handel’s oratorios or Bach’s Passions) comment on and drive forward the events of which, in the end, they are the victims. Their music is borrowed from Handel’s drama of cultural identity and religious conflict, Israel in Egypt (1739): Handel himself re-worked one of these choruses for Messiah (1741).

In Isaakyan’s reworking of the story, the magician Zoroastro appears in different guises, always as an authority figure: a star news-presenter, a domineering father, a bible-preacher, a populist politician. The choruses I selected show the public’s various reactions: unchallenging acceptance “Great was the company of the preachers”; anxious forboding “The people shall hear and be afraid… they shall be as still as a stone”; belated understanding “There came a thick darkness”; and a fascination with destructive power “He gave them hailstones for rain, fire mingled with the hail”.

“Orlando Orlando” Premiere Left to right: Hartmut Schörghofer, Gabriel Prokofiev, Georgy Isaakyan, Andrew Lawrence-King, Dmitry Bertman

Orlando, Orlando: Nominated  for the Golden Mask in 6 categories (2019)


The Art of Time: Tomas de Santa Maria on performing renaissance Fantasia

Tomas de Santa Maria’s Arte de tañer Fantasia (1565) free to download here is a teaching book for keyboard instruments and vihuela. Like Milán’s (1536) book for vihuela El maestro read more here, this publication is intended not only to teach the rudiments of notation and instrumental technique, but also to give detailed information for high-quality performance and to empower students to improvise their own Fantasias, in the strict polyphonic style of the late renaissance. Thus, the second part download part 2 here offers a complete introduction to 16th-century counterpoint.

In this post, I offer a brief overview of the contents of the two volumes of the Art of playing Fantasia, and analyse in detail Tomas’ comments on his highest priorities, Time and Rhythm, as well as his remarks on Ornamentation and on Performance Practice in general.



The Art of Fantasia


Tomas presents improvised Fantasia-playing as a renaissance Art, a term which had quite a different flavour almost half a millennium ago. Whilst the 20th century has taught us to regard art as the triumph of a lone genius over rules and restrictions applying only to ordinary folk, in the 16th and 17th centuries Art was defined as a system of coherent principles that transformed raw nature into artful creativity, full of life and grace. More about the period meaning of Art and period terminology here.

As Renaissance Art, Tomas’ fantasia is improvised within the rigorous structures of Franco-flemish polyphony, inherited and developed by such composers as Antonio de Cabezon (who checked and approved Tomas’ work), and his arte is indeed a book of rules: 78 chapters of detailed prescriptions, plus several bonus sections on key Performance Practice topics.

The remainder of this article consists mostly of extensive quotes from Tomas’ book, so for clarity my brief comments below are in blue.

Prologo: principles & fundamentals

El fin de este libro es arte de tañer fantasia – “the aim of this book is rules for improvising, divided into two parts. The first deals with all the pre-requisites that are necessary to begin improvising… the second part deals with everything necessary for this purpose, which is to improvise counterpoint, all put into a system (puesto en arte) and into universal rules (reglas universales)…

“In this first part we proceed by way of easier and clearer matters, beginning with the names of the notes (signos), but our principal intention is only to teach young professionals in this discipline (arte) what they need to put into practice, step by step from the most obvious and lightest matters, towards greater matters, and not [beginning] with the most demanding and difficult matters, which would tend to confuse and intrigue experts in such questions rather than enlightening and educating those new beginners, who like children should be nourished with light sustenance, easily digested, and later with more solid food!”

“In all the sciences and disciplines such order is essential… we see the same in Nature, which proceeds from imperfection to perfection… This has been the reason and motivation for our setting out to begin the first part with the notes, not as they have previously been analysed, but from first principles and fundamentals.”

Science, Art & Use

Tomas’ Prologue also includes a discussion of the need for arte – a coherent set of principles, contrasting this with uso – use, i.e. the habitual way of ‘just doing it’. Such use is not necessarily bad – a good habit can be a useful skill – but it must be guided by arte – rules. Our modern-day understanding of ‘art’ as the engagement with mysterious beauties beyond everyday rules is Renaissance Science.

Tomas’s personal connection to the ineffable, divine aspect of Music is proclaimed in a lengthy exposition of the role of Music in the Bible and in Christian doctrine. Reluctantly, he leaves this topic, to focus on the subject of his book – arte as a set of principles.

‘But I wish to leave this [Science], about which much more could be said, so as not to depart from my main purpose. And I say that although I have served the institution of my Order by playing organ wherever my duties took me, I considered many times the great effort required until now, and the many years taken up by learning singing and playing. Moved by emotions of love and charity, I began to investigate and re-examine how all this might be expressed as arte – a set of principles, so that in a short time and with less effort one could acheive the goal, and not merely as uso – habit, ‘just doing it’.

‘Because habit is broad and risky, whereas principles are narrow and sure. And so we see from experience that no-one without principles is perfect in their skilled discipline (facultad); because those who go without principles are like those don’t know the way and go without a guide; and like those who go in the dark without light. Since principles are the guide and the light, then it’s quite fair to say that those who do creative work (obran) without observing principles are ignorant.

‘This is the declaration of the Philosopher, who was asked what knowledge is; and who replied that knowledge is understanding the matter from its causes and first principles (primeros principios), which is what arte consists of.’

And so Tomas spent 16 years of ‘incalculable and incredible work‘, consulting with high-level colleagues, in particular Antonio de Cabezon, in order to perfect his set of principles, and teach his arte as ‘universal rules‘.


Part 1 begins with the names of the notes in plainchant (canto llano) and staff-notation (canto de organo); the three Hexachords (propriedades); the contradiction between the hard and soft Hexachords (see below); changing Hexachord (mutacion); the two pre-requisites for singing from staff-notation.

Then follows an extra section with ‘advice for maintaining the Tactus (compas) well, analysed in detail below.

Chapter 6 continues with note-values; introduction to the keyboard; semitones; black and white keys; intervals etc. Chapter 13 deals with Performance Practice, setting out eight conditions for fine playing (see below), which are discussed point by point in the following chapters.

From Chaper 20, Tomas explains how to perform polyphonic works on vihuela or keyboard, including advice on ornamentation. Then he analyses the Church Tones, Renaissance Modes, use of remote tonalities and Cadences.

Part 2 is devoted to the rules of counterpoint in 51 chapters: dissonance and consonance; suitable progressions, ascending and descending, whether in slow notes or faster; voice-leading; formal design. Chapter 52 has advice for new players (see below). The final chapter shows how to tune keyboard instruments and vihuela (in meantone).


Whilst every musician understands that one shouldn’t mix up B-flat and B-natural, Tomas’ comments on the contradiction of the hard and soft Hexachords – la contradicion que ay entre las dos propriedades de bequadrado y bemol (Part 1, Chapter 3) are interesting in the context of musical expression and History of Emotions.

‘Of the three Hexachords, the two that are B-natural and B-flat are notorious for being mutually repugnant and contrary – muy notorio ser repugnantes y contrarias entre si – and to such a degree that in no way can they suit or conform one to another nor vice versa, unless there is a particular necessity to make some perfect fifth or perfect fourth, or to excuse some dissonance of fa against mi… Finally, if we are singing or playing in B-natural, we must necessarily avoid singing or playing with B-flat [and vice versa].

‘The reason and cause of this contradiction and repugnance is because song with B-flat is a soft, sweet and smooth song (blando, dulce y soave), and on the contrary, song with B-natural is hard, strong and bitter (duro, rezio y aspero), and so – like soft and hard  – they are manifestly opposites and contraries.’

The natural Hexachord [C D E F G A, containing neither B-flat nor B-natural] is halfway between the hard and soft Hexachords, conforming with either of them’. Tomas links this ‘convenience and conformity’ to the structure of eight-note modes, which combine notes from two six-note Hexachords. ‘The natural Hexachord is halfway, a tempering and concord … with which every mode (tono) can complete its perfect operation.’


Part 1, Chapter 5 De dos documentos para en brevemente cantar canto de organo – two pre-requisites for quickly [learning] to sing from staff-notation

‘It is certain and evident that staff-notation – canto de organo – is highly important and necessary for the player, both to understand what they are playing as well as to set a work [i.e. arrange polyphony for solo instrument] and gain advantage from [studying] it. Just as a scholar to complete his diploma has to read many learned writers every day… so the player should… set works in staff-notation by selected composers every day, enriching his knowledge of new and fine things… Por falta de fundamentos se gastava mucho tiempo…

If you lack fundamental skills, you’ve been wasting a lot of time!

Tomas’ emphasis on staff-notation and deep understanding of counterpoint takes his book into territory beyond that explored by Milán in El maestro (1536). Milán uses tablature notation, which tells the vihuela-player which string to pluck with the right hand and which fret to stop with the left hand, note by note, and with careful control of rhythm. But this notation does not show the movement of the individual polyphonic voices, and Milán allows more freedom than Tomas de Santa Maria in adapting the strict rules of counterpoint to the exigencies of a particular instrument. Although Milán requires basic knowledge of staff-notation, his students learn to improvise polyphony mostly by ear and by ‘muscle memory’, by learning the stops and plucks that create the progressions and cadences of each mode. Tomas teaches staff-notation in detail and wants his students to learn counterpoint as an academic, as well as practical, exercise. But both writers encourage students to play good music, in order to learn by example, reproducing and imitating learnt musical fragments in their own improvised fantasias.

Tomas gives ‘two very brief and comprehensive rules (reglas), with which in a very short time one can easily learn and understand in depth: the first deals with  compas and the second with written note-values.’

As for Milán (read more here), so also for Tomas, the term compas combines the philosophical concept of Tactus (the slow, steady pulse governing renaissance and baroque rhythm) with the practical, physical representation of that pulse as a down-up movement of the hand (or foot) and with the notation of the duration of a down-up pulse unit by the note-value of a semibreve and by a bar of staff-notation enclosed by bar-lines. Tomas distinguishes clearly between Tactus (the complete down-up movement, corresponding to a semibreve) and semi-Tactus – medio compas (downbeat only, or upbeat only, corresponding to the duration of a minim).


Quanto al compas, que es como fundamento del canto de Organo, por quanto siempre estriva en el, sea mucho de notar que la llave y govierno de toda la musica, assi del cantar como del tañer, es el compas y medio compas, de los quales el que bien supiere usar, terna bien fundamento para bien cantar y tañer, por que el compas es ciera guia en toda la musica mensurable que por su certidumbre le dezimos ser el freno de la musica, porque nos detiene para no cantar ni tañer desatinada y desconcertadamente, sino conforme a razon, por peso y medida, y por preceptos y reglas de musica. Y assi con justo titolo el compas es llamado el govierno con que se concierta y rige toda la musica, assi del cantar como del tañer, dandole toda gracia y ser. 

‘Regarding Tactus, which is like the foundation of staff-notation, since it is always based on Tactus, it should be carefully noted that the key and government of all music, whether sung or played, is the Tactus and semi-Tactus. If you know well how to use them, you’ll have a good foundation for singing and playing well, for Tactus is a sure guide in all measured music [i.e. not plainchant], which for its certitude we can say is the musical brake which restrains us from singing or playing recklessly and in disorder, but instead rationally, by weight and measure, and by precepts and rules of music. And so it is the appropriate title to call Tactus the government with which all music (both sung and played) is brought together and ruled, giving it all its grace and its very existence.’

Tomas now links the practical purpose of Tactus as the basis of musical ensemble to its formal definition in relation to Aristotelean Time: ‘a number of movement in respect of before and after’ Physics (4th cent. BC). We should keep in mind that Isaac Newton’s theory of Absolute Time was not published until more than a century later.

‘Compas es medida, en la cantoria tomado a intento que las bozes concurren en consonancia a un mesmo tiempo. Tactus is Measure, used in choir so that the voices come together in consonance at the same time.’

‘Compas es la cantidad a tardanca de tiempo que ay del golpe que hiere en baxo a otro siguente baxo. Tactus is the amount of duration of time from one down-beat to the following down-beat.’

In the next paragraph, he links the physical hand-movement and practical purpose of Tactus-beating to the notation of musical time with bars and note-values.

‘El compas con que se mide toda la musica practica assi del cantar como del tañer fue sacado del compas con que se mide y inivela la cantidad a cuya semejança el compas de la musica practica mide el tiempo que se gasta en las figuras del canto de Organo… The Tactus that measures all practical music-making both vocal and instrumental is taken from the Tactus that measures and determines the quantity represented by a bar of musical notation which measures the time taken up by the note-values of staff-music.’

Tomas makes it abundantly clear that all measured music (i.e. all music except chant) is governed by Tactus, and the music has to conform to the Tactus, not vice versa.

‘El compas, en el qual estriva toda la musica practica… Tactus, on which all music-making is founded. ‘Toda la musica, assi del cantar como del tañer, esta subjectada y atajada al compas, y no la compas ala musica. All music, both vocal and instrumental, is governed by and founded on Tactus, and not the other way around.

About this, students who want to excel in this art should be well admonished.

‘The Tactus is divided into two equal parts, that is into two semi-Tactus… by the upbeat, so that the Tactus is always on the downbeat and the semi-Tactus on the upbeat’. Tomas emphasises that in binary metre every semi-Tactus is of equal duration.

‘There are two different types of compas in music-making – in one the Tactus is divided (as above) into two equal parts. In the other type into three equal parts: this is the compas of Proportion, also known as Triple metre, in which of the three parts that it has, two are spent on the down-stroke and the other one on the up-stroke. This is done singing two Semibreves on the downstroke  and one on the upstroke [slow, Sesquialtera proportion] or two Minims on the downstroke and one on the upstroke [fast, Tripla proportion].

Four requirements for maintaining Tactus perfectly

  • Beat time with the hand, down-up, with each stroke of equal duration

‘Even though the upstroke should not have a ‘bump’ [topar] as the down-stroke does, nevertheless the down-beat must hit as if it struck something.’ He mentions two faults to avoid: ‘often we see imprecise Tactus-beating without any ‘bump’ neither on the down nor the up, or hitting with the hand as if it struck something on down and up’. This subtle difference between down- and up-strokes is the concept of arsis and thesis.   ‘Every bar has these two beats.’

  • The hand stays down for the entire duration of the semi-Tactus

‘It is not lifted until the note on the upstroke. Similarly on the upstroke the hand stays up for the entire duration of the semi-Tactus, until the downstroke.

‘For this it is necessary to raise and lower the hand with equal regularity – una misma ygualdad.

  • The up- or down-beat and the note on which it falls are struck together simultaneously – juntamente a un mesmo tiempo 

‘The beat is not before or after the note, the note is not before or after the beat, but absolutely together at once. For this, it’s necessary that each beat, both down and up, should be struck with a certain force or impetus, and in addition both should be struck equally, that is one doesn’t strike the downbeat harder than the up, nor the up harder than the down.

  • Every bar goes as measured and determined by the measure of the first bar

‘The measure of tempo maintained in the first bar is maintained in every bar that follows, by reason, that one doesn’t take more time for one bar than for another.

‘We give this advice to new players,  that they basically count by semi-Tactus [minims] … and this way they cannot fail to play in Tactus with all the rigour that is required. because by experience we see that those who don’t play in Tactus err in the semi-Tactus.’

Note that Tomas is encouraging beginners to count relatively quickly, in minims [about MM 60], whereas more experienced players might count the whole Tactus [semibreve ~ MM30]. Modern-day musicians are so used to a fast count, that even Tomas’ easy option of 60 bpm is challengingly slow for many nowadays




‘If you want to maintain Tactus and semi-Tactus well, practise a lot maintaining it for yourself [i.e. within your own body] with the hand and with the foot… for players, maintain Tactus & semi-Tactus with your foot, since whilst playing you can’t do it with your hand.’


Mensuration signs & note-values

Part 1, Chapter 6 De las figuras

We are discussing the figuras – written notes – according to their note-value sung in compasete [indicated by C, modern ‘common time’], which is  now commonly used by everyone…. even though  [C-slash, modern ‘alla breve’] is also called compasete by many, which if taken strictly we would have to sing by whole compas [down-up Tactus, and also bar-length], which is breve or two semibreves.  Nevertheless we use it [in the same way] as C, a half-circle without the slash, with the result that using one or the other [mensuration sign, ‘time signature’] we now sing in compasete. What is strictly called compasete  in this period, which is C without slash… the Semibreve is one compas [down-up Tactus, and also bar-length]’

So by Tomas’ time, the strict definition of compasete (Milán calls it by another dimunuitive, compasillo) as C is informally extended by many to include C-slash; and the realisation of C-slash strictly according to theory, i.e. counting by breve and semibreve (which would suggest double tempo, though Tomas does not clarify this explicitly), seems to have been abandoned in practice.


Although period use of the term compas often includes the meaning ‘a bar of notated music’, Tomas’ basic explanation of note-lengths clearly shows that bar-lengths can be varied in practice. In his table, the numbers viii, iv, ii and 1 count the number of Tactus beats.



This defining exemplo shows:

  • Bar-lengths are expanded as necessary to accommodate large note-values


  • The primary meaning of compas is Tactus (as a duration of time corresponding to the movement of the Tactus-hand)


  • Note-values are defined in relation to Tactus

In theory, the relationship depends on mensuration sign, but the theoretical distinction between C and C-slash no longer applies in practice.


  • A single mensuration sign (i.e. C) allows varying bar-lengths

This contrasts with the modern use of C as a time signature requiring a consistent bar-length of one semibreve.


  • There is no assumption of maintaining duration as bar = bar.

This has implications for triple-metre proportions, which today’s performers sometimes describe as ‘bar=bar’. That might be an accurate description in some circumstances, but ‘bar = bar’ is not a period principle that can be used to determine proportions.

Tomas’ (1565) principles of rhythmic notation are entirely consistent with Milan’s theory and practice in El maestro, three decades earlier.

The fundamental quantity is Tactus. Relative durations are specified by note-lengths. The notated bar-length framed by bar-lines  is essentially a visual convenience, no more.

Two Principal Requirements for singing from staff notation

Dos cosas se reguieren principalmente para saber cantar canto de Organo.

  1. Give each note its written time-value
  2. Know which note is on the up- or down-beat, and which is not

For minims it’s easy… with each beat down or up, you sing a minim… if one minim comes with the downbeat, the next is with the upbeat and vice versa.

Tomas now gives examples for various note-values of how notation is linked to Tactus beats.

8 conditions for playing with total perfection and beauty

Book 1, Chapter 13 ‘So that all music might have that grace and essence (ser – literally, ‘being’) which it deserves, it’s necessary to play with all the delicacy that is required, which is repaid in much gold and creates yet more essence and grace. Without this, all that is played, however good it might be, will not have grace nor brilliance. Here is the clear difference between the same work played by a perfect and refined player, or played by another, imperfect and coarse; because played by the expert it will appear to be delicate and high art, and played inexpertly it seems low-class and coarse, as if it were two different pieces.

‘The conditions which thus beautify the music can be reduced to eight:

  1. Play in Tactus (compas)
  2. Place your hands well.
  3. Strike the keys well.
  4. Play cleanly and distinctly.
  5. Let the hands run well up and down the keyboard.
  6. Use appropriate fingering
  7. Play with good groove (ayre)
  8. Make good ornaments and trills (redobles y quiebros)’

‘Playing in Tactus … is the first condition’

For more on Tactus, Tomas refers his readers back to his previous remarks (analysed above).

Chapters 14-18 are specific to keyboard, in particular clavichord, technique. Chapter 14: Fingers are numbered from thumb 1 to little finger 5. Hands are curved like cat’s paws, fingers close together, thumb underneath and close to the 5th finger. All this is close to period harp-technique too. Elbows are dropped, relaxed and close to the body.

Chapter 15: strike the keys with the flesh of the finger; with impetus; equally strongly with both hands; don’t strike from too high above the key; press down into the key, but not so much as to raise the pitch; don’t raise the fingers too much away from the keys.

Chapter 16: for clarity, release one key before playing the next. Lift the finger a little after playing, but don’t take them too far away from the keys.

Chapter 17: for facility throughout the whole range, keep the hand compact, turn the hand slightly in the direction of movement (especially for fast notes), keep the active fingers close to the keys.

Chapter 18 defines Principal Fingers as those that strike the first note of trills. Thumbs are not used for black notes, except for octaves in one hand, or when there is no possible alternative. One should not use the same finger twice in succession for crotchets or (especially) quavers. Consecutive semibreves, on the contrary, are played with one finger repeating. Melodic crotchets are taken pair-wise, alternating two fingers. This is the familiar Renaissance concept of Good and Bad notes, corresponding to the accented and unaccented syllables of a song-text: more on Good & Bad here. Quavers and semiquavers are fingered four-by-four. 

Tomas analyses fingering in considerable detail, confirming the importance of fingering in creating short-term phrasing and articulation. His fingerings for two-note chords require changing fingering on consecutive thirds, which has implications for facilitating particular ornaments (page 45). 

Groove and Swing


Chapter 19 introduces the concept of ayre – particular ways to apply rhythmic freedom to fast notes, within the regular pulse of the Tactus. Ayre sometimes refers to melodic tunefulness, but more often to subtle rhythmic patterning. Depending on context, I translate it as ‘groove’ (dance patterns and/or medium- term patterning) or ‘swing’ (changeable, short term patterns), in the jazz sense of subtle rhythmic adjustments that give a particular character or elegant shape without disturbing the fundamental beat.  

‘The way to play with good ayre… requires playing the Crotchets in one way [groove] and the Quavers in three [alternative options for three different ways to swing].’ Thus these adjustments are within the fundamental steady pulse of Tactus (semibreve, down-up) and semi-Tactus (minim, down or up).

‘The manner – manera – you must have for playing Crotchets is to wait – detenerse –  on the first and hurry – correr –  the second; and neither more nor less wait on the third and hurry the fourth; and in this way for all the Crotchets. As if the first Crotchet were dotted, and the second a Quaver… and take note that the Crotchet that hurries should not be very hurried, but a little moderate – un poco moderada.



‘Of the three manners of [playing] Quavers, two are done almost the same way, which is waiting on one quaver and hurrying the other one… In one manner you begin by waiting on the first Quaver, hurrying the second; and neither more nor less waiting on the third and hurrying the fourth; and in this way all of them… As if the first were dotted and the second a Semiquaver. This manner is suitable for works that are contrapuntal throughout – todas de contrapunto – and for passages of decorative fast notes both long and short – passos largos y cortos de glosas.



The second manner is done by hurrying the first Quaver and waiting on the second; and neither more nor less hurrying the third and waiting on the fourth; and in this way all of them… As if the first were a Semiquaver and the second a dotted Quaver. In this manner, the dotted Quavers are never on the beat, but in-between. This manner is suitable for short decorations – glosas cortas – which are done like this in [composed, contrapuntal] works as well as in [improvised] fantasia.  And note that this manner is very much more galana (elegant, showy) than the other one, above.’



The noun gala and its related adjective galana occupy an area of meaning that extends from ‘decorative’ or ‘elegant’ to ‘luxury’ or ‘ostentation’. Milán discusses tañer de gala, which seems to be well towards the ‘showy’ end of this semantic spectrum, as suggested by my translation ‘bravura playing’. More on Milán here.

‘The third manner is done by hurrying three Quavers and waiting on the fourth; and be warned that the waiting has to be all the time that is necessary so that the fifth Quaver comes to be struck in time on the semi-Tactus; and in this way all of them. With the result that they go four by four… as if the three Quavers were Semiquavers and the fourth a dotted Quaver. This third manner is the most galana of all, and is suitable for long and short decorations – glosas largas y cortas.

‘Take note that the waiting on the Quavers should not be much, but just enough to show and be understood a little, because waiting a lot causes great gracelessness and ugliness – desgracia y fealdad – in the music. And similarly for the same reason, the three Quavers that hurry should not hurry too much, but with moderation, conforming to the waiting on the fourth Quaver.


The soundscape of Renaissance rhythm

Tomas’ instructions for Renaissance ayre create a rhythmic soundscape that differs sharply from 20th-century assumptions about art-music and improvisatory fantasias. He demands that the player count in minims, which should be completely steady. From other evidence, it is plausible that this count would be somewhere around minim = 60. Within that slow steady beat, crotchets are good/bad (i.e. subtly long/short), quavers are subtly shaped in one of three specific ways, the choice depending on the genre of music and the length of the decorative passagework. Whichever groove or swing is applied, it is maintained consistently throughout the passage in question.

There is no trace of 20th-century rubato, nor of its early-music derivative, phrasing that ‘goes towards’ a certain point. There is none of the hesitancy and pauses that often characterise modern-day performances of ‘improvisatory’ music: on the contrary, even if the player is genuinely improvising, Tomas and his advisers, the Cabezon brothers (as well as Luys Milán before them) expect Tactus, Groove and Swing to be maintained.

Nowadays, one might describe Tomas’ sound-world as steady pulse at approximately 60 bpm, with regular groove at the subordinate level around 120 bpm, and various options for swing at the most rapid level of rhyhmic activity, around 240 bpm. But in that pre-Newtonian age, Tomas has no concept of Absolute Time on which to base such a description; he has no clock precise enough to measure such short durations: rather, he has Tactus, which counts Aristotelean Time as ‘a number of movement in respect of before and after’ (Aristotle, Physics). The essential quality of that Tactus movement is that it is consistent – within the limits of human perception – so that Tomas’ minim is always about one second in duration (though he has no machine to measure it, and no conceptual framework for comparing it to anything more objective than his own feeling of consistency).

It is this essential consistency that allows Tomas to map specific performance practice instructions onto particular note-values (minims are steady, crotchets groove good/bad, quavers swing in one of three ways). Such linkage, which is seen also in Ortiz’s instructions for viola da gamba,  strongly implies that the absolute duration in time of any given note-value is approximately fixed within the whole repertoire:  e.g. minims are approximately one second. If the durations of note-values could vary arbitrarily (as they can in modern practice), this linkage would be meaningless.

Nevertheless, Milan indicates subtle changes in tempo from one piece to another, centred on a default tempo of ‘well measured Tactus’ that is ‘neither very fast nor very slow’. But these changes are not imposed arbitrarily by performers’ artistic choice: performers are required to follow the composer’s directions. So – in contrast to the 20th-century concept of ‘artistic freedom’ for performers, the period attitude is that there is a correct tempo, and that it is the performers’ job to find it.

Ornamentation 1: Graces

Chapter 20: How to make redobles and quiebros 

Summarising Tomas’ definitions & examples: Redoble is a reiterated upper-note trill, starting on the written note, and turned at the beginning.


Quiebro is an upper- or lower-note trill, starting on the written note, but without initial turn. Quiebros can be senzillos (simple, i.e. one flip) or reyterados (reiterated).


The difference between redoble and quiebro is that the redoble has the initial turn through its lower note.



‘Redobles are only made on complete bars, i.e. on Semibreves. And Quiebros are made on MInims and on Crotchets and, as a marvel, on Quavers. Reiterated quiebros are made on Minims, simple quiebros on Crotchets; except for one which is not reiterated, and always made on Minims on the [hexachord] pitches sol fa mi fa. This is called the Quiebro de Minima.

‘Reiterated quiebros are made on every Minim where the fingering permits. But simple ones are not made on every Crotchet, but alternately yes and no.

‘There is only one way to make Redoble … with whole-tone and semitone combined. Quiebros are made with tone or semitone, except for the Quiebro de Minima, which is always made … with the lower semitone and upper tone.  The other way would produce gracelessness and displeasure – desgracia y desabrimiento – to the ears, for which reason it must not be made where it would finish on mi…. But it can finish on any other note ut, re, fa, sol, la.

Redobles can have the semitone above or below, ‘but note that in no way is it permitted to make a Redoble with two whole-tones combined, because this is very graceless and displeasing to the ear.’

Redobles permitted and prohibited


Tomas gives specific fingerings for each ornament, for the keyboard, left and right hands.

‘These styles of redobles and quiebros … are very new and very elegant – galanos – causing such grace and tunefulness – melodia – in the music, which bears them in so many degrees – grados – and with such contentment to the ears, that it seems something quite different from playing without them, so much so that there is every reason to use them always, and not others which are old-fashioned and not graceful.

‘Simple quiebros … for ascending are made with tone or semitone below. Those for descending are made with tone or semitone above.’

Tomas describes a very fast ornament, in which the principal note is not actually repeated, but sustained whilst the auxiliary is played almost simultaneously and quickly released. There is no conventional notation for this technique (described also in some harp sources), so he does not provide an example. With this type of fast mordent, Tomas prefers the descending version (with upper auxiliary) to the little-used ascending version. 

Quiebros on Crotchets, both ascending and descending, are sometimes made on the beat, and sometimes on the off-beat, and this [on the off-beat] is the better and more elegant – galana – manner, because it gives more grace to what is played.

Tomas gives keyboard fingerings for each option, for each hand.

The upper-auxiliary Minim quiebro normally used for descending can be used ascending if the principal note is a mi.

‘Sometimes, and only descending, one can make quiebros on two consecutive Crotchets. which is done for grace and elegance. This occurs when after an ascent to a Semibreve there are two Crotchets descending. 

‘When there are ascending Crotchets which then descend, one must always make a quiebro on the highest note, which is done for descending [upper auxiliary]’ ‘Similarly, when there are descending Crotchets which then ascend, one must always make a quiebro on the lowest note, which is done for ascending [lower auxiliary].

‘Similarly, to give more grace to the music, one must always make quiebros on every Crotchet that follows immediately after a dotted Minim.’

‘So that the music should have more grace and thus give more contentment to the ears, it’s necessary that redobles and quiebros de minimas should be done by either hand, a redoble with one hand, another redoble with the other; and similarly a quiebro with one hand, another quiebro with the other; responding to each other.’ The fingerings for consecutive thirds (above) facilitate a similar effect between two voices in one hand.  ‘This is heard when both hands play Semibreves or Minims which can have redobles or quiebros, playing them one after another, which greatly adorns the music and gives it grace, especially when there is a chain of Semibreves or Minims.’

‘When the Mode – tono – avoids certain notes … the ornaments should also avoid them.’

Tomas gives technical advice for executing ornaments at the keyboard, repeating his earlier comments about keeping the fingers close to the keys. That advice might well be adapted for vihuela and harp, as keeping the fingers close to the strings.

Tomas characterises his ornamentation as ‘new’, and it is intended for the relatively short sustain of the clavichord. Fewer, or different ornaments might be appropriate to Milán’s period and/or to other instruments. Nevertheless, it seems likely that all renaissance music was ornamented considerably more than the raw notation suggests.

Tomas demands almost ceaseless ornamentation, more-or-less on every second note, as well as strict adherence to rules of ‘grammar’ for ornamentation. Such florid playing in regular Tactus, and with the groove and swing of ayre, creates a sound-world for the late 16th century that contrasts notably with 20th-century assumptions about art-music, the ‘purity’ of polyphony and what ‘improvisatory’ playing might mean.


Setting polyphonic works

Chapter 20. ‘Playing polyphonic works on the clavichord is the font and origin from which are born and proceed all the fruits and benefits, and all the art of playing for players.’

‘It should be noted that in whatever work of any kind, all the voices are interdependent and linked one to another, that no individual voice can move a single note without having specific respect and regard for all the other voices. And similarly voices are measured and counted, linking voices Tactus by Tactus, semi-Tactus by semi-Tactus.’

‘Two things have to be kept with all rigour, which rule and govern the arranger so that they never err: these are count and measure, which are interdependent… Measure is the same as compas (Tactus, also bar-length), by which all practical Music is ruled and governed.’

Tomas also explains vihuela tablature, and how to set polyphonic works for vihuela.

Tips for understanding polyphonic works

Chapter 21. ‘Brief advice for new players to master quickly any kind of work’

‘Three things are necessary to understanding any kind of work quickly, and thus to play it more perfectly.

1. Play in Tactus

‘maintaining it always with the same equality of time, i.e. not changing it from more to less nor from less to more. For this, it’s necessary to maintain Tactus with the foot and similarly to take great care with the semi-Tactus… in addition, it’s necessary to understand note-values and give each of them their full duration.’

2. Sing through each individual voice in turn

3. Understand all the Consonances and Dissonances in the work, whether in 2, 3 or 4 voices.

How to obtain benefit from studying polyphonic works

Chapter 22. ‘Five things have to be noted:

1. Understand profoundly the invention and artfulness in the contrapuntal progressions – passos – whether the response or repeat is at the fourth, fifth, octave or other interval… in two, three, four or more voices; with or without imitation. The Art of Fantasia consists of all this, which above all one has to get to know; because in everything it is only arte [i.e. a coherent system of rules]  that makes a Master. And from that it follows that all those who ignore the rules are imperfect.

2. Note the entrance of each voice, to know if it enters before the cadence, in the cadence, or after the cadence; with what invention or subject it enters; for the entry of each voice is the most delicate matter, of greatest subtlety and arte that there is in music. So this must receive great attention and care, in order to apply it in the works.’

3. Note all the styles of Cadences which are used in the works, undersanding them profoundly and memorising them, in order to make similar cadences when improvising [fantasia].

4. Note all the Consonances and Dissonances… and memorise them, in order to create varied progressions, for this is of great benefit in acheiving flow and abundance of spontaneity [fantasia].

5. When a progression is repeated, note the differences in the repeated version, whether in 2, 3 or 4 voices.

‘For new students to apply these benefits in improvisation – fantasia – it’s necessary that they practise constantly with the same progressions that they have learnt, so that with this practice – uso – they become accustomed to the [rules of] arte, and then they can easily play other progressions. Similarly, it is very advantageous to transpose a particular progression into all possible keys. For this, take note that wherever you want to transpose them, they must keep the same [Hexachord] solmization.

‘To gain the great fruits and benefit for improvising of all the above, it’s necessary to practise many times each day, with great perseverance, never mindlessly – desconsiando – but trusting for certain that work and constant practice – uso – conquers all and creates a maestro… A drop of water can carve out stone, not in one or two droplets, but falling constantly.’

Tomas recommends frequent, mindful practice, repeating the same material many times in order to perfect, memorise and internalise it. Although his comments are consistent with the modern-day understanding of learning elite skills, he expresses himself through the period meaning of such terms as uso (practical techniques) and arte (a coherent set of rules for effective creativity). What we mean nowadays by ‘art’, the ineffable mystery of the emotional power of music, is Renaissance Science. More about period terminology here and here.

Tomas’ emphasis on learning progressions and cadences echoes the approach of Milán’s El Maestro.


Ornamentation 2: Divisions

How to add divisions to polyphonic works. Chapter 23 Del glosar las obras

Renaissance ornamentation is categorised as Graces on a single note (the redobles and quiebros discussed above) and Divisions or Diminuitions – glosas –  where the interval between two long notes is ‘divided’ into many shorter notes. Tomas gives several examples for each interval, ascending and descending.

‘To add divisions – glosar – to polyphonic works one must be advised that glosas are only made on three note-values – Semibreves, Minims and Crotchets – even so, least often on Crotchets.

‘To glosar a work well, there are two things to note:

  1. All the voices should have equal amounts of glosas
  2. If the voices repeat something, the glosas are also repeated in all the voices

unless something prevents this, which is often the case.

If it is necessary to glosar Crotchets ascending or descending [four by four, stepwise], one must take the glosas for Semibreves ascending or descending a fifth. 

Helpful Hints & Improvisation

Book 2, Chapter 52.  Useful advice for new players.

Tomas’ purpose is not only to teach the instrument and basic musicianship, he also gives advice on how to learn to improvise within the demanding style of 4-voice Renaissance polyphony.

1. Practise running the hands throughout the whole range of the instrument, with appropriate fingerings, observing all the conditions and circumstances already discussed.

2. Practise making redobles and quiebros with both hands.

3. Maintain Tactus very well with hand or foot… give each note-value its precise value

4. After studying a piece well in a class, write it out just as the master taught it, with the glosas etc. Similarly, sing through each individual voice.

5. Understand well how to play the instrument

6. Take as your foundation and guide the Eight Conditions for playing perfectly [above].

7. Understand and be able to play in all possible keys

8. Practise easy works first, and then progressively more difficult ones

9. Practise transposing works into every possible key.

Similarly, try to take from each work those progressions which have graceful melodies, and memorise them, so that afterwards you can improvise on them spontaneously.

Once you are expert in all these things, try to start to play improvisations, based on some melodious progressions. In addition, try to play the progressions with different imitative counterpoints – fugas – i.e. at the fourth, fifth and octave, which greatly beautifies the music.

Similarly, try taking one voice from a work (whichever you want, soprano, alto, tenor or bass), and play it as the treble in four-voice harmony, making up three voices in your head… with a variety of harmonies, which greatly exalts and beautifies the music.

Similarly, once you are already a little expert in playing a given voice like this as the treble, trying playing it as the alto, tenor or bass [with three new voices created around it]. This suggests the Renaissance technique of composing a Parody Mass, in which the counterpoint of a pre-extant motet is re-worked and greatly extended to create an entire mass setting. This technique could be a model for improvisation in Tomas’ style, as could the fantasias compiled for keyboard, harp and vihuela by Henestrosa, ‘cutting and pasting’ contrapuntal progressions from various works into a new creation.

If you want to be a perfect player, try to apply yourself and practise little by little, playing counterpoint that has a good feeling – ayre – and graceful melody; on plainchant; and especially polyphonic music, until you are perfect at it. For this is the root and source from which grow and proceed all the skills applicable to the keyboard, and also the perfection and grace which it gives to all the music that is played.’

Tomas final words perfectly capture the essence of Renaissance improvisation: Practise by playing good music well; strive for total perfection; this will improve your skills, your improvisation and your playing of written music.

Modern-day performance

Three significant differences stand out from Tomas’ rules of arte, when compared to today’s HIP approach to fantasia and renaissance polyphony.

  1. Many modern-day performers choose not to play in rhythm.
  2. Few modern-day performers add much ornamentation.
  3. Few modern-day improvisers maintain correct counterpoint to Tomas’ standards.

Tomas’ comments on Tactus are so strongly worded that it is beyond any doubt that all the music of this period should be played in Tactus, counting regularly on a minim beat, and controlling this with the physical movement of hand or foot. Unless you frequently practise and problem-solve with a physical Tactus-beat, you are out of touch (sic) with Renaissance rhythm.

This insistence on Tactus is certainly fundamental, but it is not merely elementary. On the contrary, Tomas associates it witha perfect and refined … expert’, and with playing that is ‘delicate and high art’. 

The amount and detail of ornamentation specified by Tomas is alarming for those of us accustomed to ‘the pure lines of renaissance polyphony’. Our ears, as well as our fingers, will need plenty of practice with so many redoblesquiebros and glosas in all the voices, on almost every second note.

We might assume that improvisation excuses sloppy rhythm and bad counterpoint, even that ‘art has no rules’: Tomas’ book exists to teach the contrary!


El Maestro on Tactus: Luys Milán’s renaissance instructions for well-measured Tempo



Luys Milán’s (1536) book of music for the vihuela de mano entitled El maestro is the first collection of music for this renaissance instrument, tuned like a lute but shaped like a guitar and played similarly to the lute by plucking rather than strumming. It is also the first teaching book for any instrument, intended for complete beginners (a basic knowledge of singing from staff-notation is assumed) and structured progressively from the rudiments of tablature notation to elementary solo pieces, more demanding fantasias and dance-music, and song accompaniment in various styles.


Milán’s purpose is not only to teach how to play the instrument, but to show his students how to create their own fantasias in the formal contrapuntal style of the high renaissance. Right from the first example, his tientos (musical ‘essays’) are of the highest artistic quality, beautifully expressive as well as gramatically correct.


Of particular interest are his detailed comments for each piece, holding firmly to his declared priorities: tempo and tones. His renaissance ‘tones’ are the eight modal scales which define the ‘key note’, progressions and final cadence of each piece.

Freely available online, you can download the original print (full colour, high resolution, Spanish language and tablature) and also a transcription into staff notation, transposed for modern guitar (with some errors and omissions, without Milán’s tempo indications, with note-values halved or otherwise reduced) My recording, with Jordi Savall, is here, but it does not reflect my more recent research into Tactus.




Compas, mesura & ayre

In this post I analyse Milán’s remarks on Tempo, his highest priority. His Spanish terms are mesura (the ‘measuring’ of music in time); ayre (musical ‘feel’, rhythmic patterning, we might well translate this as ‘groove’); and (most frequently) compas. The significance of compas is broad, combining the philosophical concept of Tactus (the slow, steady pulse governing renaissance and baroque rhythm) with the practical, physical representation of that pulse as a down-up movement of the hand (or foot) and with the notation of the duration of a down-up pulse unit by the note-value of a semibreve and by a bar of tablature enclosed by bar-lines.



The pre-requisite for studying with El maestro is that the student should understand as a singer, how one must keep Tactus and Measure: basta que sepa cantando entender como se ha de traer el compas y mesura.

Tactus is the slow, steady underlying pulse, Measure is the sub-division of that slow pulse into all the various rhythmic combinations of differing note-values. Milán’s ygual compas is remarkably parallel to lutenist John Dowland’s insistence on the ‘equality of measure’ in his (1609) translation of Ornithoparcus’  (1515) Micrologus. 


It is also consistent with baroque lute/theorbo-player Thomas Mace’s (1676) Rule of Time-keeping, requiring ‘Exact, Equal, Constant, True and Even Motion… like the Balance of a good Clock… in all musical performances whatever’ read more, and with such theorists as Zacconi (1592), who characterised Tactus as r’egular, solid, stable, firm… clear, sure, fearless, and without any pertubation.’



Contrary to the present-day fashion for artistic freedom, Milán’s language (in common with most period sources) is strongly prescriptive. His oft-repeated formula is se ha de… – one has to: se ha de tañer con el compas – you have to play with the Tactus. This theoretical understanding and practical skill in maintaining steady Tactus should similarly be the pre-requisite for all modern-day HIP performances of renaissance and baroque music.

Milán begins with this essential requirement:  es menester que sepays que mesura y ayre se ha de dar a la musica – you need to know what measure and groove have to be given to the music. Pues sabemos que cosa es compas, vengamos a saber quantas de las sobredichas cifras entran en un compas: pues por esto se ha de saber el ayre y mesura – once we know what Tactus is, we can learn how many note-values fit into one Tactus: then from that you have to know the groove and measure.

In the detailed comments that follow, Milán’s starting assumption and ‘default setting’ is steady Tactus, maintained throughout the piece: ‘you have to play it all with an equal Tactus, without making any change’ – la [musica] aveys de tañer toda a un igual compas sin hazer mutacion.  That Tactus should be well-measured, neither very fast (rushed) nor very slow (spacious) – con un compas bien mesurado… ni muy apriessa  ni muy a espacio.  In practice, the Tactus is shown by a down-up movement of the hand or (more conveniently while playing) an up-down movement of foot, in steady time – El compas en la musica no es otra cosa … sino un alçar y abaxar la mano o pie por un ygual tiempo.

As notated in tablature, the combined up-down movement corresponds to the standard bar-length of a semibreve, so a single beat (either down or up) corresponds to a minim. This is consistent with many other sources, and in Spanish practice was referred to as compasillo. Milán shows one piece with bars of double length, which is a notational convenience designed to be easier on the eye, but seems to have no implications for the beat or the tempo.


Default Tactus: ‘ygual tiempo’ & ‘compas bien mesurado’


The ‘equality of measure’ characteristic of Tactus is certainly the default setting for any particular piece. This ‘equality’ probably continues between one piece and another across the whole repertoire, since Milán is careful to specify when any piece goes even slightly faster or slower. And it is these painstakingly described exceptions to the default tempo that make this vihuela book a crucial source for anyone working with renaissance music.

From the very outset, Milán is clear that the performer should not select their own tempo at will, but should respect the composer’s intentions. se ha de considerar en las siguentes fantasias la una: que se ha de tañer con el compas apressurado o espacioso como el auctor quiere – the first thing one has to consider in the following fantasias is that one must play with the Tactus fast or slow as the composer wants. [His second priority is to identify the tone].

The default setting for any given piece is an equal Tactus for the whole piece. But Milán asks for that Tactus to be selected (according to his wishes, not the performer’s whim!) within a range centred on ‘well-measured, neither very fast nor very slow.’


Milán’s range of tempo indications


compas a espacio – slow Tactus

compas bien mesurado– well-measured Tactus

compas algo apressurado – slightly fast Tactus

compas apressurado – fast Tactus

quanto mas se tañera con el compas apressurado mejor parecera – the faster the Tactus, the better

Note that it is the Tactus itself that goes slow, well-measured, slightly fast, fast or ‘the faster, the better’. The note-values within the Tactus can be anything from breve to quaver. The physical action of beating time with the hand makes even a small change of Tactus feel quite different – a different groove or ayre. And we can assume that even the biggest change of Tactus is distinctly less than a doubling/halving of tempo, since this could be better shown by halving/doubling note-values.

I suggest that Milán’s compas bien mesurado might be approximately minim ~ MM60, and that his other tempi would be subtle adjustments to that default setting. In the language of jazz, Milán’s apressurado could be ‘up-tempo’ and his a espacio ‘laid back’.

Milán states that compas batido (literally, ‘beaten’) means the same as apressurado (literally, ‘pushed’), i.e. ‘fast’.

Changes of Tactus within a piece

The reader should experience a frisson of shock at the above sub-heading: in the renaissance context of equal, steady Tactus ‘without any perturbation’, the idea of changing Tactus at all is surprising, and changing it within a piece is almost alarming. Milán recognises that to play like this ‘has litttle respect for Tactus or for most music’: it is appropriate only in a certain bravura style, tañer de gala, with long passages of fast notes, redobles, contrasted against slow harmonies, consonancias. 

This is musica con diversos redobles … y tiene mas respecto a tañer de gala, que de mucha musica ni compas – music with various fast notes… and it has more respect for bravura playing than for formal musical construction or Tactus.

Its particular style is tañer de gala con estos redobles largos – bravura playing with these long passages of fast notes.

The noun gala and its related adjective galana occupy an area of meaning that extends from ‘decorative’ or ‘elegant’ to ‘luxury’ or ‘ostentation’. Milán’s tañer de gala with its disregard for the normal rules of Tactus and musical structure would seem to be well towards the ‘showy’ end of this semantic spectrum, as suggested by my translation ‘bravura’.

Writing his own Arte de tañer Fantasia for keyboards and vihuela in 1565, Tomas de Santa Maria similarly emphasises steady Tactus in all music, and offers suggestions in Book 1, Chapter 19 for the ‘groove’ – buen ayre – of crotchets and quavers within the regular minim (semi-Tactus) beat. Crotchets always go long-short (i.e. good-bad), with the long crotchet on the Tactus beat. Quavers can go long-short, or short-long. This second style is only suitable for short passages of fast notes – glosas – but is much more galana (elegant) than the first. The third way is in groups of four quavers, short-short-short-long: this is suitable for long or short glosas and is the most galana (showy) of all. Perhaps Tomas’ third style would be appropriate for Milán’s tañer de gala. Whichever style is chosen, it should be maintained consistently through that particular glosa. More on Tomas de Santa Maria in another post.

Milán repeats many times and with small variations in wording his instructions for changing Tactus, but only in this context of tañer de gala – music which is like ‘trying out’ the vihuela, mixing harmonies with fast notes:  una musica la cual es como un tentar la vihuela a consonancias mescladas con redobles…

para tañerla con su natural ayre haveys os de regir desta maniera. Todo lo que sera consonancias tañerlas con el compas a espacio y todo lo que sera redobles tañerlos con el compas apriessa. ‘To play it with its natural groove, you have to rule yourself in this way: everything which will be harmonies you have to play with the Tactus slow and everything which will be fast notes you must play with the Tactus fast.’

Milán is quite specific that he asks for something beyond normal rhythmic accuracy – it’s not enough for him that the small note-values are faster notes anyway. He insists (many, many times) on changing the Tactus itself, so that the written contrast in note-values is exaggerated by the change in Tactus.

Modern performers might be tempted to interpret ‘fast’ as ‘twice as fast’, but this robs the listener of the sensation of a change of pulse, since the doubling of speed will be heard as halved note-values within the same underlying pulse. The shock of changing Tactus is greater if the change is noticeable, but small, and not proportional.

Similarly, modern performers might want to add accelerando or rallentando, but Milán does not suggest this, and again, the effect of an abrubt change is greater. I tell my students to

use the gear-shift, not the accelerator or brakes!

Milán is very clear that change of Tactus is only allowed in particular circumstances: Esta fantasia que sigue es de la misma arte de la passada fantasia tentando la vihuela con redobles y consonancias; y que vos he dicho de que manera y compas se han de tañer estas fantasias que mas propriamente se pueden dezir tentos – This fantasia that follows is in the same style as the preceding fantasia, trying out the vihuela with fast passages and harmonies: and I have told you in what style and Tactus you have to play these fantasias, which more properly might be called tientos [essays].

Y por esta mutacion de compas os dire que no la aveys de tañer como tañereys esta musica que de aqui adelante torna a proseguir la qual es como la del principio que la aveys de tañer toda a un igual compas sin hazer mutacion. And about this change of Tactus, I tell you that you must not play like that in the music that returns in the following pieces, which is like the pieces at the beginning, which you have to play all in an equal Tactus without making any change.


Technique & Phrasing

Milán distinguishes between two different techniques for playing fast notes. Both techniques produce an alternating pattern, but with different sound-quality: alternating two fingers dos dedos is considered more elegant than back-and-forth with a single finger dedillo, but particular melodic patterns suit one or the other technique.

y parar de tañer en cada coronado un poco – and stop playing a bit at each fermata mark.  This could be interpreted in the modern sense of breaking the time and waiting longer at the pause mark, but is more likely to imply creating a silence (literally, stop playing for a bit) within the notated value. This is consistent with the Rhetorical (i.e. word-based) principle of making the last note of the phrase ‘bad’, i.e. short/un-accented, and with period use of this sign (historically not a pause but signum congruentiae – the sign of harmonic resolution, the end of a phrase). More on good/bad notes here.

In the slow tempo that Milán requires, the vihuela’s final chord would need to be sustained for more than four seconds. There is little hope that a vihuela-player could prolong this even further, as a modern fermata: it is almost inevitable that the sound will stop before the notated duration – the usual situation for final notes in this period.



Some songs – romances – have similar passages of instrumental redobles between the vocal phrases, for which Milán suggests a variety  of performance options. The passages in fast notes can be omitted entirely, or the song can be performed with two different Tactus speeds, in the bravura tañer de gala style described for those particular fantasias. han se de tañer lo que fuere consonancias a espacio, y los redobles que ay a las finales quando acaba la boz muy apriessa – it has to be played with the harmonies slow, and the passages at the end of phrases when the voice stops very fast.

For the first romances in the book, Milán requires the singer to sing llano – plain, sustained – whilst the vihuela shouldn’t go very fast, nor very slow: la vihuela ni ha de yr muy apriessa ni muy a espacio. This might rule out the tañer de gala approach, or it might be an instruction applying to the vocal episodes only.

For each vocal genre, there are specific instructions song-by-song for the differing roles of singer/instrumentalist. Sometimes the singer may improvise ornamentation whilst the vihuela plays slow chords, and sometimes the voice is ‘plain’ and the instrument has discanto – counterpoint.

The alternative version of the first villancio has the instruction: el cantor ha de cantar llano y la vihuela algo apriessa – the singer has to sing plain and the vihuela slightly fast. Note that this is not an instruction about the compas (Tactus), it clarifies the different levels of activity in the voice-line and instrumental-part respectively. Singer and vihuela must keep the same Tactus (of course!), but within that Tactus there can be various levels of activity.

Similarly, the second villancio has the instruction: el cantor puede hazer garganta y la vihuela ha de yr muy espacio – the singer can make throat-ornamentation and the vihuela has to go very slowly. This song also has an alternative version with the singer llano and the vihuela apriessa.

For the first soneto (sonnet), Milán writes that the singer can add some trills- algun quiebro – whilst the vihuela goes at moderate pace. Other sonnets have similar instructions: el cantor glose donde huviere lugar con la boz y donde no cantar llano – the singer ornaments where he has an opportunity with the voice, and otherwise sings plain. El cantor ha de cantor llano. Y donde cabera glosar con la boz sea quiebro o trinar que dizen – the singer has to sing plain. And where it fits, to ornament with the voice, which should be quiebros or trills as they are called. The last sonnet has to be played algun tanto regozijado – with rather much rejoicing.

It’s worth noting that Milán distinguishes carefully between different genres – Spanish or Portuguese villancico, Italian soneto and the romance.  Each genre has its own performance practice. Milán leaves no doubt that romances are a special case, and the fast instrumental solo passage-work within them has to be treated differently from the rest of the song: lo que de musica se sigue despues de las finales es para solo tañer y ha de callar la boz donde acaba la cifra colorada – the music which continues after the end of phrases is instrumental only and the voice has to be silent where the coloured notation ends.


The Italian Pavan is also a particular genre with its own performance style. Milán presents 4 of his own compostions, and 2 Italian melodies set by him for the vihuela. The first pavan-like fantasia se ha de tañer ni muy a espacio ni muy apriessurado: sino con un compas bien mesurado has to be played neither very slow nor very fast, but with a well-measured Tactus. el ayre della remeda al ayre de las pavanas que tañen en ytalia – Its groove resembles the groove of the Pavans that are played in Italy.

The Pavans that follow are to be played with a slightly fast Tactus, algo apressurado. Milán asks for this dance-music to be played two or three times through. The last Pavan is in triple metre (often found in Spain, not in Italy) with the [slow, sesquialtera] proportion of three semibreves.



Milán shows two triple-metre proportions, notated with three minims to the bar and three semibreves to the bar. We can observe the notated bar-lengths of Milán’s two proportions, but there is no explanation of how to measure those triple-metre bars in time, i.e. with the hand-movement of the Tactus.

Most of the pieces (in common time), and the examples of proportions in the introduction, have no mensuration sign (time signature) at all. For changes between duple and triple metre, Milán uses these signs without explanation:


The modern reader might well wish that the Maestro would have distinguished more clearly between the inter-linked concepts of compas as bar-length, hand-movement and time-duration.  But whatever significance might be read into the mensuration signs, the musical content seems to argue against the hypothesis of  “bar = bar” as a universal rule for linking compas as notated bar-length to compas as time-duration/Tactus. A large variety of rhythms, including short note-values, is found in Milán’s proportion of 3 semibreves to the bar; whereas the episodes in 3 minims to the bar are almost entirely in the jig (canario) rhythm of dotted minim, quarter, minim. This suggests a fast proportion for 3 minims (tripla, 3 minims to one tactus beat, 6 minims to the complete down-up cycle: minim ~ MM180), and a slow proportion for 3 semibreves (sesquialtera, 3 semibreves to two tactus beats = the complete down-up cycle: semibreve ~ MM 90).

This would be consistent with what appears to be Monteverdi’s practice in the early 17th-century: read more.




For certain pieces, Milán asks for the 4th fret to be adjusted. Alçareys un poco el quarto traste de la vihuela para que el punto del dicho traste sea fuerte y no flaco – raise the 4th fret a bit so that the stop at this fret might be strong and not weak. This appears ambiguous: which way is ‘up’? Moving a fret towards the bridge raises the pitch of the note and might seem to be Milan’s meaning. But when he repeats the instruction later for a song accompaniment, he clarifies: haveys de alçar el quarto traste un poco hasta las clavijas de la vihuela – you must raise the 4th fret a bit towards the tuning-pegs. This has the effect of lowering the pitch of the stopped note, which is what we would expect for the sharps and hard-hexachord notes on this fret in the meantone temperament typical of the period.


From a History of Emotions perspective, it is interesting that the low (i.e. in-tune) major third is characterised as fuerte, and the out-of-tune high third (characteristic of modern Equal Temperament) is ‘flaccid’.


Zarambeques: Encounters between Spanish Baroque & African Traditions

Andrew Lawrence-King & Ballaké Sissoko perform Zarambeques in the Festival de Fontfroide, 16 July 2019…



The Kora

The African Kora defies categorisation in Western organology. Like a double-harp, it has two ranks of strings stretched along a long hard-wood neck, but secured on a bridge, like a lute; the bridge is supported on a leather-covered calabash which functions as a resonator. Played by jali (griots), the poets, story-tellers and musicians of the West African Mandinka people, the Kora enjoyed similar high status and associations with royalty as the medieval harp and renaissance lute in Europe.

Traditionally, the nyenmyemo, a leaf-shaped metal plate clamped to the bridge enhanced the sound of antelope-hide strings, tuned with leather tuning-rings, but modern-day players use nylon strings, machine-head tuning pegs and electronic amplification. Kora music is characterised by repeating poly-rhythmic chord-sequences – kumbengo – and brilliantly ornamented melodic flourishes – birimintingo.



According to Mandinka history, the Kora’s origins are linked to the jali Mady Fouling Cissoko in the 16th century. Its music has been transmitted by oral tradition, and the first recorded album, Mali: cordes anciennes, was released in 1970, featuring Sidiki Diabaté and Djelimadi Sissoko. Album here. Their sons, Toumani and Ballaké, recorded Nouvelles Cordes Anciennes in 1999 as homage. The new generation album here.


The new generation, Ballaké & Toumani




Spanish Baroque Harp



The Spanish arpa doblada has two ranks of strings, that cross over between neck and sound-box in the way that your fingers interlace when you clasp your hands. One row has the diatonic ‘white’ notes, the other has the chromatic sharps and flats. Harps played solo, to accompany singers, in church-music and in the first Spanish and South American operas. The Spanish harp enjoyed similar high status to the renaissance lute and baroque harpsichord in Italy and France.



The gut strings are attached to a tall sound-box, which is narrow and shallow in the treble, wide and deep in the bass. This shape, and the playing technique with the right hand high up near the neck, the left hand about half-way down the strings, produce a characteristic contrast of sound: bright and punchy in the treble, rich and resonant in the bass. Spanish harp-music shares a common repertoire with the baroque guitar, consisting of written and improvised variations – flowing diferencias – over the repeating ostinato-bass of dance-rhythm grounds – formal danzas and popular bailes.

According to Juan Bermudo’s (1555) Declaración de instrumentos musicales, some harpists had already added a complete set of chromatic strings to the one-rank renaissance harp. Surviving 17th-century instruments and such publications as Ribayaz’s Luz y norte (1677) show how widespread the double-harp became. Nevertheless, single-rank harps were even more numerous, and the various national instruments of Central and South America are the descendants of particular harps brought from Spain in colonial times.

The first Spanish harp CD album, Luz y norte, was recorded in 1994, featuring The Harp Consort directed by Andrew Lawrence-King. Album here. The same ensemble recorded the first South American opera, Torrejón’s La púrpura de la rosa (1701) – opera here – and accompanied baroque guitarist Paul O’Dette’s recording of Santiago de Murcia.


Pat O’Brien, ALK & other founder-members of The Harp Consort, at the recording sessions of Luz y Norte in 1994.





The Spanish harp-book Luz y norte and guitar intabulations by Murcia contain fascinating hints of baroque African music, identified by such titles as zarambeques, cumbées, paracumbé etc. These names [read more here], containing the characteristically African mb phoneme, are associated with dancers’ swaying hips, with harmonies in ‘blue note’ sevenths, and with rhythmic patterns that avoid the accent on the downbeat. Similar rhythms and harmonies characterise the chacona, which may have been brought along the slave routes from Africa to South America before re-crossing the Atlantic back to Spain. Certainly, Afro-Caribbean influences produced the Cuban guaracha, preserved in Juan Garcia de Zéspedes 17th-century setting as a Christmas music-drama for Puebla Cathedral, Mexico. Afro-Cuban/Mexican guaracha here. Murcia’s sub-title zarambeques o muecas hints at open-mouthed facial expressions of dramatic emotions or ritual dance.



This project brings together the splendours of African and Spanish harps, the noble history of baroque Spain and the rich traditions of ancient Africa, two distinct styles of improvised variations over ground basses, two world-renowned performers, and four centuries of artistic exchanges, to produce a unique cross-cultural, time-travelling musical encounter…

… or perhaps a distant echo of that long-lost meeting which first brought the zarambeques to Europe.


Encounters between Spanish Baroque & African Traditions


Andrew Lawrence-King Spanish Baroque Harp
Ballaké Sissoko Kora


Fantasia de Luduvico Alonso Mudarra (Sevilla, 1546)
Zarambeques o Muecas Santiago de Murcia (Madrid, 18th cent.)
Guarachas Juan García de Zéspedes (Puebla, 17th cent.)
Cumbées Murcia

Futur Ballaké Sissoko

Fantasía de consonancias y redobles Luys Milán (Valencia, 1536)
Xacaras & Gallardas Lucas Ruiz de Ribayaz (Madrid, 1677)
Folias portuguesas Anon (Cancioneiro de Paris, 16th cent.)
Paradetas Improvised after Ribayaz Luz y norte

Niandou Ballaké Sissoko
Passa Quatro Vicent Segal
Samba Tomora Ballaké Sissoko

… and perhaps, there might be a visit to the Isla de la Chacona…


The Best Practical Musick: Thomas Mace’s Rule of Time-Keeping


The Best Practical Music


In a recent online discussion in the Historical Performance Research group, I gave

a timely warning to anyone who might be tempted by the idea that Rhetorical Eloquence is somehow contrary to rhythmical or harmonic structure.

My provocation drew the hoped-for riposte, with a suggestion that 17th-century lutenist Thomas Mace thought that ‘playing in time is [only] for beginners’. This suggestion, and the mis-reading of period texts as if they supported it, is so commonly encountered, that I took up the challenge, and re-read the whole of Mace’s 1676 treatise Musick’s Monument to find out what Thomas actually wrote.





The book includes many music examples, even complete pieces and suites, in tablature. Its 272 pages are divided into three parts, on the Necessity of Singing, the Noble Lute and the Generous Viol. Concerning time-keeping, Mace’s instructions for beginners and comments for advanced players are found in the The Civil Part: or the Lute made Easie, starting at page 78:

In all musical performances whatever, if they be done according to Art, they are done according to the Rule of Time-Keeping

This alone should be enough to settle any debate. And Mace gives us plenty of further detail of how to keep time.



The inter-dependence of Time and Motion is rooted in Aristotle’s definition of Time as ‘a number of motion, in respect of before and after’. Not until a century or so later would Newton’s concept (Principia, 1687) of Absolute Time gain general acceptance. Mace’s Aristotelean time requires steady motion to drive it, and – according to the philosophy of the Music of the Spheres – the motion of music imitates the perfect movement of the stars and the harmonious nature of the human body.

On the lute, ‘an instrument on which both are hands are employed, we must therefore keep time with a foot’. Muffat gives the same advice for violinists in his preface to Florilegium Secundum (1698).



Mace’s requirement for

Exact, Equal, Constant, True and Even Motion… like the Balance of a good Clock

carries forward from 1592 the principles of Zacconi’s (hand-beating) Tactus:

regular, solid, stable, firm… clear, sure, fearless, and without any pertubation



Mace’s time-keeping foot moves just like Zacconi’s tactus-hand: down for one minim, up for the next minim, so that the complete tactus-motion occupies a semibreve.



The instruction (page 79) that there should not be ‘the least Difference’ during the piece is supported by Muffat’s repeated insistence that the vrai mouvement (true movement) of French dances should continue from the beginning to the end. And as a good teacher, Mace recommends that beginners ‘carefully practise; so that the good habi acquired ‘at the first’ will ‘ever continue’ for the rest of the player’s career.

In contrast, many of us who nowadays play Early Music, received our first training in the post-romantic school of the 20th-century, with its tacit assumption that the vacillating rhythm and wayward tempi of Rubato are the secret of advanced expression. We have to read Mace’s words carefully, if we are to escape our own assumptions and inhabit his world of Aristotelean time.

In chapter XI, Mace recommends a pendulum as an aid to learning how to keep time ‘by the most Exact, Easie and Infallible Way’, and as a test for ‘masters’ even for an ‘Artist of the Highest Form… a very Master’ that should ‘be able to keep Exact True Time’. The length of the pendulum should be adjusted so that one can count from one to four ‘with Deliberation, as a Man would speak Gravely or Soberly, and not Hastily or Huddlingly; yet not Drawlingly or Dreamingly; but in an Ordinary Familiar way of Speaking’. These four crotchet-beats, i.e. one semibreve, occupy the time for the pendulum to swing one way and then the other way, i.e. a complete oscillation.  The pendulum, as an ‘assured time-keeper’ should be the musician’s Director.

Mace’s advice for students concludes with a reminder (page 81) that

The Exact Motion of True Time-Keeping is one of the most Necessary and Main Things in Musick


In this familiar 17th-century context of true time-keeping, which is supported so strongly by period French and Italian sources, Mace’s next remark might well seem contradictory:

Although in our First Undertakings we ought to strive for the most Exact Habit of Time-Keeping that we can possibly attain unto (and for several good Reasons) yet, when we come to be Masters, so that we can command all manner of Time, at our own Pleasures, we then take Liberty (and very often for Humour and good Adornment-sake, in certain places) to break time; sometimes faster and sometimes slower, as we perceive the nature of the thing requires, which often adds much Grace and Lustre to the performance.

How are we to reconcile such Liberty with Mace’s uncompromising remarks on the ‘Exact, Equal, Constant, True and Even Motion’ of Time-Keeping’ … ‘in all musical performances whatever;?  Mace, a cleric in divine orders, greatly admired the lute-playing of John Dowland, who similarly preached

Above all things, keep the Equality of Measure. For to sing without law and measure is an offence to God himself.



And according to Shakespeare’s Richard II, sweet music becomes sour ‘when time is broke’.



It would be foolhardy to turn a blind eye to all this period context of ‘exact Time-Keeping’, and leap to the conclusion that Mace’s Liberty is an invitation to apply 20th-century rubato indiscriminately to 17th-century music. Rather we must search for evidence of precisely where the ‘certain places’ are, and of how to ‘perceive the nature of the thing’.



Above all, we need to understand Mace’s concept of Humour – not as a modern performer’s personal ‘interpretation’, but as a quality that already resides within the composition, and which the performer must perceive, so that the listener may understand, enjoy and be moved. In a philosophy of performance that goes back to the trobadors and trouveres, Mace requires players to find the Humour, not invent it.

17th-century ‘Humour‘ does not mean comedy: we might roughly define it as ‘Emotion’, ‘Mood’, or (to use another period English term) ‘Passion’. As a term for musical performance, Humour is rooted in period Science, where the doctrine of the Four Humours offers a self-consistent and practicable system for understanding and working with the psychological and physiological effects of emotions. Words and music that are heard and understood in the mind (see Enargeia, Visions in Performance) send signals (Energia) down to the body, creating changes in various body-fluids. The changing balance of those fluids creates the physiological effects of emotional change.

The Sanguine Humour is linked to blood, associated with healthy red cheeks, or a delicate blush, with love, courage, hope, with the enjoyment of music, good food and red wine. The Choleric Humour is linked to yellow bile, associated with desire, anger and the urge for strong drink. The Melancholy Humour (beloved of John Dowland and Shakespeare’s Jacques) is linked to black bile, paleness in the face (lack of Sanguinity, the opposite humour), associated with sleeplessness, too much study, unlucky love-affairs, and academic precision! The Phlegmatic Humour is linked to green phlegm, and lack of any emotional response: a wet blanket.


Of course, these four Humours are not a complete catalogue of the vast array of human emotion; rather, like the four cardinal points, North, South, East & West, they indicate primary directions within the whole area being mapped.

Like those cardinal points, the Four Humours focus attention on opposites: North & South, Sanguine & Melancholic. This supports the tendency in 17th-century music to contrast one Humour (in Italian, affetto) with its opposite (oposto) – see Motion and E-motion in the First Opera. In common with opera- and ballo-composer Cavalieri’s advice for singers, Mace’s instructions for lutenists ask for contrasts of loud and soft.

And Mace’s linking of three concepts: specific musical situations (‘certain places’), affetto (Humour), and subtleties of Time recalls Frescobaldi’s (1615) Rules for toccatas (applicable also to madrigals and other genres with contrasting movements). All too often, modern performers take Frescobaldi’s remarks as an encouragement to rhythmic anarchy and Rubato; but close reading of his carefully formulated Rules reveals both the underlying assumption of steady Tactus, and the precisely delimited circumstances under which the Tactus can change. See Frescobaldi Rules – OK?

To summarise, Frescobaldi advocates using Tactus to control the music, even if that Tactus sometimes changes. He limits the opportunities for change to the break between contrasting movements with contrasting emotions. He also allows a momentary pause, on the upbeat. Caccini suggests, and Monteverdi notates another practice of rhythmic freedom, where the accompaniment continues in steady measure, but the solo melody drifts around, like a jazz singer syncopating against the rhythm section. See sprezzatura.

And 17th-century musical Time has its own special shape, described by the concept of Arsis & Thesis and illustrated by the non-linear movement of a pendulum swing. Tactus is not ‘metronomic’, in the perjorative sense. See The Shape of Time

Time & Humour

So now let’s study Mace’s remarks about Time and Humour – all of them – and see what we can discover about the ‘certain places’ where the music might go ‘faster or slower’, and how it might thus go.

Page 97 – Brisk

In describing the character – we might well say, Humour – of the key of B major ( not Bb!), rarely encountered in 17th-century lute-music, Mace uses for this Key the words ‘Noble, Brave and Brisk-Lively’. This usage reminds us that, for Mace, the words Brisk and Lively convey character, not merely speed. It would be nonsense to write that B major is a ‘fast key’, but the quality of Brisk-Liveliness is shown by the dotted notes that Mace uses in the music example that follows.

Mace’s focus in these chapters is on correct left-hand fingering and right-hand strokes, preparing “by setting your Left Hand upon the Stops, and your Right Hand upon the String, ready to strike. yet


Strike them in their due time… according to their true Quantities

Page 101-102 – gentle

In his discussion of Full Plays (chords using many strings, for example at cadences – Full Stops), Mace describes the ‘Fashionable way of Playing them (now used)’ which ‘is much more easie’, in which the thumb plays the bass note, and the forefinger rakes down all the other strings, rather than playing each string with a different finger. He defines the word ‘rake’ as ‘smoothly stroke… very gently’. There is no suggestion of slowness, indeed Mace emphasises that an intervening short note ‘will not admit of any delay’.

From this, and my previous citation, it is apparent that Mace perceives two opposing types of Humour, in which Brisk Lively is contrasted with Smoothly Gentle, but without linking these qualities to any change between Fast and Slow.  As readers of modern English, we should be careful not to add any present-day connotations of brisk = fast, gentle = slowly, when we see these words in Mace’s next pages.

Page 103-104

Mace gives a ‘General and Certain Rule (never to be altered)’ for ornamentation – Graces or the Adorning of your Play (note the use of the word ‘adorn’, which he links also to the concept of Liberty), that ‘All Shakes’ must be made according to

The Aire and Humour of your Tuning and Lesson

He then sets down the Aire as a scale, determined by the nature of the tuning of the lute, not by the tonality of the piece at hand.

He rules out the idea that rhythm might be bent for the sake of prolonging ornaments – whatever his concept of Liberty might be, it is not this.

When I have thus continued Beating, so long as my Time will allow me

Page 109 – vibrato

I can’t resist this brief digression to note that Mace’s Sting, ‘a very Neat and Pritty Grace’… ‘for some sorts of Humours very Excellent’ is vibrato, “as to make the Sound seem to Swell with pritty unexpected Humour, and gives much Contentment, upon Cases”. Thomas believes that vibrato adds a pleasing emotional quality, but only in certain circumstances.

But the Sting is ‘not Modish in these days’.  It would seem that Dowland used more vibrato than Mace….

Page 109 – Loud & Soft

Mace gives great importance to dynamic contrasts:

Play some part of the Lesson loud and some part soft

‘which gives much more Grace and Lustre to Play, than any other Grace, whatsoever: therefore I commend it, as a Principal and Chief-Ornamental Grace (in its Proper Place)

Page 109 – the Pause

At the end of this chapter on ornamentation, we have the first mention of modification of Time.

‘The thing to be done is but only to make a kind of Cessation, or standing still, sometimes Longer and sometimes Shorter, according to the Nature, or requiring of the Humour of the Musick’

If this is done ‘in its due Place’, it is a ‘a very excellent Grace’.

In subsequent pages, Mace gives many examples of ‘due Places’ for the Pause. The effect recalls Frescobaldi’s description of the Tactus hand ‘hesitating in the air’  at certain moments. For both writers, the effect is used very sparingly, and as a one-off event. It is not applied repeatedly or continuously throughout a passage, in the manner of 20th-century Rubato. There is no suggestion of rallentando approaching the Pause, indeed Mace’s words ‘but only to make a kind of Cessation’ seem to rule out any anticipatory slowing-down.

Page 115-116 – touch, humour, key, conceit

In this discussion of improvisation, Mace celebrates the ability to manage contrasts in four inter-related qualities: touch (the sounding of one or more notes), Humour (emotional quality), key (what we would today call tonality), conceit (a musical idea, subject or theme). Once the particular key is established, ‘some little Humour’ (a few more notes, a fragment of melody) allows the listener to ‘discern some Shape or Form of Matter’.

The ‘Shape or Form’ is also called a Fugue, i.e. a contrapuntal point, a fragment of melody suitable for polyphonic imitation.

‘This term Fuge is a Term used among Composers, by which they understand a certain intended Order, Shape or Form of Notes, signifying such a Matter, or such an Extention; and is used in Musick as a theme, or as a subject-matter in Oratory, on which the Orator intends to discourse.

‘And this is the Nature and Use of a Fuge in Musick.’

‘Maintain a Fuge or Humour’

In this context of improvised playing, ‘maintaining’ seems to combine the compositorial skill of working a point of counterpoint (Fugue) and the performer’s skill of maintaining an emotional mood (Humour).

Page 117 – Fuge & Humour

‘As to the Humour of It, you may observe that it all tastes of, or similises with the first bar in some small kind; yet not too much of the same Humour…the last part is a little akin to the Fuge; yet perculiarly a Humour by itself. For you may carry on and maintain several Humours and Conceits in the same Lesson, provided they have some affinity or agreement one to the other.’

Mace criticises composers of the previous generation for too many contrasts of Humour in one piece. But in the following page (118) he declares that music is a language that can express any emotion, and that it is even more powerful than rhetorical words.

In Musick any Humour, Conceit or Passsion may be expressed, and so significantly as any Rhetorical Words or Expressions are able to do

‘If any difference be, it is in that Music speaks so transcendently, and communicates its notions so intelligibly to the internal, intellectual and incomprehensible Faculties of the Soul, so far beyond all language of words…. Those Influences that come along with it, may aptly be compared to Emanations, Communications or Distillations of some sweet and heavenly Genius or Spirit, mystically and unapprehensibly  (yet efectually) dispossessing the Soul and Mind of all irregular disturbing and unquiet Motions; and stills and fills it with quietness, joy and peace. absolute tranquillity and inexpressible satisfaction.’


On page 120, he observes the ‘Fugues, Orders and Forms’ of his first three examples, in which the Humour of the first two bars is maintained in the next two bars, then for the remainder of the piece there is ‘another Humour or Fuge’, distinct from the opening, ‘but alluding to it’.  Mace’s ideal contrast in Humour is subtle and simple, rather than dramatic or manifold.

Page 120 – Suite


A Sett or a Suit of Lessons… may be of any number as you please, yet commonly are about half a dozen. The first always … in the nature of … [an improvised] Prelude…. Then Allmaine, Ayre, Coranto, Saraband, Toy or what you please, provided that they be all in the same key; yet (in my opionion)… they ought to be something akin, or to have some kind of resemblance in their Conceits, Natures or Humours.


In the example prelude that follows, ‘the whole Lesson alludes to the same thing, and yet with pleasant variety.’  We might therefore assume that in such a piece , with no significant change of Humour, there will be no need for the Liberty of making any significant change in tempo.


Page 121-124 – The Author’s Mistress

Mace tells a touching personal story about the inspiration for the composition of this piece, 40 years previously, in passionate longing for his wife-to-be. He considers it ‘the Best Lesson in the book’ and its powerful emotional associations make it an important test-case for the realisation of Humours in performance.


Mace declares that the first two bars give the Fugue, which is maintained through the whole Lesson. The Form and Shape consists of two uniform and equal strains, but the Humour of it ‘which you may perceive by the marks and directions is not common’.  The only marks and directions in the tablature are contrasts of loud/soft, ornaments and slurs.

These three terms ought to be considered in all performances of this Nature (Ayres and the like): Fugue, Form & Humour

The Fugue is Lively, Airy, Neat, Curious and Sweet – like my Mistress.
The Form is Uniform, Comely, Substantial, Grave and Lovely – like my Mistress.
The Humour is singularly Spruce, Amiable, Pleasant, Obliging and Innocent – like my Mistress.

Mace’s verbal directions are ‘to Play Soft and Loud, as you see it marked’; ‘use the Sting (vibrato) where you see it set, and the Spinger after it’; ‘observe the slides and slurs, and you cannot fail to know My Mistress’s Humour, provided you keep True Time, which you must be extremely careful to do in all lessons:

For Time is One Half of Musick.

Thus in his best composition, a work of powerful emotions and deep personal meaning, Mace looks for expression of passions by dynamic contrasts, by subtle use of vibrato and by elegant slurs, whilst insisting on ‘True Time’.  There is no place here for Rubato. Even contrasting tempi for the two sections are not suggested, probably because the nature of the piece is uniform, without contrasts between the sections.


Page 125-126 – The Offspring


This piece was composed to create a consort, combining with My Mistress as a lute-duo. It can also be played as a solo, continuing on from a solo performance of My Mistress. When it is played as a solo:


You must for the Humour’s sake make Pauses


Mace marks where the three pauses should be made, in the last strain of the piece: on each of the pause-notes, vibrato should be added. As previously, he emphasises the need for ‘soft and loud, as you see it marked’, and to ‘take notice of the Fugues, which are … maintained to the end, yet various from each other’. The Fugues determine the Humour, the Humour requires dynamic contrasts, and (for the first time) here Mace applies his concept of Liberty, for the sake of Humour.

As we would expect, the Pauses come at the end of (short, internal) phrases, on a consonant, sustained harmony, and on the up-stroke of the lutenist’s time-keeping foot. This corresponds closely to Frescobaldi’s identification of consonant, sustained harmonies as the mark of the end of a section, and with the hesitations of his Tactus hand being also on the up-stroke.

Page 126-129 – Uniformity & Contrast

Mace is teaching the student to improvise, as well as to perform composed music. So he emphasises that renaissance compositorial skill, of working out a contrapuntal point (managing a fugue) and creating ‘a True and Handsome Form or Shape’. Uniformity of Form consists of matching the number of bars between strains, and having an even number of bars in each.

The Fugue or Humour may be whatever one wants, yet they should be neat and spruce, and they should be maintained uniformly and evenly.

Uniformity is especially desirable in short dance movements: Allmaines, Ayres, Corantoes, Sarabands should always be Uniform and Even. But longer pieces – Preludes, Fancies, Pavans etc – often have ‘Humours of Pauses and Flourishes in a wild way, according to their Nature’.

Some pieces have ‘Fansical, Humorous or Conceited Names’ yet have regular ‘Forms, Shapes, and Order of their Time, or Proportion’ and may be called Allmaines or Ayres.

Mace now describes the various movement-types in a suite, He criticises some improvised Preludes as ‘confused-wild-shapeless-kind of intricate-play … in which no perfect Form, Shape or Uniformity can be perceived…. and has an unlimited and unbounded Liberty … of Forms, Shapes and all the rest.’

Pavans are ‘very Grave and Sober; full of Art and Profundity’.

Allmaines are ‘very Airy and Lively’;

Galliards ‘are performed in a Slow and Large Triple Time …. grave and sober’.

Corantoes are ‘shorter … and quicker triple-time, full of sprightfulness and vigour, lively, brisk, cheerful’.

‘Sarabands are of the shortest triple-time, but are more toyish and light than Corantoes’

A Tattle de Moy ‘is much like a Sarabande, only it has more of Conceit in it’ as if ‘speaking the word Tattle de Moy, and of Humour.

‘Chiconas are only a few conceited humorous notes at the end of a suite, very short … commonly of a Grave kind of Humour’

‘Toys or Jiggs are light-squibbish things, only fit for Fantastical and Easy-Light-Headed people’

Common Tunes are popular street songs: Mace praises them as ‘very excellent and well-contrived, neat and spruce’.

‘The Ground is a set number of slow notes, very Grave and Stately … expressed once or twice very plainly … then several Divisions upon it.’

We must understand the word ‘conceited’ in its period meaning, as ‘full of clever and witty ideas’.

Page 130 – Another Liberty

Mace calls the fourth lesson a Coranto, ‘and properly…. by the Time and Shape of it.  However [Mace] would have it played played in a Slow and Long proportion, for the Nature of it is far more Sober than a Coranto.’

‘The Fugue is seen in the first 3 notes, and perceptible’ throughout. ‘The Form is Even, Uniform and Perfect. The Humour is a kind of sorrowing, pitying and bemoaning.’

We can see something of Mace’s underlying assumptions from these instructions. He considers that there is a standard tempo for a Coranto, but that for the sake of the Humour one should adopt a different tempo, in this case slower. His ‘slow and long proportion’ might be a specific tempo, Sesquialtera proportion (rather than the usual Tripla) based on a standard tempo of common time.

Here we see one ‘certain place’ where Liberty is appropriate: for the sake of the Humour, a particular piece may be played slower (or faster) than the standard speed. Nevertheless, within that piece, the (unusual) tempo would be maintained. This application of Liberty still satisfies the absolute requirement for accurate time-keeping. Mace mentions the possibility of varying the length of the tempo-pendulum, and Frescobaldi allows certain movements to be faster or slower, but still ‘facilitated by Tactus’.

In short, performers may take an unusual tempo, if the Humour suggests it, but that tempo should be maintained accurately.

Page 130 – Finding the Humour

One short paragraph gives valuable advice for finding the ‘General Humour of any Lesson’,

by observing ‘its Form or Shape’. If it is ‘Uniform and Retortive’ with ‘Short Sentences’, then ‘you will find it very easy to humour a lesson by playing some sentences loud, and others again soft, according as they best please your own Fancy; some very Briskly and Couragiously and some again Gently, Lovingly Tenderly and Smoothly’.

Here the performer has free choice of where to apply Loud and Soft. But there is no indication of contrasting tempi. From Mace’s usage in previous chapters, we know that ‘Brisk’ and ‘Gently’ do not imply changes of tempo, but are character words, linked here and elsewhere in the treatise to Courage or to Love, Tenderness and Smoothness. Notice that according to the doctrine of the Four Humours, these are all aspects of the one, Sanguine Humour. So the contrast of Loud and Soft is not so great, as to require change of tempo.


Page 130 – The Pause


The ‘choicest lustre… in such Humours’ is given by making ‘your Pauses at Proper Places, which are commonly at the end of such sentences, where there is a Long note.’

This advice correlates well with Mace’s own practice, as observed in earlier chapters.


Page 131 – A Humour

This is another coranto-like piece, which Mace calls ‘A Humour’.


The Fugue or Subject-Matter … is throughout maintained. with handsome and various intermixtures. The Form is Uniform (each Strain within itself), though not all of the same number of bars’.


Here, the strains vary in humour.

‘Sometimes (for Humour-Sake) more Pleasant and Delightful… Humorous and Conceited… and seems to mock, or mowe, or jest; to be blyth or merry, as if it were telling some jiggish story, and pointing at this or that body … In the four last bars … you must pause and use the stinging Grace [vibrato] a pretty while; and then softly whirl away and conclude.’

This delightfully whimsical description conveys a vivid impression of the character of the piece, without resort to any suggestion of tempo change, until the Pause just before the end. Notice that the Pause is all the more prolonged in this witty and active piece; and that after the pause the ending ‘whirls away’ softly, but not slowly.

In this piece, the Liberty is not that the time is altered, but that the Humour is so witty that the performance departs from the normal mood of a Coranto.

‘And although it be Coranto-Time, yet (in regard of the Conceitedness of the Humour) I give it that name.

The title over the tablature reads ‘The fifth lesson of the first set, being a Coranto, but called I like my Humour well

Page 132 – A perfect Coranto

‘This … perfect Coranto … has its Fuge ,,, throughout maintained. Its Form is Uniform … the Humour is Solid, Grave and very Persuasive… Expostulating the Matter with great Ferventness, which you must humour by performing Soft and Loud-Play in Proper Places, where you may easily perceive such Humour to lie’.


Page 133 – Tattle de Moy


Mace helps students to find out for themselves Fugue, Form and Humour. But notice that students should find these elements, and not invent them for themselves.

The Fugue is in the first two bars. The Form is absolutely Perfect and Uniform … Its Humour is Toyish, Jocond, Harmless and Pleasant, and as if it were one playing with or tossing a ball up and down; yet it seems to have a very Solemn Countenance, and like unto one of a Sober and Innocent condition or disposition; not Antic, Apish or Wild etc’.

‘Remember [as always] to play Loud and Soft …  Briskly and Gently, Smoothly, as your fancy will (no doubt) prompt you’

Memento, that Soft and Loud play is a Chief Grace.

Mace encourages students to persist, even if his advice at first seems strange – this is welcome support for today’s Early Music performers too!

These ways of discourse will seem strange to very many at the first, because they are unusual.


Page 142 & 147 – Observations

The Humour must be found out, by playing Soft and Loud, and making your Pauses

‘When you meet with such Seeming-Single-Moving-Walking things; and find Affinity between parts and parts, or bars and bars… then Soft and Loud play is the most necessary for to Humour it…

In modern English: if you find a movement that seems rhythmically consistent, with affinity between one part, or one bar and another, then the way to Humour it is by dynamic contrasts.

‘Many drudge and take great pains to play their lessons very … fast [but] you will perceive little Life or Spirit in them…. they do not labour to find out the Humour, Life or Spirit’


Page 149 – a Grave Galliard

For the preceding Coranto, Mace writes ‘Loud and Soft, which is enough’.

The next piece has the form of a Galliard, but should be played ‘in a very Sober and Grave Proportion; for it has a most singular Humour in the way of Expostulating Grief and Sorrow’. Here again, the Humour suggests taking the Liberty to play in an unusual tempo, but there is no suggestion of rhythmic irregularity.

The Galliard on page 171 is marked ‘Play this Lesson in very slow time’


Page 152 – Slow with Pauses

‘Play it slow, make your pauses, and observe the Humour’

Otherwise, pauses seem to be used mostly in fast pieces.


Page 153 – Tattle de Moy


‘Find the Humour yourself, by Soft and Loud play’


Page 170 – Crackle the crotchets

This special effect on three-note chords consists of arpeggiating each chord, causing them to ‘sob’ by slacking the left hand grip as soon as the note is struck, suddenly deadening the sound. Mace is careful to specify that this is all done in such a way

‘so as not to lose time, but give each crotchet its due quantity’



It is beyond debate that the underlying context for all Mace’s advice is of regular Time-Keeping. That time-keeping is by Tactus, counting by minims in common time, and with proportions for triple time. There are standard expectations for the speed of common time, and for the appropriate proportion for particular dance-types.

The model of perfect time-keeping is the Pendulum. The practical means of time-keeping is by moving the foot, down for one minim, up for the next.

The performers’ role is not to impose their own ‘interpretation’ on the piece, but to find out the Humours that are already there. Keeping Time is essential, for finding out the Humours.

The principle means of expressing changes in Humour is dynamic contrast.  A secondary means of expression is the Pause, in particular places.

Dfferent movements can have different tempi, even tempi that are unexpected for the dance-type that the piece seems to resemble, if the Humour demands it.

Fast pieces, and even some slow pieces, can have one or pauses before the end, but the conclusion ‘whirls away’ without rallentando.

I see no evidence at all for Rubato within a phrase or movement.  Mace’s Liberty of ‘fast and slow’ is between one movement and another, or between the standard tempo for a certain dance-type and the specific tempo for a piece in a particular humour. The only other Liberty in time-keeping that he mentions is the Pause.

All this is consistent with what we read in other sources of this period, whether English, Italian or French.

I give Thomas the last word:

Keep True Time


In modern English: Keep True Time, which you must be extremely careful to do, in everything you play (page 124).