Table of Contents
Biomechanical Riding & Dressage: A Rider's Atlas

( a "modest proposal" under construction)

I offer here some structural explanation of the "rollkur-like" training technique the FEI now calls "hyperflexion."

It is the proposition here that LIGHTNESS includes minimal aids (or the slightest indications from its rider if the horse prefers) and the appropriate response to them. These are sophisticated, subtle, tactful relationships in the horse/rider system. Lightness begins in the sensibilities of the rider. If it were easy to achieve, you could buy 'essence of lightness' in a bottle at the store.

Lightness is
a) based on attention and confidence in both horse and rider,
b) achieved by relaxed use of core muscles to support the column of the body and to provide the "elastic pillar" for half halts,
c) allowed by understanding that the horse needs to breathe, thus aids should not ask it to maintain positions that interfere with this vital function (includes not only severe positioning of the head but tactless use of the legs on the rib cage housing of heart and lungs).

First, you will probably want to become familiar with the report below.

The Use of Over Bending (“Rollkur”) in FEI Competition
31 January 2006 during the FEI Veterinary
Committee meeting at the Olympic Museum,

Second, here is the language from the report that proposes substitution of the term "hyperflexion" for rollkur. This is an excerpt from a longer discussion.

The Workshop considered the term "hyperflexion" and the following working definition was agreed:
1) Hyperflexion of the neck is a technique of working/training to provide a degree of longitudinal flexion of the mid-region of the neck that cannot be self-maintained by the horse for a prolonged time without welfare implications.”
2) There must be an understanding that hyperflexion as a training aid must be used correctly, as the technique can be an abuse when attempted by an inexperienced/unskilled rider/trainer."

Third, the technique being discussed includes not only over bending (lateral or right to left motions of the horse's head/neck) but hyperflexion (downward motion of the horse's head/neck - upward motions would be called hyperextension).

1 - Horses (and other cursorial or running mammals) are connected so that the hindquarters can effectively transmit impulse forward.
2 - Horses have been shown to generate more impulse with hind legs than front legs, although front legs do participate in impulsion. In walk, the limbs behave as inverted pendulums, with the mass vaulting over a pivot point. In the ballistic gaits of trot and canter, the limbs are modeled as inverted spring-loaded pendulums.
3 - Hind legs are larger and more muscular than forelegs.
4 - The materials of the limbs (bones and their links of fascia, ligaments, muscles, cartilage) are affected by strains. The character and duration of the strain are significant in developing biological and mental responses.

1 - A system of riding that works from hindquarters --->> toward the bridle is logically and biomechanically sound.
2 - A system of riding that works from forequarters toward the hindquarters contradicts the structures of the horse.
3 - A system of riding that maintains strain beyond what a horse will impose on itself (FEI words in bold italics) will engender risk of injury rather than develop gymnastic capacity or willing response to aids.

When measurements are made, investigators try to explain the overall character of a gait such as walk, trot or canter by modeling (or attempting to model) contributing subfactors (see Table One below based on Dickinson, et al. and Cavagna, et al.). This is more easily said than done: isolating “just” the contribution of muscles and tendons to energetics of a gait may require invasive technology and the measuring probes or protocol may affect the outcome. Muscles behave in a number of ways. They may lengthen or shorten in the sense we usually talk about, but they may also exert considerable force when behaving as struts where the change in length is minimal. Thus not all questions we would like to have answered unambiguously have been settled to the satisfaction of most people.

Arguments about the "welfare of the horse" are then inserted into the normal uncertainty that accompanies scientific investigation. It is at least reasonable to avoid techniques that are likely to invoke unwilling reactions from horses, especially if there are alternatives. The alternative to "hyperflexion" or rollkur is the classical technique of forward, down, out (FDO). Rather than using the rein to adjust the horse's posture (riding from front to back), it is adjusted from the hindquarters forward by the seat and legs, with the result received in the rein. The mass of the head and neck then produces the stretching of the whole horse-system. The success of FDO is built on its concordance with equine architecture.

Ligaments and fascia exhibit a property during dynamic or passive stretching called “hysteresis.” Range of motion is deeply involved with the amount of stretching a system such as a leg may undergo without damage. The actual result achieved is affected by the history of a body part action: that is, what it has been doing affects what it does at the present moment. That is why dressage gaits require predictability (regularity, straightness which simultaneously affect behavior of the center of mass).

It is precisely the hysteresis response (passive stretching) under the conditions of hyperflexion or rollkur that is suspect here. If a horse is kept under these conditions longer than it is willing to do so, then there must be a felt reaction in the nervous system to the extreme positioning of all connective tissues. It is beyond this discussion whether or not this is a pain response, but the unwillingness of a horse to maintain a position may be taken as a warning that limits are being pushed, with potential negative results on the wellfare of the animal.

The characterization of pain, especially 'proof positive' that it exists in a specific circumstance, is difficult and beyond the scope of this web site. That caveat is an acknowlegement of its importance as an issue in training horses or humans. For a formal discussion of the many subtle aspects of pain, its creation and amelioration, I recommend The Body in Pain: The Making and Unmaking of the World by Elaine Scarry. It is phenomenological in its approach and densely written, but reading it brings rewards of understanding.

Further, the range of motion combined with tempo in dressage gaits is crucial in its impact on bony and elastic structures. While it appears possible to produce exaggeratedly slow tempo with hyperflexion combined with the impression cadence by means of "suspension," examination of video sequences of horses in rollkur at 45 strides per minute shows marked positive disassociation of diagonal pairs that achieves the slow tempo in extended, collected trots and passage. The separation of hoof beats by longer intervals occurs instead of more suspension, which would require the horse to be higher in the air in order to achieve slower tempo. If diagonal pair timing is considered, the time interval separating hoof beats (or toe down) can occur in three ways:
- the hind leg of the pair lands first (positive disassociation),
- the two legs land simultaneously, and
- the front leg of the pair lands first (negative disassociation).

Each timing combination has consequences for the way forces act on the structure of the horse. Here we will only consider positive disassociation, a feature often touted as desirable in trot and canter. In my videos and in other videos I have examined, most often observed positive disassociation centers around 0.03 second (obviously there is some variation), even at fairly slow tempos of 53 strides per minute.

The horse in tempos significantly slower than 51 strides per minute is landing on one leg at a time for nearly 0.09 second (as opposed to 0.03 sec), placing the strain of its mass + rider on each leg for a protracted time (force measured in newtons or 1 N = 1 kilogram x 1 meter x sec-2), rather than having forces distributed between diagonal pairs that are nearly simultaneous.

Suspension in trot (16-37 milliseconds) or canter (54-87 milliseconds) reflects limits placed on horses by physics (Clayton, H. M. 1997. Classification of collected trot, passage and piaffe using stance phase temporal variables. Equine Veterinary Journal Supplement 23, 54-57; Clayton, H. M. 1995. Comparison of the stride kinematics of the collected, medium, and extended walks in horses. American Journal of Veterinary Research 56, 849-852; Clayton, H. M. 1994a. Comparison of the stride kinematics of the collected, working, medium, and extended trot. Equine Veterinary Journal 26, 230-234; Clayton, H. M. 1994b. Comparison of the collected, working, medium, and extended canters. Equine Veterinary Journal Supplement 17, 16-192) Once in suspension, they return to earth with the acceleration of gravity. This corresponds to 32.174048 feet per second per second (ft/sec2) in the English system of measurement. Accleration of gravity does not change, vertical distance in suspension shows little variation and video analysis shows that limbs are loaded singly during the very slow tempos. A horse’s limbs support its body weight so the average limb vertical force during a limb duty cycle is equal to its body weight.

The tensioned riding that accompanies hyperflexion cannot operate outside the realm of physics and the properties of biologic structure. Thus not only is rollkur raising concern where tissues of head and neck are concerned, in its observed practice it raises concerns about the effects it has on the wellfare of limbs. It would appear that veterinary supervision of riding sessions incorporating hyperflexions would be wise.

A. To explain the importance of BENDING to LIGHTNESS, I offer the following animation of my simple computer model with some deep fascial connections (bands/envelopes of fibrous tissues connecting soft body parts) between hindquarters and forehand shown. Muscles influential in generating bend are those that work in concert with the deep fascia connecting impulsion from hindquarters to the trunk. The longissimus dorsi muscle is an example.

A trained horse working in collection, where maneuvering is tested, needs only indications from its rider, not interference. In addition, aids are not compulsions, as a horse may refuse (block) the requests from its rider. In that case, the situation is dangerous for horse and rider (see FEI caveat above). Injuries to fascia or disease of those tissues (for example, fibromyalgia) are painful.

The animation compares two frames and shows lateral bend at the moments of walk "vee" where there is maximum right to left shift of the rib cage. Lateral bending in the thoracic spine is accompanied by some axial rotation because of vertebral shape. I have connected some of the deep fascia (crucial connective tissue for vertebrates) to show how THE HORSE IS CONSTRUCTED TO BE DRIVEN FORWARD FROM THE HIND LEGS. Why else might the hind legs be larger and more muscular than the forelegs? The structure of the forehand is stabilizing and the structure of the hindquarters emphasizes thrust. The whole running (cursorial) animal is integrated for dynamic balance between thrust and stability. It is one object of dressage to train this natural capacity for agile movement. The need for lightness is more than aesthetic, it is a requirement of mobility in a gravitational field with friction (footing). Without lightness, subtle aids are unable to inform the horse of what is requested, and the rider's mass is busy wiggling about, disturbing the balance of the conjoined centers of mass.

TABLE ONE: Behavior of limbs in the context of their connections to trunk and neck. Fascia are not muscles, but are sheets/bands of fibrous connective tissue enveloping, separating, or binding together muscles, organs, and other soft structures of the body. They are critical architectural components of vertebrates. Fascial injuries or diseases are painful (fibromyalgia sufferers will attest to that, as well as those with plantar fasciitis).
General diagram of important deep fascial connections in forehand and hindquarters. It is this deep fascia, combined with muscles, tendons and ligaments that are arranged to transmit motion of legs in the vertical plane to movement of the trunk (for example, lateral bending) in the horizontal plane. Fascia are sheets or bands of fibrous connective tissue enveloping, separating, or binding together muscles, organs, and other soft structures of the body: bones have fascial envelopes that allow links to connective tissues. Connective tissues form a continuum of structures when examined at the cellular and molecular levels. Two frame animation comparing deep fascia during right and left "vee" of walk, where bend is maximum (accompanied by some axial rotation). Even at this small size, the model can resolve lateral motion motion in the rib cage. During gaits, the rib cage changes shape in three dimensions. This is enabled by the cartilaginous portions (yellow) of the ribs and sternum in conjuction with the joint at the top of each rib. These joints allow motion during breathing plus during movements of the legs.
Model of a leg in WALK behaving as an inverted pendulum. Forces generated in contact with the ground reflect combined masses of horse and rider. Tempo affects how strains are distributed, depending on gait and timing of one, two or three legs coming into contact with footing. After: Dickinson, M. H., Farley, C. T., Full, R. J., Koehl, M. A. R., Kram, R., and Lehman, S. 2000. How Animals Move: An Integrative View. Science 288: 100-106. Model of a leg in a ballistic gait (trot, canter, gallop, pronk). After: Cavagna, G. A., Thys, H., and Zamboni, A. 1976. The sources of external work in level walking and running. The Journal of Physiology: 262, Issue 3 639-657.

For the unbelievers of the way thrust from hindquarters operates, there is this graph --- click on it to see it illustrated with a canter stride. (Graph modified from: Wilson, A., et al., 2000. Optimization of the Muscle-tendon Unit for Economical Locomotion in Cursorial Animals. In: Skeletal Muscle Mechanics (W. Herzog, editor). pp. 517-547, J. Wiley & Sons. Wilson and colleagues redrew the data from Niki et al., 1984. A Force Plate Study in Equine Biomechanics. 3. The vertical and fore-aft components of floor reaction forces and motion of equine limbs at canter. Bulletin of the Equine Research Institute 21, pp. 8-18.).

B. Here is the computer model in rollkur or hyperflexion, based on overlays of video frames of Olympic/National Champioship riders in warm up. (under construction)


C. If the computer model indications are reasonably insightful, hyperflexion blocks forward impulse, precisely because it concentrates on the forehand. Short steps behind are typical on video recordings for this technique of highly tensioned riding. It might be that efforts to control horses in this way produce consistent test results, but there have been incidents when horses worked in rollkur refused to go forward in the competition arena. However, this page only looks at the deep fascia. For details of the internal structures of the head during various postures, this is the link.

Aids used in rollkur/hyperflexion produce a horse that is possibly off the contact or evasive of bit connection (see the cover photo of the FEI Report). Hyperflexed horses are often seen working with the curb bit parallel with the ground on a tight rein (although the rein is slack in the FEI illustration). In stills extracted from the videos, many of these horses can be seen to have opened the mouth to some degree, to have withdrawn the tongue from the bit and to be leaning on the bit(s) with the lower molar. In dressage, the bit is a communication device for the tongue: withdrawing it lessens or avoids a continuous connection in the "circle of aids." A contact where there are faults with the tongue, the jaw is held shut and the lower molar is against the bit, is not light, but is called "boring on the rein." As Nuño Oliveira says in Classical Principles of the Art of Training Horses:

“I wish to convey to the riders who are interested in dressage the idea of the pursuit of LIGHTNESS, which has always been the major preoccupation of all the great Écuyers who have enriched the art with their experiences and teachings. Each of them have left written works in which they emphasize the concern for lightness and the delicate use of the aids.

Unfortunately, at this present time [1983], this idea is a little lost. If some riders think about this subject and try to have light horses, I will be greatly rewarded. For I have been able to make these noble horses work as happier animals. I have also wished to give to those who are beginning to ride their horses an idea of progression. Of course, there are special cases where it is necessary to have a very great knowledge in order to be able to surmount the problems. Observation of the horse, the reflection and reading, not only of that which is written today but also the works of the Old Masters, and practice, can give you all those understandings.