Tempo in Baroque Music and DanceAuthor(s) : Esther Coorevits and Dirk Moelants

Tempo in Baroque Music and DanceAuthor(s): Esther Coorevits and Dirk Moelants
Source: Music Perception: An Interdisciplinary Journal , Vol. 33, No. 5 (JUNE 2016), pp.
523-545
Published by: University of California Press
Stable URL: https://www.jstor.org/stable/10.2307/26417332
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/10.2307/26417332?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
University of California Press is collaborating with JSTOR to digitize, preserve and extend
access to Music Perception: An Interdisciplinary Journal
This content downloaded from

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
All use subject to https://about.jstor.org/terms
Tempo in Baroque Music and Dance 523
TEMPO IN BARO Q UE MUSIC AND DA N C E
R
ESTHER COOREVITS & DIRK MOELANTS EGARDLESS OF STYLE OR REGION OF ORIGIN,
Ghent University, Ghent, Belgium music and dance are closely related art forms.
Not only is dance almost always accompanied by
GROWING INTEREST IN STUDIES ON THE RELATIONSHIP music, the corporeal aspects of specific music genres
between music and movement has given rise to many influence the way we interpret and perceive music.
paradigms and theories, including embodied approaches Insight into the relation between dance and music is
that provide interesting methodologies in studies on particularly relevant for the Baroque period for a num-
music and dance. Insight into the relation between dance ber of reasons. First, the direct link between music and
and music is particularly important for the Baroque dance practice is very important in this era. An impor-
period, as a direct connection between music and dance tant share of Baroque music was composed with the
was omnipresent, even if music was not used to dance to. intention to serve as dance music in a social context
Many types of Baroque dances existed, each of them with or in theatrical settings. Second, the influence of dance
particular dance steps and a specific character, requiring music is very prominent in other genres; instrumental
a specific tempo. However, in music performance prac- music often uses dance movements as a model, but also
tice today, the link with the original dance movement is in vocal genres such as cantatas or oratorios, references
often lost and the tempo variation can be very large. The to dance movements are prominently present. There-
aim of this study is to compare the interpretations of fore, music of the Baroque era cannot be fully under-
dancers and musicians regarding Baroque music and stood without insight in the dance practice of that
dance in an experimental setting. First, we investigate the period. Up to now, scholarly research on Baroque
influence of dance movement on the musical interpreta- dances has mainly focused on either dance or music
tion of a series of Baroque dances. The pieces were and was generally based on written sources, while few
recorded both with and without dance accompaniment attempts have been made to shed light on the mutual
and the tempo and timing in the different versions were interaction between dance and music in an actual per-
compared. In the second part, dancers performed a par- formance context.
ticular choreography to music that varied in tempo. In this paper, two empirical studies that explore the
Video analysis and questionnaire data were used to eval- interaction between dance performance and the inter-
uate the different performances. The results were com- pretation of Baroque music are presented. Most research
pared with the tempi of music recordings of the same on timing in classical music has focused on post-Baroque
dance types, showing a clear difference between music repertoire (e.g., Clarke, 1982; Clynes, 1987; Palmer, 1989;
and dance performance. Musicians adapt their interpre- Repp, 1992; Todd, 1985). Examples from the corpus of
tation when performing together with the dancers, and Baroque music have been used to illustrate aspects of
the optimal tempo range found for certain Baroque expressivity (De Poli, Rodà, & Vidolin, 1998; Povel,
dances coincides only partly with the tempi commonly 1977), interpretation of musical structure (Cook, 1987),
found in music recordings. The direct link between and motor skills (Shaffer, 1981), but not in the context of
music and movement and its mutual influence illus- historical performance practice. Empirical studies that
trates the importance of an embodied approach in deal with issues such as timing and expressivity in his-
music performance, where in this case dance movement torical performance practice are quite rare. Although
gives concrete information for a ‘‘historically informed’’ phenomena such as overdotting (Fabian & Schubert,
performance. 2004), onset asynchronies in harpsichord performance
(Gingras, Asselin, Goodchild, & McAdams, 2009), or
Received: November 24, 2014, accepted June 5, 2015. ‘‘notes inégales’’ (Moelants, 2011), have been studied in
the past, to our knowledge, the relation with dance move-
Key words: performance studies, dance & music, tim-
ment has not yet been taken into account in research on
ing & tempo, movement, historical performance practice
timing and expressivity.
Music Perception, VOLUM E 33, ISSU E 5, PP. 523–545, IS S N 0730-7829, EL ECTR ONI C ISSN 1533-8312. © 2016 B Y THE R E GE N TS OF THE UN IV E RS I T Y O F CA LI FOR NIA A LL
R IG HTS RES ERV ED . PLEASE DIR ECT ALL REQ UEST S F OR PER MISSION T O PHOT O COPY OR R EPRO DUC E A RTI CLE CONT ENT T HRO UGH T HE UNI VE R S IT Y OF CALI FO RNIA P R E SS ’ S
R EPRIN TS AND P ERMISSI ONS WEB PAG E , HT T P :// W W W. UC PRESS . EDU / JO URNA LS . PHP ? P ¼R E P RI NTS . DOI: 10.1525/ M P.2016.33.5.523

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
524 Esther Coorevits & Dirk Moelants
At the same time, we see a growing interest in studies development of dance as an art discipline was France.
on the relationship between music and movement in Court dancing, refined and graceful in style and tech-
more general terms. The topic appears in different nique, symbolized French culture and was considered
areas, such as musical development (Phillips-Silver & superior to any other dance form in Europe (Little &
Trainor, 2005), music education (Campbell, 1991), reha- Jenne, 2001). The dissemination of French culture
bilitation (Thaut & Abiru, 2010), musical communica- through noble dance was not the least accomplished
tion (Moelants, Demey, Grachten, Wu, & Leman, 2012), by French dancing masters who were employed at every
or auditory-motor coupling (Leman et al., 2013). This notable European court. To quote Pierre Rameau
embodied approach has given rise to a large spectrum of (1725): ‘‘Il n’y a point de cour dans l’Europe qui n’ait
methods that can be applied to study the relationship un maı̂tre à danser de notre nation.’’ [There isn’t any
between dance and music. Studies such as these cover European court that hasn’t employed a [French] danc-
a wide range of interests in the field of psychology, for ing master.] Along with the dance, they spread the
example, the development of rhythmic synchronization music accompanying it. Thus, both the European aris-
(Zentner & Eerola, 2010), the study of beat deafness tocracy and court musicians and composers came into
(Phillips-Silver et al., 2011), or the relation between contact with French dances and their music. Also, music
dance and induced or perceived emotion (Burger, Saar- theorists of the 17th century encouraged musicians to
ikallio, Luck, Thompson, & Toiviainen, 2013; Van Dyck, refine themselves in the art of dancing in order to feel
Maes, Hargreaves, Lesaffre, & Leman, 2013). From the natural rhythmical flow and pulse of the music
a musicological point of view, considerable interest has (Muffat, 1698).
been directed towards the relation between structural The chief tools in the dissemination of court dance
aspects of music and movements (e.g., Burger, Thomp- were the many dance treatises and manuals that were
son, Luck, Saarikallio, & Toiviainen, 2013; Van Dyck, published and spread throughout Europe. In 1700,
Moelants, et al., 2013). Other studies have focused on a dance notation system developed by Pierre Beau-
the relation between dance and music in specific gen- champ was published by Feuillet in his Chorégraphie,
res, such as samba and charleston (Leman & Naveda, ou l’art de décrire la danse (1700). It elaborated an
2010) or flamenco (Maduell & Wing, 2007). Artists extended and complex dance vocabulary and was the
also have been interested in the relationship between most popular system at that time. Its chief components
dance and music. This interest has resulted in colla- are symbols describing the movements of the dancers’
borations with engineering departments to create feet and legs, the distribution of these symbols along the
interactive environments for dance performance (e.g., dance path where the bars of music are indicated, and
Camurri et al., 2000). However, until now, the relation the corresponding music tune at the top of the page,
between historical dance and music performance has showing the relation between music and dance (Pierce,
not been a topic of empirical studies. On the other 1998).
hand, several studies have stressed the importance of The vision of the dancer-choreographers was that
dance for the interpretation of Baroque music, based writing down dances in a fixed system could ensure the
on the study of written sources, eventually combined correct and complete transferal of dances from one
with personal experience (Franko, 2011; Little & Jenne, generation to another (Franko, 2011). The codified
2001). They did not, however, systematically investi- steps — the path of choreography and its relation to the
gate actual performances. music — give us a lot of information to apprehend court
In this paper, we will first go deeper into the relation dance. But at the same time, the complexity of these
between Baroque dance and music, give an overview of systems can obscure the correct interpretation. Dance
some aspects of dance that have a crucial influence on masters were aware of this issue and wrote treatises
the interpretation of the music, and present two experi- about the interpretation of dance notation. Famous
ments in which the relation between music and dance is examples include Traité de la Cadence (Feuillet &
explored in a real performance context. Pécour, 1704) and Le Maitre à Danser (Rameau,
1725). Still, a lot of aspects remain open to interpreta-
Dance and Music in the Baroque Era tion, especially the relationship between choreography
and music. Two scholars who made a major contribu-
During the 17th century, dance reached its status as tion to research on the relation between Baroque music
a serious art discipline in the Western tradition, which and dance are Wendy Hilton (Hilton, 1977, 1981, 1986)
eventually would lead to the ‘‘romantic’’ ballet of the and Meredith Little (Little, 1975a, 1975b; Little & Jenne,
19th century (Franko, 2011). The leading nation in the 2001). Their studies explore the different aspects of

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
Feuillet notation and derive characteristics of both involves many dimensions, one of the main problems
dance and music to unfold the interaction between cho- when dealing with the interpretation of Baroque dance
reography and music tune. music seems to be the choice of the tempo. In the next
But the influence of dance practice goes beyond section, the role of tempo in dance and music perfor-
actual dance music. During the 17th century, stylized mance will be explored by means of historical sources
dance music emerged that contained more elaborated and contemporary research, to provide a framework in
melodic and rhythmic traits, intended for concerts which we designed the two studies presented in this
without actual dance performance (Burkholder, Grout paper.
& Palisca, 2010; Spitzer & Zaslaw, 2005). One of the
questions here could be whether the original dance Tempo and Its Relationship to Choreography
practice is still traceable in the performance of these
works. Despite the absence of choreography, dance and Tempo-indications in Baroque music are mostly
music were not entirely disconnected, even if the music expressed verbally and, by consequence, they are open
was written for solo-instruments such as the lute or the to interpretation. They can refer to the character, move-
viol, which were normally not used to accompany ment, or expression in which the music should be
dance performances. Both musicians and composers played or performed, for example vivement or gay,
had practical knowledge of ballroom and theatre dances, andante, or allegro (Cyr, 1998). Apart from verbal indi-
which inevitably influenced their compositions and per- cations, the time signature or meter of a music piece
formances. Dance music, whether it was intended for bears a connotation of tempo too. In general, the speed
ballroom or concert performance, maintained its air de of the counting unit (the denominator of the fraction)
mouvement1 (Bayle, 1992): ‘‘L’esprit est le même et l’exé- slows down when the unit gets smaller. A bar in 3/2 is
cution si précise musicalement, qu’on connait d’abord de then supposed to be played slower than a bar in 3/4.
quelle espèce chaque aire est et qu’on se sent comme This way, tempo and meter are closely related to each
inspirer, même malgré soi, l’envie de danser.’’ [The spirit other. Second, a pitfall when reading tempo-indications
is the same and the performance musically so precise that is the metrical level the indication is pointing to. A dance
one recognizes at first of which type the tune is, and gets could be interpreted as ‘‘slow’’ at the level of the half note,
inspired, even despite yourself, by the urge to dance.] but as ‘‘lively’’ on the level of the quarter note. Also,
(Muffat, 1695). J. S. Bach, for example, wrote a large composers and music theorists did not always take the
amount of music of which not only the title but also the notated rhythm into account when writing down the
musical traits imply a connection to French court danc- musical meter. In his Nouvelles Suites de Pièces de Cla-
ing (Hsu, 2012; Little & Jenne, 2001; Qureshi, 1994). vecin (1727), Rameau notates a courante with the metric
Some of these dances were also actively performed to indication 3, using three half notes per bar, while the next
accompany social dances at the court, but other popular dance is a sarabande, again in 3, but with three quarter
dances in instrumental music (e.g., the allemande) were notes per bar. This ambiguous relationship between
not performed as choreographies anymore in the 18th tempo, meter, and rhythm does not facilitate the inter-
century. So, one could ask if composers and musicians pretation of Baroque dance music. But, by exploring
still bore the original dance traits in mind when compos- metrical and rhythmical structures, it is possible to find
ing and playing this kind of dance music (Little & Cusick, some indications for the tempo and character of the
2014). music (Little & Jenne, 2001). In any case, playing music
This mutual interplay between dance and music in the appropriate tempo was considered essential to
opens a lot of questions concerning interpretation, good musical taste, as Étienne Loulié points out in his
both for dancers and musicians interested in historical Éléments (1696): ‘‘Mais je me flatte que ceux qui ont le
performance practice. What is the appropriate phras- goust fin & qui ont éprouvé combien un Air perd de sa
ing, articulation or bow technique for each dance? beauté lorsqu’il est executé trop viste ou trop lentement,
How is the character of each dance expressed in the me sçauront bon gré de leur donner un moyen . . . pour
music? Though the interpretation of dance and music en connoı̂tre le veritable mouvement . . . ’’ [I appreciate
that those who have good taste and have experienced
1
how much an Air loses its beauty when it is performed
In French sources from the Baroque period, the expression air de too fast or too slow, will be willing for me to give them
mouvement is often used to indicate not only the tempo but also the
character of a musical piece. This is a clear example of how (dance)
a means . . . to know its actual character and tempo . . . ]
movement was closely related with music performance during this In dance music, the correct tempo is essential for the
period. character of the dance, the corresponding choreography,

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
and the distinction of different dance genres (Donington, The first study is a case study that looks at the per-
1989). French music theorists were aware of this problem formance of one specific piece, comparing the rhythmic
at that time, and developed an instrument to define aspects of a purely musical performance and perfor-
tempo-indications more exactly, relying on the move- mances with the actual dance. Here, we want to examine
ment of a pendulum (de La Chapelle, 1737; Loulié, the actual (historical) dancers‘ influence on the timing
1696; l‘Affilard, 1705; Meyerhöfer, 1996; Pajot, 1735). of the music performance. In the second study, we
Pendulum data were preserved for a small selection of choose two popular historical dances that are often
dances, but they are not as unequivocal as our present found in music compositions: the courante and the sar-
metronome numbers. The numbers could refer either to abande. We focused on the cradle of Baroque dancing,
the length of the pendulum or the duration of a beat, the France, working only with French music and French-
metrical level they refer to is not always clear, and more- style choreographies. In a dance experiment, specialized
over, it is not clarified if they refer to a complete (back Baroque dancers performed a typical choreography of
and forth) or half a swing of the pendulum. This can have both dances in a wide range of tempi. They gave feed-
major implications for the interpretation of the tempo of back on each performance, and afterwards, aspects of
the music and hence, the correct interpretation of these their performance were examined through video anal-
pendulum data is still under discussion (Harris-Warrick, ysis. This approach gives us a good view on the relation
1993; Heijdemann, 1984; Hilton, 1981; Jerold, 2010; between dance and musical tempo. The results of this
Kroemer, 2001; Schwandt, 2013; Van Biezen, 1984). This study are then compared to tempi found in commer-
is where actual dance practice can give insight into the cially available music performances of the same dance
difficult question of how to deal with tempo in Baroque types, which allows us to compare common perfor-
dance. mances with the preferences of the dancers.
First, the tempo of the music should be appropriate
for the execution of dance steps, especially the leaps. Experiment 1: The Influence of Dance on Music
These can be performed slow or fast, but it is physically Performance
impossible to remain suspended in the air infinitely.
Also, one measure of music may contain many steps, For this case study, a piece with a number of different
which would take a dancer more time and thus a slower dance characters in different tempi was required. In
tempo. In general, Baroque court dance should be per- collaboration with the dancers, ‘‘The Submission’’ was
formed with an air of effortlessness and elegance, and selected, a ballroom dance created in 1717 by the
all dance steps and movements should be fluently English dancing master Kellom Tomlinson (c.1690-
merged into each other. A skilled dancer will be able 1753) to music by Jean-Baptiste (John) Loeillet (1680-
to perform the dance steps more quickly without get- 1730) (the first page of the original edition is shown in
ting untidy and will not lose the smoothness in per- Figure 1). Tomlinson was an English dancing master
forming steps very slowly (Little, 1975b). Also the who was educated in French style dancing and pub-
character of the dance is important for the tempo: cho- lished The Art of Dancing, the English translation of
reographies containing sliding steps and few hops bear Feuillet’s Chorégraphy. His choreographies survive in
a more dignified, serious character and therefore will be Feuillet-Beauchamp notation and hence, ‘‘The Submis-
performed more slowly than choreographies contain- sion’ is a choreography in French style, though com-
ing many leaps and jumps. The way a specific step is posed in England (Thorp, 2012). It consists of three
executed or articulated, for example, a deep and slow musical parts, each with its own character, on which
plié (bending of the knees) as opposed to a fast and one choreography is danced. A slow movement in 3/4
crispy one, tells a lot about the character of the dance is followed by a lively menuet, also in 3/4, while the last
and depends largely on the tempo of the music (Little, movement is a fast rigaudon in 4/4. The two dancers
1975a). In research on Baroque dance music, some that performed the choreography were experienced spe-
scholars already took the notated choreography into cialists in historical dance and ‘‘The Submission’’ was
consideration (Cobau, 1984; Harris-Warrick, 1993; part of their repertoire, as they had performed it repeat-
Hilton, 1977; Little, 1975a; Little & Jenne, 2001; Qure- edly in public performances. They were coupled with
shi, 1994), but up to now, systematic research on actual a music ensemble consisting of a violin, a viola, a viol.
dance practice is lacking. In this paper, two studies are and a harpsichord. The leading violinist had recorded
presented which explore the relationship between the piece on CD (Loeillet, 2012), the other musicians
tempo, music practice, and dance performance in a his- performed the music for the first time, but had the
torical context. opportunity to practice in advance.

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
tempo of the whole, we based our analysis on the viola

da gamba part (bass). The bass part articulates every
strong beat and, in contrast to the harpsichordist and
the violinist, the gamba player does not add ornaments,
so the basic time structure is found most clearly in this
part. To determine the length of each event, a manual
onset detection system using the program praat
(Boersma & Weenink, 2013) was used. Praat analyzes
pitch and intensity automatically and provides a spec-
trogram notation visualization, with parallel audio feed-
back. The use of this program makes it relatively easy to
locate the onsets with a precision of a few milliseconds
when using recordings made with contact microphones.
In order to obtain a constant view on the timing, the
data from the individual time intervals were transformed
to a series of ‘unit times’ by dividing their total duration
by their notated note length. As such, the duration for
each eighth note is obtained, corresponding to the lowest
level in the metric grid. From these values we can easily
compute durations at other metric levels. The series of
unit times gives us information on the timing at the level
of the individual event (Figure 2). Next to this, the tempo
at the notated beat level was computed, based on the
duration of the individual events, and transformed to
FIGURE 1. First page of the choreography “The Submission” in Feuillet a ‘beats per minute’ (BPM) scale. The results of the tim-
notation (K. Tomlinson) with the music tune (J. Loeillet) at the top of the
ing analysis are summarized in Figure 3.
page.
Results & Discussion

Method
A first analysis showed that the difference between the
PROCEDURE two recorded instrumental parts without dance was
Initially, dancers and musicians rehearsed separately. As very small and never significant. Therefore, only the
such, they fixed their own ideas to the interpretation of second, final version will be used in the following anal-
the dance suite. At the end of the rehearsal, the music ysis (labeled ‘‘no dance’’). Using the unit times for each
performance was recorded twice, the musicians still event in the viola da gamba part (excluding the final
being unaware of the choreography. Subsequently, the events of the first two parts, as they contain relatively
dancers joined the ensemble. First, musicians and dan- long breaks in some of the versions, cf. infra), we found
cers performed the dance together, without any prior a significant difference between the four recorded ver-
communication. After this first joint performance, the sions in each of the three movements. For the slow
dancers provided feedback and worked together with movement, F(3, 368) ¼ 96.86, p < .001, and the menuet,
the musicians to come to a performance in which every- F(3, 377) ¼ 121.71, p < .001, post hoc Scheffé tests
body agreed on the interpretation, which took two more showed that all differences between versions were sig-
run-throughs. Individual recordings of the musicians nificant at a p < .001 level, except the difference between
were made using contact microphones. In addition, the the last two versions, which is nonsignificant. For the
complete ensemble-sound was recorded and a video rigaudon, F(3, 552) ¼ 18.82, p < .001, the image is
recording was made for control purposes. totally different, as it is the very last version which is
significantly different (p < .001) from the other versions,
ANALYSIS while none of the other three versions differ signifi-
The five recorded versions of the piece were analyzed: cantly from each other.
two instrumental versions without dance and three ver- Both graphs in Figure 3 clearly show the succession of
sions with dance (one without and two with feedback). the three parts as well as the differences between the
In order to provide an optimal view of the timing and four versions.

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 2. Illustration of the method of analysis. On top the score of the first four measures of “The Submission” followed by the analysis in praat
(soundwave, spectrogram with indication of the amplitude and annotated onsets with interonset intervals), and the unit times and beat times
calculated from these ioi’s (all in ms). All numbers are rounded to the nearest millisecond.
In the first two movements the musicians immedi- movement continued. The plié (bending of the knees),
ately played faster when the dance was added as com- for example, notated as the first movement in the cho-
pared to the version without dance. Even without oral reography of the menuet, should be performed at the
communication (dance1), musicians felt that the dance end of the last bar of the slow movement, so the élevé
demands a particular tempo and they adapted their (rising movement) that follows the plié can be per-
playing. After communicating with the dancers, the formed on the accentuated first beat of the menuet.
tempo was adjusted even more and performed even A similar thing happens at the beginning of the rigau-
faster (dance2). After this session, the dancers were don, where the landing after a leap marks the first beat.
happy with the tempo for the first two parts, so the As such, following the feedback by the dancers, the
musicians kept the tempo equal (dance3). However, they three dances are concatenated without a break and
thought the final part was performed too fast now and smoothly merged into one performed choreography.
therefore, the third part was played consistently slower A final effect concerned the lowering of the variance
in the last version. This shows that the adaptation to the in the versions with dance, notably in the first two parts.
dance is not merely a question of speeding up. Rather, it Figure 4 shows the standard deviation of the tempo in
is related to the specific temporal character of the dance, the three different parts for the four versions, expressed
which requires a precise match with the tempo of the as the percentage of the mean tempo. Clearly the vari-
music, regardless of the direction of the change. ability decreased in the versions with dance. Also, in this
Two other changes could be observed. First, a change case a different attitude between dancers and musicians
in the transitions between the parts was noticeable. could be perceived. Whereas musicians tended to exag-
When recording the music without dance, the musi- gerate phrasing and rhythmic contrast, dancers actually
cians tended to take a break between the parts. In the preferred a more stable performance in which they
first version including dance, this effect largely could maintain the cadence of their movement.
remained, as the musicians stopped and waited for the Three takes were necessary to come to a music per-
dancers to signal the new start. However, the dancers formance that was suitable for the choreography.
perceived the choreography as a whole and needed flu- This illustrated the critical relation between music and
ent transitions between the different parts, as the dance dance. Three issues were identified: the main tempo, the

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 3. Comparison of the tempi used in the four versions of “The Submission.” On the top (a) error bars show mean and 95% confidence interval
of the mean, based on the unit length of every event. The line chart on the bottom (b) shows the linear evolution at the beat level.
transition between different parts, and the variability of On the other hand, we saw that musicians were able
the tempo. This shows that musicians and dancers have to adapt their timing and keep the same tempo and
different views on the music and illustrates the impor- timing in successive performances of the same piece.
tance of knowledge about dance practice for musicians In the next chapter, we will focus on the first of the
concerned with ‘‘historically informed’’ performance. issues mentioned: the main tempo, which is closely

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 4. The evolution of the variance over the four performances in the three parts of “The Submission.” The markers represent the standard
deviation of the tempo of all beats, expressed as a percentage of the global tempo of each performance.
connected to the character of the dance and the feasi- the surviving choreographies, leaps and pirouettes can
bility of the choreography. be found. Though the character of the sarabande is
mostly described as dignified and serious, more frivo-
Experiment 2: Courant and Sarabande in Music lous sarabandes exist that mostly refer to the initial
and Dance Spanish character of the dance (Lancelot, 1995).
The following study consists of two parts: first, the
The second experiment focuses on the courante and the tempo of recorded sarabandes and courantes by French
sarabande, two popular dance types that were regularly composers, roughly covering the period 1650-1750, was
used in the typical dance suites from the Baroque analyzed. In the second part, we looked at choreo-
(Fuller, 2012). The courante typically found in the graphic performances to explore how dance movement
French Baroque suite is a very noble, serious dance. The is influenced by tempo. Dancers were asked to perform
emphasis is on the metrical ambiguities and the contra- the same courante and sarabande repeatedly, in a wide
puntal structure, which makes the courante a sophisti- range of tempi. Questionnaires and video analysis were
cated, stylized dance (Little & Cusick, 2014; Veilhan, used to determine the ‘optimal’ tempo. Finally the
1977/1979). Being Louis XIV’s favorite court dance, it results of the dance experiment were compared with the
was named ‘‘the king’s dance’’ and it was said to be the tempi found in the music recordings.
most ‘‘noble’’ dance of all. Its dance vocabulary is sober
but rhythmically intricate, and leaps and jumps are Analysis of Music Recordings
absent (Lancelot, 1995).
Like the courante, the sarabande is a serious, noble METHOD
dance. The most typical characteristic trait is a dotted Material. Five hundred and seven sarabandes and 437
pattern, which provides the sarabande its gracious char- courantes composed by Jacques Champion de Chambon-
acter. This dance is surrounded by a melancholic atmo- nières (c. 1601-1672), Louis Couperin (c. 1626-1661),
sphere but is rhythmically less ambiguous than the Jean-Henri d’Anglebert (1629-1691), Jean-Baptiste
courante (Hudson & Little, 2014). The dance steps used Lully (1632-1687), Marin Marais (1656-1728), Elisabeth
in the sarabande are usually more complicated, and in Jacquet de la Guerre (1665-1729), François Couperin

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
(1668-1733), Jacques-Martin Hotteterre (1674-1763), skewed (0.64 for the courante, 1.75 for the sarabande),
and Jean-Philippe Rameau (1683-1764) were collected indicating that there is quite a large group of slower
using Spotify. In many cases, the selection contained sarabandes, which is counterbalanced by a small group
different recordings of a piece, played by different per- of fast pieces.
formers, but only recordings on period instruments
were retained. Using commercially available recordings Analysis of Dance Performance
without focusing on particular performers and avoid-
ing adaptations for modern instruments (e.g., piano or METHOD
modern guitar) assures that the collection gives a rep- Material. To ensure that all dancers started without
resentative view on tempi used in historically informed prior knowledge of the choreography, the choreogra-
performances of French Baroque music. It is important phies for both dances were newly composed for this
to note that musicians following the ideas of histori- study. Based on surviving choreographies and with
cally informed performance practice do want to recre- advice from four dance masters, basic choreographies
ate the music as it sounded at the time it was written, for a courante (Figure 6a) and a sarabande (Figure 6b)
contrary to a so-called ‘‘romantic’’ performance style in were designed in Feuillet notation, containing the char-
which the music is a starting point for a personal inter- acteristic steps of these dances. The dance path (trajectory
pretation of the musician. In this context, finding the the dancers follow on the dance floor) for both dances
tempo that matches the original dance movement is was quadratic, bringing the dancers back to their starting
highly relevant and we can assume that the musicians point (at the bottom left) at the end of the choreography.
want to match their performance with the dance’s orig- At the corners of the dance path, a tournement (turn -
inal character. indicated by a little hook) was added to the dance steps, to
An analysis of the scores shows that the sarabandes indicate on which feet the dancer should turn.
mostly (97.4%) have ‘‘3’’ as metric indication, which The courante consists of two phrases, each of them six
roughly corresponds with the modern 3/4-meter (Corr- bars long (bars are indicated with perpendicular lines on
ette, 1738; de Brossard, 1703; Houle, 1987). A few exam- the dance path) containing the following dance steps
ples were found written in 3/2 (N ¼ 8; 1.58%) or 6/4 (units of movements that go together): Pas court / Pas
(N ¼ 5; 0.99%). The courantes are mostly (94.1%) long / Pas long / Pas long avec ouverture de jambe / Pas
notated in 3/2, with some (N ¼ 26; 5.9%) notated in long / Coupé - Coupé // Pas court / Pas long / Pas long /
3/4. Whereas the tempi of the sarabandes were very Pas long avec ouverture de Jambe / Pas long / Tems de
similar despite the metrical difference, the group of courante.
courantes in 3/4 was clearly performed faster, with an The two dance phrases are almost equal, except for
average of 144.5 bpm against 68.9 bpm for the 3/2 the last bar. The first step of the choreography is a pas
group. As this 3/4-meter does not correspond to the court (first bar - full rectangle), which consists of two
meter used in the dance study, these 26 courantes were step-units: (1) a tems de courante (at the right side of the
excluded from further analysis. dance path) and (2) a demi-jeté (left side of the dance
path). The tems de courante comprises a plié (bending of
Procedure. Each piece was played back using Spotify with
the knees - inclined line in the circle at the bottom), an
the score at hand. The tempo was determined by music
élevé (upwards movement after the plié - horizontal line
experts who tapped along with the recording following the
in the circle at the bottom), a glissé (sliding of the feet on
notated meter from the score. Having the score at hand
the ground - circle in the middle), and the displacement
guarantees that the notated beat level is annotated rather
of weight to the other feet. The demi-jeté consists of
than the perceived tempo, which could be at another met-
a plié, élevé, and the displacement of weight. The second
ric level. The average tempo for each piece was notated
step in the choreography is the Pas long (second bar,
along with the details of the piece and the performers.
dashed rectangle). This step is similar to the pas court,
but the tems de courante is replaced by a pas coupé (3)
RESULTS (first unit in the pas long). The pas coupé consists of the
The tempo distributions for both dances are very sim- same movements as the tems de courante, but here,
ilar (see Figure 5), with means of 68.86 and 68.47 bpm there is already a displacement of weight in the begin-
for the courante and sarabande, respectively. The range ning, during the élevé. Hence, in the pas coupé, there is
of tempi is larger for the sarabande, which is reflected by two times a displacement of weight, while in the tems de
a higher standard deviation (18.7 for the sarabande courante, this occurs only once. The pas long avec ouver-
and 10.2 for the courante). The distribution is also more ture de jambe (fourth bar) is basically a normal pas long,

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 5. Histograms showing the distributions of tempi measured in recordings of French Baroque sarabandes and courantes.
FIGURE 6. Newly composed choreographies of the courante (a) and sarabande (b) in Feuillet notation.

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
but this step is embellished with the outwards opening version, some ‘‘notes inégales’’ (lengthening the first note
of the leg. In short, a bar in the courante always com- of a pair of eighth notes relative to the second, similarly to
prises two step-units, either (1) & (2), (3) & (2) or (2) & swing in jazz performance; cf. Moelants, 2011) were
(2). Only the last bar is an exception, as it consists of added as well, in compliance with the historical perfor-
only one step-unit, the tems de courante, which is the mance practice of French Baroque music. These versions
usual way to end a courante. The dance steps in the were transformed into MIDI files using Sibelius 5 and
courante are rather straightforward, but the complexity rendered in a variety of tempi. Starting from 80 BPM, we
is in the rhythmic structure, as two step-units are to be moved two ‘‘tempo-octaves’’ up and down based on
danced in one bar of ternary meter (3/2). a division of the tempo-octave in six equal steps, follow-
The choreography of the sarabande also contains two ing a log-2 scale, but rounded to the nearest whole num-
identical phrases, but here, each bar contains only one ber. Eighty BPM was chosen as a proper starting tempo
step-unit. Thus, two phrases of 8 bars in the sarabande for both dances, supported by literature on performance
have a similar length as two phrases of 6 bars in the practice of both dance and music (Bayle, 1992; Doning-
courante. The following steps were used in the sarabande: ton, 1998; Hilton, 1981; Quantz, 1752; Veilhan, 1977/
Tems de Courante / Pas coupé / Pas coupé en deux mou- 1979). This resulted in 25 tempo conditions (20 - 22 -
vements / Pas de bourrée emboı̂té / Tems de courante / 25 - 28 - 32 -36 - 40 - 45 - 50 - 56 - 63 - 71 - 80 - 90 - 101 -
Contretemps en avant / Pas de bourrée / Pas coupé // Tems 113 - 127 -143 - 160 - 180 - 202 -226 - 254 - 285 - 320
de Courante / Pas coupé / Pas coupé en deux mouvements BPM) that were applied to the quarter note of the melody
/ Pas de bourrée emboı̂té / Tems de courante / Contretemps of the sarabande, and the half note of the courante. At the
en avant / Pas de bourrée / Pas coupé. start of each version, three beats were given to allow the
A new dance step that was not used in the courante is dancers to prepare themselves.
the pas de bourrée (7th bar - rectangle). It contains three
steps (three times displacement of weight), of which the Procedure. Sixteen dancers (3 males and 13 females)
first one contains a plié and élevé. Another new step is participated in the experiment. Their experience in
the contretemps en avant (6th bar - dashed rectangle). It Baroque dance ranged from 1-3 years (N ¼ 6), 3-6 years
also starts with a plié, but now the élevé is extended with (N ¼ 3), to more than 6 years (N ¼ 7). Only three of
a leap (two lines in the circle), after which the dancer them were not able to read Feuillet notation. The exper-
lands on the same foot that already carried the weight. iment took place in different dance schools in Ghent
The Contretemps en avant is closed with a normal step. (B), Sint-Niklaas (B), Utrecht (NL), and Paris (F), in
The dance steps used in the sarabande are generally ballet rooms with an appropriate surface and size.
more embellished and varied than the ones we find in The dancers received the choreographies in advance
the courante (Lancelot, 1995). An example of such an and got the opportunity to rehearse them first with their
embellishment is given with the pas coupé en deux mou- dance master. Then, they were asked to perform the
vements (3rd bar - dashed rectangle). This is a normal sarabande and courante on the music presented to
pas coupé, but the second displacement of weight is them. In each tempo condition, they performed the
embellished with an extra plié and élevé (dashed circle). choreography twice with a short pause in between. After
Another embellished step is the pas de bourrée emboı̂té the two performances, they were asked to evaluate the
(rectangle at the right bottom). Again, this is a pas de tempo of the music on the basis of 1) how well the
bourrée, but instead of stepping three times forward, tempo expressed the character of the dance, 2) the flu-
only the last step goes in a forward direction. ency or smoothness of their dance movements, 3) if it
The music used for these choreographies was based was easy to synchronize, 4) if they could perform the
on existing melodies as found in surviving dance man- choreography correct and precise, and 5) how comfort-
uals. This was done to ensure that the music was suited able they thought the tempo was, all on a scale of 0 to 3.
for real dance performance. The melody of the dance Alternately, a faster and a slower tempo were played
suite ‘‘La Duchesse’’ (1701) was adopted for the cou- until the dancers indicated that it was impossible to
rante, and in the sarabande the music of ‘‘La Royalle’’ perform the choreography in the given tempo. The
(Pécour, 1712) was used. For both pieces, an orchestra- tempi were not randomized, because it was not known
tion was composed based on historical performance beforehand how fast or slow the dancers would be able
practice: a bass line with a typical basso continuo to perform both choreographies. The choreography was
accompaniment was added (see Figure 7) and the pieces a solo dance, hence the participants performed the
were orchestrated for transverse flute and violin (top dance alone and the evaluation of the performance was
line), harpsichord, and viola da gamba. In the final done individually to access the personal experience of

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 7. Basic scores of the courante “La Duchesse” (a) and the sarabande “La Royalle” (b). In the versions used in the experiment some additional
performance elements were added, such as an extra bass instrument, some rhythmic articulation and a few ornaments, mimicking the performance
style of specialized musicians.
each individual. The whole procedure was recorded on assigned a score of 12, in accordance with the number of
video. bars in the dance. Each dancer was rated individually for
all the performed tempi in both dances. The methodol-
Analysis. Two researchers that were familiar with ogy for the evaluation was objectified as much as possible
Baroque dance independently did the analysis of the by clearly delineating the criteria for the analysis.
video recordings of the dance performances. The eval- Within each bar of music, the synchronization of
uation was done manually and comprised three criteria: the dancer to the music was evaluated for each beat. For
the synchronization of the dancers to the music, the sarabande, the quarter note was assumed as the beat
the rhythmic timing or rhythmical cadence, and the level, for the courante the main beat was counted on the
precision and smoothness. For the sarabande, every level of the half note. As such, both dances had 3 beats
performance got a score on 16, while the courante was per bar. If the dancer was clearly visible out of sync for

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 8. Pas de Courante in Feuillet notation. On the left (a), the symbol indicating plié is marked, in the middle (b), the symbol for the élevé is
indicated. The pas coupé is indicated in the black rectangle on the right (c), the demi-jetté is indicated with a blue rectangle.
one of the three beats, the whole bar was evaluated as élevé or rise after the bending of the knees and the
unsynchronized. In case of doubt, synchronization for displacement of the weight of the dancer to the other
the whole bar was assumed. feet that should coincide with the first beat of the bar
To judge synchronization in Baroque dance perfor- (Figure 8b). This is a first reference point for
mance, some basic knowledge on dance notation has synchronization.
to be applied. Particular movements in a dance step have Then, the pas de courante can proceed in two ways.
to be performed on an exact moment in the musical bar, From the dance notation, it is clear that this dance step
as other movements can be performed in different rhyth- consists of two main parts (Figure 8c). The first unit,
mical ways (Feuillet, 1700; Hilton, 1997, 1977; Hsu, also known as pas coupé, comprises two times the dis-
2012). To clarify this, a typical dance step from the cou- placement of weight and is connected with a curved line
rante (know as the pas de courante) is analyzed here as an in the middle. The second unit only consists of one
example. Figure 8 shows this dance step as it was used in displacement of weight, accompanied with an élevé, and
the choreography of the courante. The duration of the is called the demi-jeté (blue rectangle in Figure 8c). It is
step comprises one bar of music (indicated by the hori- the timing of the second displacement of the pas coupé
zontal lines on the bottom and the top of the figure) or and the displacement in the demi-jeté that can vary,
three half notes. It is composed of different movements, according to the interpretation of the dancer (Feuillet &
indicated by the different symbols along the dance path. Pécour, 1704).3 The latter can occur on the third beat or
The first symbol starting from the bottom indicates third half note within the bar, or on the fourth quarter
a plié or bending of the knees (Figure 8a). This bending note of the bar, which divides the bar into two equal
should be performed before the first musical beat of the parts on the level of the quarter note, syncopating the
bar, though the placement of the symbol suggests oth- three-beat measure. The second displacement in the pas
erwise (Hilton, 1981; Pierce, 1998).2 It is the end of the coupé can also vary in timing. On the one hand, it can be
executed on the second half note or second beat, on the
2 other hand, it can be performed on the fourth quarter
This was a convention established in the 17th century, when the
Beauchamp-Feuillet system for notating dance was developed. Dancers
always ‘‘anticipate’’ the musical bar with a plié (bending of the knees), but
3
in order to keep the notation on paper clear and readable, they chose to This was discussed also the 7th Summer School on Baroque Dance
write the plié at the beginning of the bar instead of before. (July 1- 6, 2013) - Paris, with the dance master Béatrice Massin.

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
note, again causing syncopation or a subdivision of the RESULTS

bar into quarter notes. A more detailed explanation on To test whether the results could be generalized inde-
this matter can be found in Feuillet (1700) and Hilton pendent from the dancer’s artistic background, a Wil-
(1977, 1997) but is beyond the scope of this paper. What coxon rank-sum test with pairwise comparison (95%
is discussed in this example of the pas de courante can be CI) was performed, comparing the results from the dif-
applied to the other dance steps present in the courante ferent groups of dancers. None of the results showed
and sarabande. It illustrates that synchronization in a significant interaction between groups with a different
Baroque dance performance is rather complex, as it training (according to the dance master) or level of
deals with differences in interpretation of the choreog- experience (beginner, 0-3; intermediate, 3-6; advanced,
raphy by the dancer. They can vary the steps they per- > 6 years of experience), so the results for all partici-
form within the choreography, and as such, each pants are presented together.
performance is different. For each dance step in both The results of the video analysis show that the dan-
choreographies, the reference points for synchroniza- cers achieved the highest quality at 80 BPM in the
tion (especially élevés and displacements of weight that performance of the courante (M ¼ 88.9%), and at
should occur on the beat) were determined to use them 113 BPM in the performance of the sarabande (M ¼
in the video analysis. 80.8%) (Figure 9). The sarabande got relatively high
Next to synchronization, the technical quality of the quality scores for all the tempi between 80 and 127
dance performance in terms of precision and smooth- BPM, in contrast with the courante where the peak is
ness was evaluated for each tempo. Precision expresses much sharper. Also, it can be seen from Figure 9 that
whether the dancers made a clear distinction between the dancers synchronized best at 90 BPM in the cou-
the different dance steps and movements within the rante (M ¼ 88.9%) and at 101 BPM in the sarabande
choreography or not. There should be a clear differ- (M ¼ 82.9%). Between 56 BPM and 143 BPM, all
ence, for example, between gliding and normal steps. If dancers were able to perform the sarabande and
all the different parts of the movements in a bar were nobody was able to perform it slower than 40 BPM
perceivable in the performance, the bar was counted as and faster than 226 BPM. For the courante, all dancers
precise. Smoothness expresses if the dancers perform performed the choreography between 56 and 127 BPM
without interruptions or hesitations in their move- and limits were found at 36 BPM and a 180 BPM.
ments. When no such event was clearly perceivable When interpreting the results, beware that in the
in the execution of the dance steps within a bar, the extreme tempi, not all 16 dancers managed to perform
bar got a positive score for smoothness. When the the choreography.
scores of the two researchers differed, they discussed The results of the evaluation by the dancers seem to
the results in order to come to an agreement on the support the findings from the video analysis, though
bars that got a different score. Summing the scores for there are small differences. The average score for all the
all bars gave a total score for precision and smoothness criteria was the highest at 80 BPM for the courante
for each tempo condition. Together, these two para- (2.88/3 or 95.8%) and at 90 BPM for the sarabande
meters resulted in a rating for the technical quality of (2.73/3 or 91.1%) as can be seen in Figure 10. Though
the performance. the answers on the different questions that were evalu-
Apart from the video analysis, the evaluation of the ated after each tempo were highly correlated (Spear-
dancers was taken into account. These ratings on a 4- man’s Rank-correlation test, p < .05 for all five
point scale (0-3) gave an overview of how the dancers questions), the separate questions are taken into con-
themselves experienced the choreography in a given sideration in order to compare the results with the video
tempo. This subjective experience could then be com- analysis. The peak for experienced technical quality in
pared with how an external observer perceived the per- the courante is reached at 80 BPM, which is also the
formance in the video analysis. Especially their ratings tempo at which the dancers experienced optimal syn-
of synchronization, smoothness, and precision could be chronization. At this tempo, the perceived technical
compared with the results of synchronization and tech- quality seems to be the highest as well, as was found
nical quality in the video analysis. For this purpose, the in the video analysis, while perceived synchronization got
scores for smoothness and precision were considered the highest scores at 90 BPM (Figure 11a). Figure 11b
together, similar to the video analysis. This two-fold shows that the experienced technical quality and syn-
evaluation of the dance performance resulted in an chronization by the dancers peaked at 90 BPM in the
overview of how tempo influences the dance perfor- sarabande, while in the video analysis, the quality of
mance of the courante and the sarabande. performance scored best at a 113 and synchronization

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 9. Scores for synchronization and technical quality as found in the video analysis of the courante (a) and sarabande (b).
at 101 BPM. Thus, experienced and perceived technical perception of the dancers than is shown by a more objec-
quality and synchronization seem to lay close to each tive analysis of the recordings.
other, though there can occur a difference of one or two Character and comfort, the two criteria that were only
tempo levels. The most striking difference between the assessed by means of evaluation by the dancers, are
evaluations by the dancers and the video-analysis are the displayed in Figure 12. These highlight other, more sub-
much steeper slopes in the first one. Clearly, the subjec- jective aspects of quality, as opposed to the pure tech-
tive experience of synchronization and quality during the nical aspects of performance. Again, in the courante, the
performance is much more affected by the negative peak for character and comfort lies at 80 BPM and for

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 10. Mean score for all criteria evaluated by the dancers in the courante and sarabande.
the sarabande at 90 BPM. However, when the tempo of Monteclair, 1736; Freillon-Poncein, 1700; L’Affilard,
the music moves more towards the extremes, the curve 1702; Loulié, 1696; Masson 1705; Pajot, 1735) where the
for both criteria is steeper as compared to the mean courante is generally characterized as a somewhat slower
scores for all the criteria. In the courante, this is the case dance than the sarabande. These findings can now be
for how comfortable they thought the tempo was, both in compared with the resulting tempi from the analysis of
the faster and slower tempi. In the sarabande, the ratings the music recordings.
for comfort also lower in the slow tempi, but in the fast The general results from the music recordings and
tempi, it is the character of the dance that is lost first. This dance performances are summarized in Figure 13, where
was validated with a Wilcoxon signed rank test, where the we see the distribution of tempi found in the music
ratings for character and comfort are significantly differ- recordings (grey bars, with the columns centered around
ent from the median score in slow tempo (< 63 BPM) for the tempi used in the dance experiment), the average
both the courante (p < .05) and the sarabande (p < .001). score of the evaluations by the dancers (experience, black
In fast tempi (> 127 BPM), the WRT gave similar results line), and the average score of the technical quality
(p < .001 for both variables in both courante and sara- (smoothness and precision) and synchronization para-
bande). Hence, the subjective experience of the dancers meters from the video analysis (perception, grey line).
seems to be more sensitive to tempo changes than the Although there is a certain overlap, we see a clear
pure technical aspects of the dance performance. discrepancy between the peaks of the tempo distribu-
When comparing these different aspects, the optimal tions of the recorded music and the data from the dance
tempo zone regarding dance performance for the cou- performances. Especially for the sarabande, a large part
rante seems to lay around 80-90 BPM; for the sarabande of the recordings are performed at a tempo that is not
it is found around 90-101 BPM. The courante is a rhyth- really comfortable for the dancers. Although many of
mically more complex dance, but the dance vocabulary the recorded pieces are not actually written to accom-
used in the sarabande differs from the more sober cou- pany actual dance performances, we can assume that
rante in that it contains leaps and more lively steps (like the character of the original dances was preserved in
the pas de bourrée). This could suggest that a slightly the music (Little & Jenne, 2001; Wetzer, 1988). As the
faster tempo is more appropriate for the sarabande, character is largely influenced by the tempo, musicians
a finding that seems to be consistent with the description who want to perform the music as closely as possible to
of these two dances in historical sources (d’Alembert the original intentions of the composer (as in histori-
1752; de Brossard 1703; de La Chapelle 1737; de cally informed performance practice), should bear this

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 11. Synchronization and quality of performance as experienced by the dancers and perceived from the video analysis in the courante (a) and
sarabande (b).
in mind. In this case, this could convince the performers music by musicians substantially. Verbal communica-
to play courantes and especially sarabandes at a faster tion between dancers and musicians can help to gener-
tempo than they are generally used to. ate a music performance that fits the choreography best.
Even when bodily communication alone is used during
Conclusions performance, musicians experience an influence of the
dance movement on their interpretation, which mainly
The results of the first experiment confirm that dance affects their tempo and timing. This is of course closely
movement can influence the interpretation of dance related with a sense of cross-modal understanding of

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 12. Evaluation by the dancers of the criteria ‘character’ and ‘comfort’ compared to the mean score for all criteria in the courante (a) and
sarabande (b).
musical expression. Musicians communicate musical However, (Baroque) dance is not part of the regular
expression partly through bodily movements, and curriculum in music education, and musicians are often
likewise, their expression can be influenced by other not aware of the corporeal aspects of dance music.
musicians’ or dancers’ body movements that bare a par- Hence, an essential part of this music is lost. The results
ticular expression (Leman, 2008). In this light, it is also of the second experiment show that the interpretation of
striking that the musicians manage to keep a tempo the sarabande by musicians is in general slower than the
precisely once their performance fits the movement of tempo dancers prefer. A similar but less pronounced
the dancers. trend is observed for the interpretation of the courante.

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
FIGURE 13. Comparison of the experience and perception of the dance quality and the tempi found in the performances of French courantes (a) and
sarabandes (b) from the Baroque period. Lines show the normalized average ratings of the dancers (“experience”) and the video analysis
(“perception”), following the scale of left axis. Columns follow the scale of the right axis, and give the percentage of pieces fond in each tempo
category, with the tempo of used for dance stimulus as midpoint.
Moreover, the relationship of tempo between courante Lambert, 1702; Freillon-Poncelin, 1700; Masson, 1705;
and sarabande as defined by the dance experiment is in Montéclair, 1736; Muffat, 1698; Rameau, 1725; Rous-
accordance with the proportions of the time signatures seau, 1768). In the research field of historical music
3/4 en 3/2 and the characteristics of these two dance performance, the information gathered from dance and
types as described in historical sources of the Baroque choreography could be of great importance for the real-
period (d’Alembert, 1752; de Brossard, 1703; de Saint- ization of a historically correct interpretation. This idea

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
is supported by musicians who did make the effort to dances, but also to dances from the Renaissance (e.g.,
approach Baroque music from the point of view of cho- basse danse, pavane) or the Middle Ages (e.g., estam-
reography (Hsu, 2012, Qureshi, 1994), but also by musi- pie, saltarello).
cians that play Baroque music and reflect on shape and Outside the traditional field of early music, this
movement. Leech-Wilkinson and Prior (2014) noted, method could also be applied to folk dances, often
‘‘Shape is . . . very important . . . it’s partly tied in with found in music of the 19th and 20th century.
gesture, for me, particularly in Baroque music. There’s Undoubtedly, composers like Chopin, Brahms, Liszt,
a sense of something having a gesture . . . and it’s quite Dvořak, or Bartok were familiar with the central
physical . . . and I think that maybe comes from a lot of European dance movements figuring in their compo-
music being originated in the dance . . . ’’ sitions. But how much of the original dance character
On the other hand, the question still remains how is left in contemporary interpretations? Apart from
strictly the character of the dance was maintained when the relation between musical tempo and rhythmic
there was no active dance practice involved. As already character, embodied approaches could also shed new
mentioned, court composers were aware of the aspects light on breathing and prosody in vocal music and its
of dance performance, also when they composed con- effect on phrasing and articulation. In the study of
cert music, but an evolution to more complex forms instrumental music, changes in the construction of
inevitably leads to compositional and interpretational instruments (e.g., modern violin versus Baroque vio-
freedom. Also, not all dances were commonly danced lin), and the related changes in bodily engagement
during the Baroque period, which begs the question of with the instrument, would be another example of
whether dance movement did influence composition a research topic where empirical methods using
and performance in such cases. Nevertheless, collabora- embodied paradigms could help us gain insight into
tions between dancers and musicians can lead to new the nature of the differences between historical and
and fruitful insights in the puzzling language of Baroque modern practices.
dance.
In adopting an empirical approach using embodied Author Note
paradigms by comparing dance practice with the per-
formance of dance music, we were able to address The authors would like to thank Lieven Baert and Jose-
a complex issue encountered by many performers phine Schreibers, the two dancers involved in the first
dealing with music from the Baroque period: the experiment, as well as the musicians Stefaan Smagghe
choice of the right tempo. Studying music performance and Kersten Cottyn. Special thanks goes also to Guil-
from an empirical and embodied point of view creates laume Jablonka, Béatrice Massin, Dorothée Wortelboer,
new opportunities for the reconstruction of historical and Johan Goessens, the four dancing Masters who
music practices. This paper is explorative and provides taught us a lot and helped us with the choreography
a methodology and some examples of the results this and the second experiment. This research was partly
type of approach could lead to, which opens possibil- realized under the FWO-project ‘‘Foundations of
ities for further research. The first study is a case study Expressive Timing Control.’’
and should be repeated with other dances, different Correspondence concerning this article should be
ensembles and dancers, and performed in different addressed to Esther Coorevits, IPEM-Department of
circumstances to see if the exploratory results found Musicology, Technicum Blok 2, Sint-Pietersnieuwstraat
in this study can be generalized. The method used in 41, 9000 Ghent, Belgium. E-mail: esther.coorevits@
the second study can be applied to other Baroque ugent.be
References
B AYLE , C. (1992). Interprétation et musique en danse baroque. B URGER , B., S AARIKALLIO, S., LUCK , G., T HOMPSON , M. R., &
[Interpretation and music in baroque dance] In D. R. T OIVIAINEN , P. (2013). Relationships between perceived emo-
Wilson (Ed.), The marriage of music and dance (n.p.). tions in music and music-induced movement. Music
Conference Proceedings of the National Early Music Perception, 30, 517-533.
Association. B URGER , B., T HOMPSON , M. R., L UCK , G., S AARIKALLIO, S., &
B OERSMA , P., & W EENINCK , D. (2013). Praat: Doing phonetics by T OIVIAINEN , P. (2013). Influences of rhythm-and timbre-
computer [Computer program, Version 5.3.51]. Retrieved June related musical features on characteristics of music-induced
2, 2013 from http://www.praat.org/ movement. Frontiers in Psychology, 4, 183.

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
B URKHOLDER , P., G ROUT, D. J., & PALISCA , C.V. (2010). A history F EUILLET, R. A., & P ÉCOUR , L.- G . (1704). Traité de la cadence.
of Western music (8th ed.). New York: W. W. Norton & Company. Recüeil de dances [Treatment of the cadenza. Review of
C AMPBELL , P. S. (1991). Rhythmic movement and public school dances]. Paris: Chez le sieur Feüillet.
music education: Conservative and progressive views of the for- F RANKO, M. (2011). Writing for the body: Notation, reconstruction,
mative years. Journal of Research in Music Education, 39, 12-22. and reinvention in dance. Common Knowledge, 17(2), 321-334.
C AMURRI , A., HASHIMOTO, S., R ICCHETTI , M., R ICCI , A., F REILLON -P ONCELIN , J.-P. (1700). La véritable manière à jouer en
S UZUKI , K., T ROCCA , R., & V OLPE , G. (2000). Eyesweb: perfection du haut-bois [The only way of playing the oboe with
Toward gesture and affect recognition in interactive dance and perfection]. Paris: J.Collombat.
music systems. Computer Music Journal, 24, 57-69. F ULLER , D. (2012). Suite. In D. Root (Ed.), Grove music online.
C LARKE , E. F. (1982). Timing in the performance of Erik Satie‘s Retrieved from http://www.oxfordmusiconline.com/
‘Vexations.’ Acta Psychologica, 50, 1-19. G INGRAS , B., A SSELIN , P. Y., G OODCHILD, M., & M CADAMS , S.
C LYNES , M. (1987). What can a musician learn about music (2009). Communicating voice emphasis in harpsichord per-
performance from newly discovered microstructure principles formance. In J. Louhivuori, T. Eerola, S. Saarikallio, T.
(PM and PAS)? In A. Gabrielsson (Ed.), Action and perception Himberg, & P. S. Eerola Proceedings of the 7th Triennial
in rhythm and music (pp. 201-233). Stockholm: Royal Swedish Conference of European Society for the Cognitive Sciences of
Academy of Music. Music (pp. 154-158). Jyväskylä, Finland: ESCOM.
C OBAU, J. (1984). The preferred pas de menuet. Dance Research HARRIS -WARRICK , R. (1993). Interpreting pendulum markings
Journal, 16(2), 13-17. for French baroque dances. Historical Performance, 6, 9-22.
C OOK , N. (1987). Structure and performance timing in Bach’s C H EIJDEMANN , F. (1984). De tempo-aanduidingen bij L’Affilard en
Major prelude (WTC): An empirical study. Music Analysis, 6, Pajot (d’Onzembray) [Tempo indications from L’Affilard and
257-272. Pajot (d’Onzembray)]. In K. Vellekoop (Ed.), Tempo in de
C ORRETTE , M. (1738). L’école d’Orphée [The school of Orpheus]. Achttiende Eeuw [Tempo in the eighteenth century] (pp.
Paris: Mme. Boivin & Mr. Le Clerc. 26-36). Utrecht: Stimu.
C YR , M. (1998). Performing Baroque music. Portland OR: H ILTON , W. (1977). A dance for Kings: The 17th-century French
Amadeus Press. ‘‘courante.’’ Its character, step-patterns, metric and propor-
D’ ALEMBERT, J. L. R. (1752). Éléments de musique théorique et tional foundations. Early Music, 5, 161-172.
pratique, suivant les principes de M. Rameau [Elements of H ILTON , W. (1981). Dance and music of court and theater:
theoretical and practical music, following the principles of Selected writings of Wendy Hilton. New York: Pendragon Press.
M. Rameau]. Paris: Jombert. H ILTON , W. (1986). Dances to music by Jean-Baptiste Lully. Early
D E B ROSSARD, S. (1703). Dictionaire de musique [Music dictio- Music, 14, 51-63.
nary]. Paris: Christophe Ballard. H ILTON , W. (1997). Dance and music of court and theater:
D E L A C HAPELLE , J.-A. (1737). Les vrais principes de la musique Selected writings of Wendy Hilton. In L. Tomko (Ed.), Dance
[The real principles of music]. Paris: Veuve Boivin. and music series (Vol. 10). New York: Pendragon Press.
D E M ONTÉCLAIR , M.P. (1736) Principes de musique [Principles of H OULE , G. (1987). Meter in music, 1600-1800: Performance,
music]. Paris: veuve Boivin. perception, and notation. Bloomington, IN: Indiana University
D E P OLI , G., R ODÀ , A., & V IDOLIN , A. (1998). Note-by-note Press.
analysis of the influence of expressive intentions and musical H SU, C. (2012). The solo string works of J. S. Bach: The relation-
structure in violin performance. Journal of New Music ship between dance and musical elements (Unpublished doc-
Research, 27, 293-321. toral dissertation). Louisiana State University.
D E S AINT-L AMBERT, M. (1702). Les principes du clavecin conte- H UDSON , R. & L ITLLE , M.E. (2014). Sarabande. In D. Root (Ed.),
nant une explication exacte de tout ce qui concerne la tabulature Grove music online. Retrieved from http://www.oxford
et le clavier [Principles of the harpsichord containing an exact musiconline.com/
explanation of everything concerning tablature and the key- J EROLD, B. (2010). The French time devices revisited. Dutch
board]. Paris: Christophe Ballard. Journal of Music Theory, 15, 169-189.
D ONINGTON , R. (1989). The interpretation of early music. K ROEMER , J. (2001). Le cronometre de Monsieur Feuillet
London: Faber and Faber. (Unpublished doctoral dissertation). University of Vienna.
FABIAN , D., & S CHUBERT, E. (2004). Expressive devices and L ANCELOT, F. (1995). La belle dance. Catalogue raisonné fait en
perceived musical character in 34 performances of variation 7 l’an 1995 [Court dance. Catalogue raisonné made in the year
from Bach‘s Goldberg variations. Musicae Scientiae, 7, 49-71. 1995]. Paris: van Dieren Éditeur.
F EUILLET, R. A. (1700). Choregraphie: Ou l‘art de décrire la dance L‘ AFFILARD, M. (1702). Principes très faciles pour bien apprendre
[Choreography: Or the art of describing dance]. Paris: Chez la musique [Easy principles for the good understanding of
l‘auteur et chez Michel Brunet. music] (4th ed.). Paris: Christophe Ballard.

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
L‘ AFFILARD, M. (1705). Principes très faciles pour bien apprendre M OELANTS , D. (2011). The performance of notes inégales:
la musique [Easy principles for the good understanding of The influence of tempo, musical structure, and individual
music] (5th ed.). Paris: Christophe Ballard. performance style on expressive timing. Music Perception,
L A D UCHESSE . (1701). In Representation de differents especes 28, 449-460.
des danses [Representations of different types of dances] M OELANTS , D., D EMEY, M., G RACHTEN , M., W U, C.-F., &
(pp. 174-179). Bibliothèque nationale de France, Département L EMAN , M. (2012). The influence of an audience on perfor-
des manuscrits, Français 14884 [Online manuscript library of mers: A comparison between rehearsal and concert using
the National Library of France]. audio, video and movement data. Journal of New Music
L EECH -W ILKINSON , D. & P RIOR H. M. (2014). Heuristics Research, 41, 67-78.
for expressive performance. In D. Fabian, R. Timmers, & M ONTÉCLAIR , M. P. (1736). Principes de musique [Principles of
E. Schubert (Eds.), Expressiveness in music performance: music]. Paris: Veuve Boivin.
Empirical approaches across styles and cultures (pp. 34-57). M UFFAT, G. (1695). Florilegium Primum [First Florilegium],
Oxford, UK: Oxford University Press. Augsburg: Jacob Koppmayr.
L EMAN , M. (2008). Embodied music cognition and mediation M UFFAT, G. (1698). Florilegium Secundum [Second Florilegium].
technology. Cambridge, MA: MIT Press. Passau: Georg Adam Höller.
L EMAN , M., M OELANTS , D., VAREWYCK , M., S TYNS , F., VAN PAJOT, L.-L. (1735). Description et usage d’un métromètre, ou
N OORDEN , L., & M ARTENS , J.-P. (2013). Activating and machine pour battre les mesures & les temps de toutes
relaxing music entrains the speed of beat synchronized walk- sortes d’airs [Description and use of the métromètre, or
ing. PLoS ONE, 8, e67932. a tool for beating the bars and beats of all sorts of melo-
L EMAN , M., & N AVEDA , L. (2010). Basic gestures as spatiotem- dies]. Histoire de l’Académie Royale des Sciences, 1732,
poral reference frames for repetitive dance/music patterns in 182-196.
samba and charleston. Music Perception, 28, 71-91. PALMER , C. (1989). Mapping musical thought to musical per-
L ITTLE , M. E. (1975a). The contribution of dance steps to formance. Journal of Experimental Psychology: Human
musical analysis and performance: ‘‘La bourgogne.’’ Journal of Perception and Performance, 15, 331-346.
the American Musicological Society, 28, 112-124. P ÉCOUR , G.- L . (1712). La royalle. In Nouveau recüeil de danse de
L ITTLE , M. E. (1975b). Dance under Louis XIV and XV: bal et celle de ballet, contenant un très grand nombres [sic] des
Some implications for the musician. Early Music, 3(4), meillieures entrées de ballet [New collection of court and ballet
331-340. dances, containing a big number of the best ballet entrées]
L ITTLE , M. E., & C USICK , S. G. (2014). Allemande. In D. Root (pp. 1-5). Paris: M. Gaudrau.
(Ed.), Grove music online. Retrieved from http://www.oxford P HILLIPS -S ILVER , J., & T RAINOR , L. J. (2005). Feeling the beat:
musiconline.com/ Movement influences infant rhythm perception. Science, 308,
L ITTLE , M. E., & J ENNE , N. (2001). Dance and the music of 1430.
J. S. Bach. Bloomington, IN: Indiana University Press. P HILLIPS -S ILVER , J., T OIVIAINEN , P., G OSSELIN , N., P ICHÉ , O.,
L OULIÉ , É. (1696). Elements ou principes de musique mise dans un N OZARADAN , S., PALMER , C., & P ERETZ , I. (2011). Born to
nouvel ordre [Elements and principles of music, put in a new dance but beat deaf: A new form of congenital amusia.
order]. Paris: Christophe Ballard. Neuropsychologia, 49, 961-969.
L OEILLET, J. (2012). The submission. On The submission - P IERCE , K. (1998). Dance notation systems in late 17th-century
Sonatas, suites and concertos of the Loeillet family [CD]. France. Early Music, 26, 287-299.
Lummen: Etcetera Records. P OVEL , D.- J. (1977). Temporal structure of performed music:
M ADUELL , M., & W ING , A. M. (2007). The dynamics of Some preliminary observations. Acta Psychologica, 41, 309-320.
ensemble: the case for flamenco. Psychology of Music, 35, Q UANTZ , J. J. (1752). Essai d’une méthode pour apprendre à jouer
591-627. de la flûte traversières [Essay of a method for learning how to
M ASSON , C. (1705). Nouveau traité des règles pour la composition play the flute]. Berlin: Chretien Frederic Voss.
de la musique [New treaty of rules for the composition of Q URESHI , R. J. (1994). The influence of Baroque dance in the per-
music]. Paris: Christophe Ballard. formance of Johann Sebastian Bach‘s Six Suites a Violoncello Senza
M EYERHÖFER , D. (1996). Lichtenberg und die Sammlung Basso (Unpublished doctoral dissertation). Rice University.
wissenschaftlicher Geräte des Frankfurter Patriziers Johann R AMEAU, P. (1725). Le maı̂tre a danser [The dancing master].
Friedrich Armand von Uffenbach [Lichtenberg and the Paris: J. Villette.
collection of scientific instruments of the patricians R EPP, B. H. (1992). Diversity and commonality in music per-
from Frankfurt Johann Friedrich Armand von Uffenbach]. formance: An analysis of timing microstructure in Schumann’s
In Lichtenberg-Jahrbuch 1995 (pp. 113-160). ‘‘Träumerei.’ Journal of the Acoustical Society of America, 92,
Saarbrücken: SDV. 2546-2568.

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00
R OUSSEAU, J.- J. (1768). Dictionnaire de musique [Music dictio- VAN B IEZEN , J. (1984). Het tempo van de Franse Barokdansen
nary]. Paris: La venue Duchesse. [The tempo of the French Baroque dances]. In K. Vellekoop
S CHWANDT, E. (2013). L’Affilard, Michel. In Grove Music Online, (Ed.), Tempo in de Achttiende Eeuw [Tempo in the eighteenth
Oxford University Press. Retrieved August 1, 2013 http:// century] (pp. 7-25). Utrecht: Stimu.
www.oxfordmusiconline.com/subscriber/article grove/music/ VAN D YCK , E., M OELANTS , D., D EMEY, M., D EWEPPE , A.,
15813 C OUSSEMENT, P., & L EMAN , M. (2013). The impact of the bass
S HAFFER , L. H. (1981). Performances of Chopin, Bach, and drum on human dance movement. Music Perception, 30,
Bartok: Studies in motor programming. Cognitive Psychology, 349-359.
13, 326-376. VAN D YCK , E., M AES , P. J., HARGREAVES , J., L ESAFFRE , M., &
S PITZER , J., & Z ASLAW, N. (2005). The birth of the orchestra: L EMAN , M. (2013). Expressing induced emotions through free
History of an institution, 1650-1815. New York: Oxford dance movement. Journal of Nonverbal Behavior, 37, 175-190.
University Press. V EILHAN , J.-C. (1979). The rules of musical interpretation in the
T HAUT, M. H., & A BIRU, M. (2010). Rhythmic auditory stimu- Baroque era. (J. Lambert, Trans.). Paris: Alphonse Leduc.
lation in rehabilitation of movement disorders: A review of (Original work published 1977)
current research. Music Perception, 27, 263-269. W ETZER , J. (1988). Dansen met je oren [Dancing with your ears]
T HORP, J. (2012). New light on Kellom Tomlinson: Two articles by (Unpublished doctoral dissertation). Utrecht University.
Jennifer Thorp. Dance Research, 30, 57-79. Z ENTNER , M., & E EROLA , T. (2010). Rhythmic engagement with
T ODD, N. (1985). A model of expressive timing in tonal music. music in infancy. Proceedings of the National Academy of
Music Perception, 3, 33-58. Sciences, 107, 5768-5773.

154.59.125.36 on Tue, 02 Jan 2024 10:38:03 +00:00

Tempo in Baroque Music and DanceAuthor(s) : Esther Coorevits and Dirk Moelants

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tempo in Baroque Music and DanceAuthor(s) : Esther Coorevits and Dirk Moelants

Uploaded by

Copyright:

Available Formats

Tempo in Baroque Music and DanceAuthor(s): Esther Coorevits and Dirk Moelants

Stable URL: https://www.jstor.org/stable/10.2307/26417332

This content downloaded from