WHAT IS DANCE? · CHAPTER 04 OF 18 · A Body Begins
A dancer does not normally begin a turn by reciting the positions of every joint. The movement can be familiar, precise, and responsive without arriving as a paragraph of instructions. From outside, this may look like instinct. From inside, it may feel like knowing where the body is going. Both descriptions point toward something real, but neither explains how it happens.
The body is continually supplying information about its position, movement, contact, and orientation. The nervous system combines that information with expectations shaped by earlier experience and the action being attempted. Movement changes what is sensed; new sensory information helps shape what happens next. Dancing depends on this continuing relationship.
When we say that a body “knows,” we are using a convenient metaphor. Muscles do not secretly compose instructions in language, and there is no miniature dancer inside the head directing another dancer outside it. There are interacting processes across the nervous system and body. The remarkable thing is that their work can become available to us as an apparently simple experience: this is where I am, this is what I am doing, and this is what comes next.
Where is the hand when nobody is looking?
Vision gives us abundant information about the body, but it is not our only source. People ordinarily have a sense of where their limbs are without continuously looking at them. That capacity is part of proprioception: information and perception concerning bodily position and movement. It helps distinguish a moving hand from an unseen object whose location we must guess entirely from memory.
Signals associated with muscles, tendons, joints, and skin contribute to the wider account of the body’s state. Their roles differ, and no single receptor supplies a complete photograph. The nervous system must interpret changing information in relation to other information. A felt position is an achievement of this system, not a label physically attached to a joint.
A classic experiment by Guy Goodwin, David McCloskey, and Peter Matthews made this interpretation visible through error. Vibration applied to a muscle’s tendon altered participants’ judgments of the angle at the elbow. The reported position shifted in a direction consistent with the muscle having lengthened. Manipulating sensory input could therefore change where the limb seemed to be.[1]
The finding is not an invitation to reproduce the procedure. Its explanatory value is that a seemingly immediate bodily certainty depends on information that can be altered. We do not directly inspect our body from an infallible inner viewpoint. We perceive it through biological processes that usually work well enough to make ordinary action possible.
For a dancer, this means that the visible outline is only part of the movement’s reality. The dancer also has a changing sense of relationships among body parts. An arm may be experienced through its direction, effort, connection to the torso, and arrival in a familiar configuration. A viewer sees the resulting action but does not receive the same information.
This difference helps explain why watching a movement and being able to perform it are separate achievements. A clear image can show where a hand goes without supplying the complete bodily experience of getting it there. Understanding dance therefore requires attention to both the outward event and the mover’s access to it.
There is also a difference between knowing where a limb is and knowing where one wants it to go. A target can be perfectly clear while the present bodily arrangement is uncertain. Conversely, a person can feel the current arrangement without having selected the next action. Movement has to connect these kinds of information. The distinction is useful when someone assumes that a visible instruction should translate immediately into an accurate performance: the destination is only one part of the problem.
Several senses contribute to one situation
Dancing usually involves more information than any one sense can supply. Vision helps locate other people and features of the environment. Touch supplies information about contact. Proprioception contributes to the sense of the body’s configuration. The vestibular system contributes information relevant to head movement and orientation. Hearing may provide timing, cues, or the sound of the dancer’s own actions.
These sources do not simply take turns. Their information can be combined, compared, and used differently according to the task and conditions. A partner’s changing contact may be especially informative at one moment; a visible pathway may matter more at another. A fixed statement that dancers always depend on one particular sense would overlook this flexibility.
In an experiment about judging object height, Marc Ernst and Martin Banks found that participants’ combinations of visual and haptic information resembled a model that weighted cues according to their reliability. The study concerned a specific perception task, not dancing. It nevertheless provides a clear example of sensory information being integrated rather than accepted from a permanently dominant channel.[2]
Imagine an explicitly hypothetical performance in which the light changes. The body does not suddenly lose every source of information because one source becomes less useful. What information remains available, and how successfully it can be used, will depend on the person, the movement, and the situation. The example illustrates a question about sensory organization; it is not advice to remove visual information while practising.
The surrounding space also supplies information through its consistency. A familiar surface, a known distance, or a recognizable partner can make an event more predictable. Change those conditions and an apparently familiar sequence may demand renewed attention. What a dancer knows is partly a relationship with a particular environment, rather than a capacity sealed inside the body.
A change of viewpoint offers another ordinary complication. An action demonstrated by someone facing us must be related to our own body’s sides and directions. A reflection presents another arrangement again. The visible information may be clear while its relationship to an intended action still requires interpretation. In an imagined rehearsal, two people might agree perfectly about what they saw yet initially translate it into different directions. This would not require defective perception; it could arise from using different reference points.
This is why “the senses” should not be imagined as separate reporters sending independent newspaper articles to a central editor. Their contributions participate in a continuing estimate of the situation. The result is useful enough for action precisely because it can bring several incomplete sources into relation.
Action has to look ahead
Information takes time to travel and be processed. If every movement waited for a complete report of the previous movement’s consequences, coordination would face a serious problem. The world would keep changing while the system waited. One way of understanding movement control is that it combines incoming information with predictions about what an action is likely to produce.
A prediction here need not be a conscious guess. It can be a process that anticipates consequences on the basis of the current state, the action being generated, and earlier experience. The word “model” is often used for an internal relationship that performs this work. It does not imply that the brain contains a tiny diagram of the body or runs the same software as an engineer’s laptop.
Daniel Wolpert, Zoubin Ghahramani, and Michael Jordan examined this idea in a 1995 study in which participants estimated hand position after movements made in darkness and under imposed forces. The pattern of errors supported an account combining predicted movement consequences with sensory information. The experiment concerned hand movement, but it helped establish a framework for investigating sensorimotor integration.[3]
Consider a hypothetical dancer reaching a known position at the end of a phrase. The action can begin before the final position has been sensed, because its consequences are partly anticipated. Incoming information can then help adjust the estimate. Prediction and feedback are partners in this explanation. Treating either as sufficient on its own would miss why movement can be both prepared and responsive.
Prediction is especially useful when the body or surroundings have regularities. Familiar movements, repeated timing, and consistent contact can make consequences easier to anticipate. Unfamiliar circumstances can make an existing expectation less accurate. A system that only predicted and never updated would become increasingly detached from what was actually happening.
The important idea is therefore modest and powerful: action is not merely a reaction to the past. It is also organized around what is about to happen. Dancing makes this visible whenever a body seems already prepared for the next change before an observer has finished noticing the present one.
Feedback is information, not a scolding
In ordinary conversation, feedback often means another person telling us what went wrong. In movement, the word has a broader meaning. It concerns information about the body’s action and its consequences. Contact, sound, vision, and bodily sensation can all contribute. Feedback can confirm an expectation as well as reveal a difference.
Imagine a dancer expecting a movement to finish in one place and sensing that it has finished slightly elsewhere. The difference can matter for the next action. It need not produce a verbal thought or a visible interruption. Depending on the situation, the person may adjust within the continuing phrase, revise a later attempt, or deliberately incorporate the difference.
This example also shows why “error” needs a reference. Error relative to what? A planned position, a shared timing, a teacher’s instruction, an artistic intention, or a practical need to remain supported? The same deviation can matter greatly for one objective and hardly at all for another. Measurement becomes useful when the relevant objective is clear.
Not every correction should be imagined as a separate emergency response. In a living system, movement and information continually influence one another. The dancer’s apparent continuity may depend on many adjustments that do not appear as distinct events to a viewer. Smoothness can therefore be the result of ongoing regulation rather than the absence of correction.
There is a useful contrast with a printed route on paper. The route remains the same whether the road is clear or blocked. A moving person can use information from the present situation. Choreography can be stable while its performance remains responsive. The plan gives structure; sensing and action make an actual event under actual conditions.
We should also allow for deliberate variation. A dancer may notice a difference and choose to explore it rather than eliminate it. Improvisation does not abolish the relationship between intention and feedback. It changes what counts as a useful consequence. Information becomes material for another choice, and the next movement may be shaped by something that was not specified at the beginning.
There is no single dance centre
Brain research is often presented through a coloured image and a promise that the location of a human capacity has been found. Dance resists this kind of simple address. It can involve timing, spatial organization, sensory processing, movement generation, memory, attention, and interaction. These functions depend on connected systems rather than a single place labelled “dance.”
Steven Brown, Michael Martinez, and Lawrence Parsons studied a constrained form of dancing with PET imaging. Ten amateur tango dancers lay in the scanner and made patterned foot movements on an inclined surface. Comparisons among tasks implicated several cortical, subcortical, and cerebellar regions in different aspects of the activity.[4]
The restrictions are part of the finding’s meaning. Participants were not freely travelling across a dance floor or responding to partners. The study isolated some components under conditions that made measurement possible. It offers evidence about those components, not a complete neural portrait of dance in every setting.
A brain region becoming more active during one task than another does not, by itself, establish that the region exclusively performs the task’s named function. Several operations may differ between the conditions. Nor does an activity map tell us what a movement means to its performer or community. Those questions require additional forms of investigation.
The positive lesson is more interesting than a location claim. Dancing is possible because ordinary capacities can cooperate in demanding and flexible ways. Systems involved in sensing, moving, and timing are brought into a particular organization. We need not invent an entirely separate faculty for every human activity in order to appreciate what makes that activity distinctive.
Distributed organization also helps us avoid a false contest between brain and body. A movement is not explained by placing all significance in one and treating the other as a passive attachment. Neural processes act through a body whose structure and interactions matter; the resulting action supplies further information. The dancer is the living person participating in this relationship, not a brain operating a disposable instrument.
Attention can change its scale
A beginner may experience a sequence as several separate problems: where to place a foot, which way to turn, when to begin, and what an arm should do. Someone familiar with the sequence may attend to a larger phrase. This difference is a useful way to think about practical knowledge without pretending that expertise means the disappearance of attention.
Attention can shift among levels. A dancer might focus on a particular joint relationship, then on a pathway through space, then on the timing of a group. A familiar action can free attention for something else, while an unfamiliar detail can bring the focus back to the mechanics. These are illustrative possibilities, not a rule that every dancer experiences learning in the same order.
The phrase “muscle memory” can be helpful shorthand if we remember its limits. It does not mean that a muscle stores a complete dance independently of the nervous system. The familiar performance depends on learned relationships among perception, action, and context. Its knowledge is expressed through the body, but it is not located in muscle tissue as a hidden written score.
Verbal explanation remains useful. A teacher can direct attention, clarify a sequence, or offer an image that changes how an action is understood. Yet a sentence is not the action itself. Knowing the instruction “turn at this point” and being able to perform the turn within a phrase are different achievements. Language can guide the acquisition of a skill without replacing bodily experience.
The title of this chapter points to that distinction. A dancer may do something before being able to explain exactly how it was organized. This does not make the act irrational. It means that some useful knowledge is available in performance before, or without, a full verbal account. We encounter the same difference in many familiar skills.
There is no need to romanticize the inarticulate. Dancers can be highly reflective about their work, and technical descriptions can reveal important relationships. The aim is to acknowledge several forms of understanding. Practical, verbal, visual, and analytical knowledge can support one another while remaining different kinds of access to the same movement.
Agency belongs to the whole activity
If a person does not consciously calculate every adjustment, are they still choosing the movement? The question assumes that agency requires direct verbal supervision of every bodily detail. That would be an unusually demanding definition of action. We routinely pursue intentions through processes we cannot describe moment by moment.
A dancer can choose a phrase, follow a partner, respond to an invitation, or decide to stop without separately specifying every muscle’s contribution. The intentional activity and its many supporting processes operate at different levels. Lack of conscious access to each detail does not erase the person’s participation.
Agency can also be shared without becoming mysterious. In an imagined duet, one dancer offers a visible or tactile cue and the other responds. Each response changes the situation for both. The resulting movement may be neither a unilateral command nor a coincidence. It can be a continuing coordination in which both people have something to perceive and something to contribute.
This does not require mind-reading. Partners can use observable information, learned expectations, and agreed conventions. They can also misunderstand one another. The possibility of misunderstanding is part of what makes responsiveness meaningful. Coordination is an achievement made within uncertainty, rather than proof that two nervous systems have become one.
Descriptions of following should not be confused with the absence of choice or consent. A person can willingly take up a responsive role, and the social conditions of that role matter. The mechanics of coupling tell us little by themselves about whether participation is welcome, compulsory, equal, or hierarchical. Those are further questions about people and situations.
For now, the important point is that sensing and acting are not enemies of intention. They are among the means through which intention becomes an event. The dancer’s agency can be expressed in what is attended to, how an invitation is answered, which possibilities are pursued, and how a movement is allowed to develop.
The limits of a laboratory become useful
A laboratory often simplifies an activity so that a question can be answered clearly. A finger moves, a hand estimates its position, a participant judges a shape, or a dancer performs a restricted pattern. These tasks can seem far from the richness of a dance floor. Their distance is both a limitation and a source of experimental control.
The right question is what survives the simplification. A study may reveal something about combining sensory cues, while leaving cultural meaning outside its design. Another may clarify timing without involving expressive movement. We can use the result for the component it investigates and resist claiming that the entire human activity has been explained.
Conversely, the richness of real dancing does not make component research irrelevant. A complex performance still depends on perceiving position, generating movement, and responding to events. The challenge is to connect levels carefully. We can identify a plausible mechanism and then ask what further evidence would show how it operates in a more natural setting.
Different study designs also answer different questions. A comparison between experienced and inexperienced dancers can identify a difference without establishing exactly how that difference arose. Practice, prior experience, selection, and other factors may need to be separated. An experiment that changes one condition can offer stronger evidence about that change, while still remaining limited to the task studied. These distinctions help us appreciate research without turning every association into a story about how dance transforms a brain.
This chapter has followed that approach. Proprioception helps make the body available to itself. Several senses contribute to an estimate of the situation. Predictions and incoming information work together. Distributed processes support action, while the person participates in a meaningful activity. No single study establishes the whole account, and no single metaphor contains it.
One special kind of prediction now deserves closer attention. A dancer may need to know not only where an action will finish, but when it should arrive. The next event can be anticipated even when no sound occurs at that exact instant. A body can keep time, divide it, stretch across it, or share it with someone else. We have reached the point where sensing movement becomes sensing a pattern in time.
What to notice
Notice a familiar action that you can recognize or perform more easily than you can describe in detail. The gap is not an absence of knowledge. It invites questions about the information, expectations, and coordination supporting the act. When you next watch dance, consider how much of that work remains invisible—and then listen for the moment the dancer appears to know is coming.
Sources and further reading
- G. M. Goodwin, D. I. McCloskey and P. B. Matthews, “Proprioceptive Illusions Induced by Muscle Vibration: Contribution by Muscle Spindles to Perception?” (1972). A controlled manipulation of limb-position judgment; not a study of dance or a suggested practical exercise.
- Marc O. Ernst and Martin S. Banks, “Humans Integrate Visual and Haptic Information in a Statistically Optimal Fashion” (2002). Evidence for reliability-sensitive cue combination in a particular perceptual task; no claim that all dance perception is optimal.
- Daniel M. Wolpert, Zoubin Ghahramani and Michael I. Jordan, “An Internal Model for Sensorimotor Integration” (1995). Experimental support for combining prediction and sensory information; the chapter’s dance application is an explanatory inference.
- Steven Brown, Michael J. Martinez and Lawrence M. Parsons, “The Neural Basis of Human Dance” (2006). PET research with ten amateur dancers and constrained foot movements. It does not identify a unique dance centre or reproduce unrestricted performance.
Continue the book. The next chapter follows naturally from this one. Open the complete chapter index.