The High Leg of Eighteenth-Century Ballet

The High Leg of Eighteenth-Century Ballet

Dance Magazine (7 Jan. 2019) recently ran an article called “The Story of How Ballet Legs Got Higher, and Higher, and Higher” (click here to read). The view presented there is informed by a kind of teleological evolutionism, such that the range of leg movement in the course of ballet’s history is seen as simply getting bigger and bigger. This is a completely mistaken notion: a careful examination of relevant primary sources reveals a much more complicated development. This blog-post then, which is based on my scholarly study in progress entitled The Technique of Eighteenth-Century Ballet, is intended to debunk the myth and show some of the wayward changes in fashion, by focusing mainly on theory and practice in the eighteenth century.


Ballroom Versus Theatrical Dance

It is incontrovertible that eighteenth-century ballet employed a number of different leg heights, from the very low to the very high. This fact, however, has not been widely embraced in both scholarly and popular presentations wherein older ill-informed views have been uncritically accepted. Part of the problem here has been the unfounded assumption that the dance of the ballroom and that of the theater were more or less the same as far as technique goes.

While many movements were shared by both kinds of dance, it is quite clear, however, from both textual and pictorial sources that the execution of movements in the theater differed from that in the ballroom, or as Magri (1779: 1/137) puts it in his handbook for amateurs, “generally speaking, all these steps, being theatrical, are performed on stage differently, not as we have minutely shown.” One of the major ways in which the two differed was in the range of movement: that is, movements on stage were exaggerated, while those in the ballroom were contained. Mattheson (1739: 37), for example, writes broadly that

the art of gesture is as indispensable to the art of dance as the feet themselves. A composer who is poor at judging dances, whether they belong, say, to the choric [i.e., social] or hyporchematic [i.e., theatrical] styles – the difference lying more in the positions than in the steps, turns, or springs – will not do well at all here, for from his notes must stem comic or serious gestures.

Other sources clarify that by “positions” here was meant in part an expansive execution, with arms and legs held higher on stage than in the ballroom, and jumps reaching greater heights than on the ballroom floor. Rameau (1725: 70) writes in his ballroom handbook that “as I have undertaken, however, only to give instructions on how to do the different steps belonging to ballroom dancing [danses de ville], I am obliged not to go into these [theatrical] steps, which are executed in a grander way.” That danses de ville meant ‘ballroom dance’ is made clear by Essex’s translation of the phrase (1728: 40) as “Ball Dancing,” and by Ratier’s definition (1759?: 38) “the dance of the salon, or la danse de ville.

The professional dancer and choreographer John Weaver (1712: 162-63) also notes specifically that dance steps on stage differed in the performance, so much so that a theatrical execution would have been “rough and ridiculous” in a ballroom, that is, too exaggerated for so small a space:

SERIOUS Dancing, differs from the Common-Dancing [i.e., ballroom dance] usually taught in Schools, as History Painting differs from Limning. For as the Common-Dancing has a peculiar Softness, which would hardly be perceiveable on the Stage; so Stage-Dancing would have a rough and ridiculous Air in a Room, when on the Stage it would appear soft, tender and delightful. And altho’ the Steps of both are generally the same, yet they differ in the Performance: Notwithstanding there are some Steps peculiarly adapted to this Sort of Dancing, viz. Capers [i.e., cabrioles], and Cross-Capers [i.e., entrechats] of all kinds; Pirou[e]ttes, Batteries [i.e., battements], and indeed almost all Steps from the Ground [i.e., jumps].

The Spectator (25 Aug. 1712) likewise touches upon the overdrawn movements cultivated by theatrical dancers:

the Dancing [i.e., the dancers] on our Stages are very faulty in this Kind; and what they mean by writhing themselves into such Postures, as it would be a Pain for any of the Spectators to stand in, and yet to hope to please those Spectators, is unintelligible.

Indeed, dancing in too contained a manner on stage was in bad taste. As the famed choreographer Jean-Georges Noverre puts it (1760: 344), “confined steps, slight or ‘shrunken’ movements, in short, an execution that is too small, this is equally offensive to good taste.” The professional dancer and choreographer Giovanni Gallini (1762: 146) also speaks against mincing – and flaccid – dance movement generally, for the true art of dance “equally reprobates an ungainly rusticity, and a mincing, tripping, over-soft manner.”

The difference in the manner of executing movements on the ballroom floor versus the stage, as outlined above, continued into the following period. Carlo Blasis (1831: 489, 492-93) notes that “the private dancing [i.e., ballroom dancing] derives its origin from the theatrical dancing” and that

as to the movements of the body [in the former], they are nearly the same as those practiced by stage dancers, with this difference only, that they should not be carried to that grandeur and elevation, should have less impulse, and be modified, and adapted to the circles of Private Dancing. The legs ought to be raised from the ground but very little above the method of the second position; however, gentlemen may raise them something higher: the peculiar style of their dancing being more powerful and unrestrained, will admit of more elevated steps.

The professional dancer Théleur (1831: 100-101) in like manner comments on the difference in range of movement between stage-dancers and social dancers, noting that ballroom dance was the simpler part of theatrical dance executed “in miniature”:

Room-dancing is nothing more or less than that which is used for the stage, but executed in a more quiet style, avoiding all extravagances, or large steps; consequently, the practice to acquire this part of the art is considerably more easy than that of the public-dancer; it consists of the bending in the first position [i.e., fifth position], the little battemens from the first ground station to the fifth half aerial [i.e., from fifth to second position off the floor], the little battemens to, and from, the third half aerial station [i.e., battements sur le cou-de-pied], the ronds de jambe on the ground, the little ronds de jambe, the changes of the feet, the assemblées, the temps levés, the chassés, &c. selecting the more simple part of the practice laid down for the theatrical dancer, employing all the grace, &c. but in miniature.

For further discussion, see Fairfax (2003: 15-79).


References to a High Leg

The two heights of leg normally used in ballroom dancing were those wherein the toe of the gesture foot was raised to the heights of the ankle or the calf of the supporting leg. These are either explicitly prescribed in the ballroom handbooks or shown in surviving illustrations of ballroom dancers (fig. 1).

Figure 1. Ballroom dancers, 1735.

As was made clear in the foregoing section, the range of movement on stage was normally grander than in the ballroom, and this then implies that heights of leg above those of the ankle and calf were common in the theater. And, indeed, the heights wherein the toe of the gesture leg was raised until roughly level with the hip or above – to the height of the shoulder or even above the head – are alluded to or actually depicted in surviving illustrations of theatrical dancers. In his chapter on “serious dance” (and not the comic and grotesque), Bonin (1712: 169), for example, writes that “cabrioles can also be cut out in front, wherein the feet must come to lie almost level with or opposite the hips.” The same height is mentioned by Taubert (1717: 728) in the same context; cf. the height shown in figure 2.

Figure 2. Auguste Vestris, London 1781, an engraving after a design by Nathaniel Dance.

In the écart (i.e., the splits in the air), Ferriol (1745: 1/127) has the dancer “jump and while in the air fully open the legs as much as possible” (fig. 3). Pauli (1756: 20) sketches a grand rond de jambe thus: “the ouverture de jambe is performed when the leg does a rond in the air at the height of the thigh.” In the gargouillade, Magri (1779: 1/124) has the performer execute the circular movements with “the legs and thighs in an even line parallel to the floor.”

Figure 3. A caricature of a comic dancer apparently doing the splits in the air, circa 1780s, George Dance.

Compare as well the height of leg depicted in figure 4, which arguably shows poses, hence the reduced turnout, the bent legs and unpointed foot, and the more fanciful arrangements of the arms.

Figure 4. Dancing dwarfs: left 1716; right 1720s.


It also needs to be stressed here that raising the foot of the gesture leg to the height of the hip was not restricted merely to comic and grotesque roles, even in the early eighteenth century. Both textual and iconographic evidence indicates that a high carriage of both the leg and arms was also characteristic of lofty roles. In the passage cited above from Weaver (1712: 162), it is stated generally that “SERIOUS Dancing, differs from the Common-Dancing [i.e., ballroom dance] usually taught in Schools, as History Painting differs from Limning,” and that in contrast to ballroom dancing,  it “would have a rough and ridiculous Air in a Room, when on the Stage it would appear soft, tender and delightful. And altho’ the Steps of both are generally the same, yet they differ in the Performance.” It is apparent from Taubert (1717: 559), moreover, that in “serious” dance, second position of the arms was formed at the height of the shoulders, as in classical ballet:

Just as in this other branch of dance, namely serious dance, all the steps, even those which are borrowed from la belle danse [i.e., ballroom dance], are done high with considerable force (not to mention here the sundry capers, or powerful springs into the air wherein the feet must work away in the air for most of the time), so the arms here as well, which should necessarily accord with the legs, must always be held high, to the sides in a straight line with the shoulders [fig. 5].

Figure 5. A detail from Hogarth’s The Analysis of Beauty (1753), showing a dancer with his arms in second position.

Since not “all the steps” used in ballet were springing steps, “high” in the context of non-springing steps can mean only ‘with a high leg.’ Indeed, see figure 6, which shows a posed dancer in the role of a “Roman” (by its very nature, highly unlikely to be a comic or grotesque character), with the gesture leg raised behind to the height of the hip.

Figure 6. Left: a posed dancer in the role of a “Roman” (Lambranzi 1716: 2/21); right: the same pose reconstructed to show better the lines of the body.


References to an Overhigh Leg

References to an overhigh height – above the hip, to the height of the shoulder or above the head – are found in Magri, for example, who had been a professional dancer in the so-called grotesque style from the late 1750s to the early 1770s. Consider his description (1779: 1/40-41) of the grand battement (the bolding is mine in the following quotations):

Figure 7. A reconstruction of Magri’s grand battement.

There is another kind of battement, which is said to be a high battement dégagé. Placed as usual in fifth, raise the right leg, for example, keeping the knee well stretched and turned out and the cou-de-pied arched, disengage the leg at least to the height of the shoulder, then take it down behind the left but see to it that it does not go beyond the foot on the floor; otherwise the body will be visibly contorted, and raising it again to the height of the first, take it back down in front, this second one done heeding the same warning. As many of these as possible are repeated thus. Not going beyond the foot on the floor will be difficult because of the force of the descent; this must be done with care, curbing this force of the descent, which is easy to achieve if done with caution. There are two movements here, the first lifting the leg and the second lowering it. These are also done as quickly as possible in order to acquire the desired benefit and fine dégagé of the thigh.
I have proven myself with these battements and have gone higher than the head. Indeed, I held up my left hand so that it was raised perpendicular, and I touched the palm of my left hand with the right foot, or rather cou-de-pied, a clear indication of having disengaged the leg well [fig. 7]. Take care, however, not to practice these battements violently; do them after you have made the sinews soft and flexible, and do not be heedless of keeping the foot on the floor well supported. In practicing these in a heat once, my beating foot displaced the other on the floor, and falling flat on my face, I broke my nose. With the same carelessness, Cesarini had the ill-luck to break an arm.

In one version of the grande révoltade, Magri (1779: 1/126) notes of the gesture leg, at the height of the jump, that “its thigh [is] taken up so high that the knee passes close to the face with the leg pointing upwards and the foot going above the head.” And most impressive of all, both feet are taken above the head in the grotesque jump the spazzacampagna (lit. ‘blunderbuss’), reconstructed in figure 8, which is based on the following two descriptions and other relevant material not presented here:

The caper known as the spazzacampagna begins in fifth position. Bend both knees, and in rising into the air, draw up the legs under the body as much as possible without taking the feet apart from this fifth. Then with them both coupled together in fifth, stretch them forwards to just above the head, which is drawn down a little to hide behind the feet. From here both legs are stretched out to the sides, opening and stretching as much as possible; raise the head and straighten the body at the same time. As you start to descend, begin to bring the feet together until they are in fifth again in coming down, but with the foot in front that was behind at the beginning. To do this caper, you need to precede it with a grand brisé or a grande [demi-]sissonne, which will be able to give force to the jump. Usually the fifth positions in which it begins and ends are forced, for when greater force is needed for height, an exact position is not to be looked for but rather that which will give greater force to the jump. (Magri 1779: 1/127)

The Spacciato Campania [sic]. This was always the most outstanding tour de force of the grotesque dancer. A tempo [i.e. preparation] for it with a few steps in a running start and a forward spring is needed. With both feet on the toes in first position, spring up at the very same time and do a high jump. At the same time as the spring into the air, the right foot does a tour de jambe to the right from front to back; the left does a tour de jambe to the left from front to back. The right arm does a circular movement to the right from front to back, and the left arm does a circular movement to the left from front to back. Land again in first position. Not a little strength and practice are needed here in order to do these four circular movements with the arms and legs very high at the same time. When highest in the air, the body looks like a frog tossed up into the air with all fours stretched away, which may well be the source of the name. (Roller 1843: 221-22)

Fig. 8. A reconstruction of the spazzacampagna (land not shown).

Clearly, raising the gesture foot to the height of the shoulder or higher was already established by the beginning of the eighteenth century and continued into the early nineteenth century, as is apparent from figure 9.

Figure 9. Left: “Monsieur Dubreil dancing the role of Scaramouche,” before 1713; right: the comic dancer Charles Masurier dancing with his mistress in Blache’s ballet Les meuniers, 1824.

As mentioned above, Magri was a dancer in the so-called grotesque style, a genre which by definition was to be as contorted and exaggerated as possible (Fairfax 2003: 124-61), and these overhigh heights almost certainly were proper to this style in theory. (Eighteenth-century ballet was by convention divided into four distinct styles – and thus into four distinct techniques – commonly called the serious, half-serious, comic, and grotesque. For further details, see Fairfax (2003: 81-188).)


Excursus on the Upper Body

Figure 10. Parisot in a pose, London, 1790s.

I might mention here only very fleetingly that there appears to have been a stylistic difference in the way the upper body was managed when a high straight leg extension to the rear was executed. In the high styles (the serious and half-serious), the dancer seems to have regularly performed here a forward upper-body inclination, so that the whole body came to be horizontal to the floor when the gesture leg was highest (figs. 10-11). Cf. Bonin’s remark (1712: 169) in connection with the cabriole en arrière as described in his chapter on the high styles:

If I wish to do this cabriole to the back, however, the body and the feet must come to lie in a horizontal line [fig. 11], but I have even seen it done with the heels visible above the horizontal line so that the head and feet formed a transverse line.

Figure 11. A reconstruction of the opening movements in a cabriole to the rear in the half-serious style according to Bonin (1712: 169).

In the low styles (comic and grotesque), however, the torso seems to have been regularly held as upright as possible to produce a more contorted line in the body (fig. 12). Further evidence relevant here will be outlined in my forthcoming study.

Figure 12. Left: posed dancer in the grotesque role of Scaramouche, 1716; right: grotteschi, 1823.


The Dupré Precedent

In the 1730s, Louis Dupré, a famous dancer in the so-called serious style (a slow graceful terre-à-terre style), set a precedent for plentiful use of overhigh leg extensions in this genre, and he was widely imitated by other serious dancers. Dupre’s elasticity is alluded to in a number of sources, such as Casanova (1961: 2/141), who saw Dupré perform at the Paris Opéra in 1750, at the end of his career: “Truly it was an elastic body, which became bigger as it unfolded.” Noverre (1760: 342-44), who had been a student of Dupré, clearly indicates that many imitated these lofty extensions (the bolding is mine in the quotation):

The déploiements [i.e., unfoldings] of the leg and the temps ouverts [i.e., movements into an open position] were doubtless suitable to Monsieur Dupré; the elegance of his figure and the length of his limbs went wonderfully together with the temps développés and daring steps of his dancing, but what suited him will not suit dancers of middling height, yet everyone wanted to ape him. The shortest legs were forced to run through the same spaces and describe the same circles as those of this celebrated dancer, thus, the loss of stability: The hips were never in their place, the body wavered continuously, the execution was ridiculous, I thought I saw Thersites imitating Achilles. . . .
This fault [of disproportionate movement], Monsieur, is very fashionable among serious dancers, and as this style holds greater sway in Paris than everywhere else, it is very common there to see the dwarfish dance with movements of gigantic and ridiculous proportions. I would even go so far as to say that those who are gifted with a majestic figure sometimes misuse the extent that their limbs can reach and the ease with which they cover the stage and make their movements stand out. These exaggerated déploiements alter the noble and peaceful character that la belle danse [i.e., the serious style in this context] should have and deprives the execution of its softness and gentleness.

Figure 13. Costume design by Boquet for the serious dancer Gaétan Vestris, circa 1760s.

Elsewhere, Noverre (1760: 183) again implies that these high leg extensions were to a height well above the waist. In speaking (ill) of the tonnelet, which was a kind of male tutu worn by especially serious dancers throughout much of the eighteenth century and which in this case was not uncommonly made with pronounced lateral projections (fig. 13), he writes that “I would do away with those stiff tonnelets, which in certain dance positions, bring the hip to the shoulder, as it were, and which obscure [the body’s] contours.” Carrying the gesture foot markedly above the height of the waist would naturally cause one edge of the tonnelet to rise with the leg, thus bringing “the hip” (i.e.., the outer edge of the tonnelet) “to the shoulder.”

And so, the not infrequent use of overhigh extensions, which almost certainly were originally proper to the grotesque style, spread to the serious style in practice. The latter style in theory, however, was to cultivate only graceful movements and positions and avoid any contortion and caricature which might “alter the noble and peaceful character” requisite here.


The Vestris Precedent

Figure 14. Auguste Vestris c.1781, Thomas Gainsborough and Gainsborough Dupont.

The course of ballet history was radically changed by Auguste Vestris (1760-1842), a superstar dancer (fig. 14) who developed a highly idiosyncratic style of dance wherein elements from the four traditional styles were blended into a single composite style, during the heady days of the French Revolution. His style was marked by, among other things, much use of exaggeration, such that “his leg would rise to the height of his head” as in the grotesque style (Berchoux 1808: 20), a penchant caricatured in figure 15. (For a discussion of the freedom granted dancers to cultivate a unique personal style, see Fairfax (2003: 243-55).)

Figure 15. Two caricatures of Auguste Vestris circa 1800: left by Isabey; right by George Dance.

His tremendous popularity induced a widespread imitation of his personal style and ultimately resulted in the demise of the four traditional styles and the emergence of a single composite style in the early nineteenth century. As Noverre remarks (1807: 2/127-28), “everyone became imperfect and unfaithful copiers,” such that he could ask,

what has come of this unreasonable and capricious aping? The dancing at the Opéra is now of the same color, the same style, the same genre. There is only one manner of execution. This art has driven out variety in order to adopt the most unbearable monotony.

As part of this widespread imitation and “unbearable monotony,” limbs now were not infrequently raised overhigh with all characters, soloists and corps, men and women alike. That is to say, the exaggeration that had originally been proper only to the grotesque style, but then became common in the serious style, was now found everywhere. Noverre (1807: 2/167) complains of typically having “to see sixty arms raised well above the head and thirty straight legs carried in one spontaneous movement to the height of the shoulder.” In like manner, Roller (1843: 19), who had been a professional dancer in the 1790s, speaks broadly of “the loathsome contortions and grotesque movements to which ballet has now degenerated, little by little since the French Revolution,” and which are lampooned in figure 16.

Figure 16. A caricature of a pas de deux in Giselle, entitled Grise-Aile (‘Tipsy-Wing’), from Le musée Philipon, 1841.

Compare also Costa’s description of a grand rond de jambe (1830: 29-30), with my bolding:

From second position, the left foot sliding, and bending, comes into fifth in front, and raising the heel, the foot comes to be in front of the ankle of the right leg. Straightening the knees, raise the left foot horizontally almost to the height of the chest, also rising on the toe of the foot that is on the floor, and in this way off the floor, the said left leg is carried sideways to the height of the shoulder, which are a forced fourth and second, as with dégagés. Lower the leg little by little to second, and in setting down the heel of the right foot, which supports the body, the left repeats the unfolding, that is, the same movement.

A fuller discussion of these changes can be found in Fairfax (2003: 275-91).

By the beginning of the twentieth century, the theoretical norm of hip-height seems to have been favored, at least in the Russian school. While Bernay (1890: 199) still gives an overhigh height for a grand battement – “the said left foot then leaves fifth and rises to the height of the left temple, still forming a continuous straight line with the leg” – Alexandra Danilova (1903-1997), however, notes in her memoirs (1986: 40) that

at that time, during my years at the school, we didn’t lift the legs high – it was considered not classical, rather daring, a little bit vulgar. “You are not in the circus,” our teachers would scold if développés or grands battements got too big. Just a teeny bit above the waist was as high as we were allowed. The Victorian attitudes still prevailed.

Contrast Danilova’s statement with that of the grottesco Magri from the eighteenth century, who prescribes that the foot of the gesture leg in a grand battement be raised “at least to the height of the shoulder.” Or contrast it with that of Noverre, who writes that serious dancers after Dupré commonly performed “with movements of gigantic and ridiculous proportions.” Danilova attributes the popularization of overhigh leg extensions in the twentieth century to the influence of George Balanchine, who “wanted the legs higher” – higher once again, it could be said, in light of the foregoing.



The remarks given above concerning eighteenth-century practice refer specifically to male ballet technique. The history of women’s technique for this time frame is somewhat different and more nuanced, but I have not dealt with it here, as the purpose of this already lengthy blog-post is to debunk the myth that very high extensions of the leg were foreign to early ballet. (For a partial treatment of changes to women’s technique, see Fairfax (2003: 219-42).)




Berchoux, J[oseph]. 1808. La danse, ou la guerre des dieux de l’Opéra. Second ed. Paris: chez Giguet et Michaud.
Bernay, Berthe. 1890. La danse au théâtre. Paris: E. Dentu.
Blasis, Carlo. 1831. The Art of Dancing. Translation by R. Barton. London: E. Bull.
Bonin, Louis. 1712. Die neueste Art zur galanten und theatralischen Tantz-Kunst. Frankfurt and Leipzig: Joh. Christoff Lochner.
Casanova de Seingalt, Jacques. 1961. Histoire de ma vie. 6 vols. Wiesbaden: F. A. Brockhaus.
Costa, Giacomo. 1831. Saggio analitico-pratico intorno all’arte della danza per uso di civile conversazione. Turin: Stamperia Mancio, Speirani eCompagnia.
Danilova, Alexandra. 1986. Choura, the Memoirs of Alexandra Danilova. New York: Alfred A. Knopf.
Essex, John. 1728. The Dancing-Master. London: J. Essex and J. Brotherton.
Fairfax, Edmund. 2003. The Styles of Eighteenth-Century Ballet. Lanham, Maryland, and Oxford: The Scarecrow Press, Inc.
Ferriol y Boxeraus, Bartholomè. 1745. Reglas utiles para los aficionados a danzar. Copoa: a costa de Joseph Testore.
Gallini, Giovanni-Andrea. 1762. A Treatise on the Art of Dancing. London: printed for the author.
Hogarth, William. 1753. The Analysis of Beauty. London: Printed by J. Reeves for the author.
Magri, Gennaro. 1779. Trattato teorico-prattico di ballo. Naples: Vicenzo Orsino.
Mattheson, Johann. 1722-1725. Critica musica. 2 vols. Hamburg.
Noverre, Jean-Georges. 1760. Lettres sur la danse, et sur les ballets. Stuttgart and Lyon: Aimé Delaroche.
Noverre, Jean-Georges. 1806-07. Lettres sur les arts imitateurs en général, et sur la danse en particulier. 2 vols. Paris: chez Léopold Collin.
Pauli, Charles. 1756. Élémens de la danse. Leipzig: Ulr. Chret. Saalbach.
Rameau, Pierre. 1725. Le maître à danser. Paris: chez Jean Villette.
Ratier, Joseph. 1759?. Observación 1 sobre el arte de la danza. Manuscript Mn Barbieri Mss 14059. Madrid: National Library.
Roller, Franz Anton. 1843. Systematisches Lehrbuch der bildenden Tanzkunst und körperlichen Ausbildung. Weimar: Bernh. Fr. Voigt.
Taubert, Gottfried. 1717. Rechtschaffener Tantzmeister. Leipzig: bey Friedrich Lanckischens Erben.
Théleur, E.A. 1831. Letters on Dancing, Reducing this Elegant and Healthful Exercise to Easy Scientific Principles. London: printed for the author.
Tomlinson, Kellom. 1735. The Art of Dancing. London: printed for the author.
Weaver, John. 1712. An Essay Towards an History of Dancing. London: printed for Jacob Tonson.