Argentine Tango: A Conversation Between Two Bodies

Video example: Valseando with a sense of freedom

In tango, we often speak of leaders and followers. But during a giro, for example, something can happen that does not fit comfortably into the simple description of one dancer leading and the other following. In the video above, at one point the dancers let go of each other and one begins to move around the other. The other responds independently, alternately turning towards and away from their partner. One dances the short accents of the rhythm section while the other follows the longer movements of, say, a violin line. They are dancing together, but they are not saying exactly the same thing. And yet, together, they express the music—perhaps even more fully than if they were doing the same thing.

A large part of the attraction of Argentine tango, and of dancing together in general, lies in precisely that freedom, in this physical conversation. Two people can create a complex dance together without choreography, sometimes without ever having met before. They do not know what movement will happen five seconds from now. Often they do not know one second in advance. And yet they can walk, turn, accelerate, slow down and stop without the dance falling apart.

When we hold each other in an abrazo, as the embrace in Argentine tango is called, we usually call the two traditional roles leader and follower. This is understandable and practical. At a traditional milonga, dancing in the ronda, there is someone primarily responsible for maintaining the line of dance, watching the space ahead and initiating new movements. But those words can also tempt us into a division of roles in which, eventually, all initiative resides with the leader and the follower is mainly allowed to... follow.

As with some other social patterns, once we become deeply accustomed to this, it can be difficult to escape. You can see this even within tango as it is already danced today. We often have to relearn that the person who initiates a movement does not determine everything that happens afterwards. The other dancer can, may, and at certain moments must make their own choices about timing and movement. The leader has to take into account not only the physical space the follower needs in order to move at all, but also the speed at which that happens, delays, embellishments, and the deliberate continuation of a movement for just a little longer. Good tango should therefore never become a physical version of Simon Says, in which one body gives instructions and the other obeys them as accurately as possible.

And yet there remains a genuine asymmetry in traditional tango. Usually the same person is expected to propose each new direction. The other has considerable freedom within whatever emerges from that—certainly compared with many other partner dances—but rarely opens a new branch of the dance themselves.

Could we change precisely that? Could the language tango has developed over more than a century—walking, sensing weight, turning, opening space, feeling one another through the torso, listening to music and constantly improvising—also be one of the richest starting points for dancing freely together?

To find out, we first need to understand a little better what leading and following actually consist of, which differences between the dancers are genuinely necessary or useful, and how initiative might pass from one dancer to the other without losing tango's clarity. Preferably without continually having to reverse the entire role, including who occupies which side of the abrazo.

A dance nobody designed

First, a little history. Tango was not invented by anyone. The dance emerged towards the end of the nineteenth century in Buenos Aires and Montevideo, in an urban world where European immigrants, criollo cultures and Afro-Rioplatense communities encountered one another. Music and movement from different traditions became intertwined and changed along the way.

There is a familiar story that men mainly learned tango with other men because Buenos Aires had far more men than women. This is not entirely invented. There were indeed large groups of male immigrants, and photographs and descriptions exist of men dancing tango together. But there is no evidence, in that simple form, that tango originated as a men's dance, or that men practised exclusively with other men so they could later dance with women.

For our purposes, something else is more interesting: men did dance with men. That alone makes it clear that the two technical roles cannot have depended on a male and a female body. The later association of man leads, woman follows is a social tradition, not a natural law of movement.

The movement language itself also changed continuously. Early descriptions contain movements we still know today alongside all sorts of things that later disappeared. Tango was not a system that was first designed and then had to be performed correctly. Dancers found solutions, copied them from one another, changed them, and allowed others to disappear.

At the heart of tango—and of what makes the dance so attractive to many dancers, myself included—is the way we can improvise together. And that is possible partly because tango has developed a relatively clear physical language for doing so.

Not knowing what comes next

Argentine tango has a remarkable quality: almost nothing about the next step has to be predetermined. Beginners often learn a salida básica, but this is not a cycle to which the dance constantly has to return. The term that perhaps describes the real foundation better is el caminar: walking.

Two dancers can express an entire tango with little more than walking together. Preferably with a little more attention and panache than when shuffling off to the supermarket, but even within that simple act of walking lies an enormous range of possibilities. As an improvising musician, this was one of the things that immediately made my heart beat faster.

Tango is, of course, much more than walking. After one step there may be another step, but there could also be an ocho, a giro, a parada, a cross, or something completely different. Even when the next step or movement happens is not fixed. You can step on every beat, use double time, slow down, follow a melody or remain still while the orchestra continues. You can even initiate a step and not complete it, as in an amague.

Rather than learning the dance only as a collection of steps that can be strung together, we can therefore see it as a continuously changing situation in which we use a toolbox of techniques. At every moment, two bodies occupy particular positions in relation to each other. The weight is here, the free leg there, the torsos are at a certain angle, there is a certain distance between them, and perhaps some movement or rotation remains from what has just happened.

From that situation, some things are easily possible and others are not. One movement changes the situation, and with it what can happen next.

Researchers such as Michael Kimmel describe something similar when they write about the possibilities experienced dancers recognise in a situation while dancing. Recent research into the geometry of tango likewise shows how strongly the dancers' positions and weight distribution determine which movements subsequently become available.

An experienced dancer therefore does not simply know many figures. They become increasingly good at seeing and feeling what can happen from here. And there is another layer: not everything that is possible is also a good idea at that particular moment. A large turn may be technically possible when there is hardly any room behind you. A sacada may fit geometrically but become uncomfortable because of the size of the step. A movement may be technically perfect and still say absolutely nothing about the music playing at that moment.

Again and again, this produces a simple process:

state → possibilities → preferences → new state.

To me, that is a much more interesting basis for improvisation than a catalogue of figures.

The music helps us choose

Music plays a much larger role in this than merely telling us when to move. A sharp rhythm from bandoneons or piano makes different movements attractive than a long violin line. A sudden silence may invite us to do absolutely nothing. A melody that builds tension over several bars may give us reason to let a movement continue for longer.

The more tango you dance and listen to, the more you learn to feel these relationships. As a musician, I may have become consciously aware of them a little sooner. Certain rhythms almost automatically evoke a particular subset of all possible movements for me. Not because the music dictates exactly which movement I should make, but because some possibilities simply feel much more logical and attractive at that moment than others.

The music therefore does not enter the picture only after we have decided what to do. It changes the landscape from which we choose.

And both dancers hear the same music. That does not mean they hear the same thing. One may be absorbed in the rhythm while the other follows a melodic countervoice. That is precisely where tango can become beautiful. Two bodies do not have to keep saying the same thing about the music. As long as their movements are compatible, they can express different voices.

That is also one reason tango works surprisingly well with many different kinds of music. Classical tango music can be extraordinarily rich in rhythm, melody, countermelody, phrasing and contrast between instruments. Much non-tango music is simpler in those respects, something traditionally minded tango dancers sometimes rightly find less musically interesting. Modern music, on the other hand, can be much richer in sound, bass, stereo image, texture and the physical impact of good production. Tango does not prescribe which of those qualities we have to dance, and can therefore handle a surprising range of music. At some milongas you can see this reflected directly: classical, neo- and non-tango alternate throughout the evening.

Music also helps the dancers communicate. The end of a phrase, a silence, a strong accent or a change of instrumentation is heard by both. Such moments become natural places for a movement idea to end, continue or make way for something else. Music is therefore not merely decoration. It is almost a third participant in the conversation.

Traditional tango already has two voices

We do not need to dismantle the existing dance before we can make tango more reciprocal. Much of what we need is already there. Take an ocho. One dancer may initiate the movement, but that by no means determines everything that follows. The other organises their own axis, makes the pivot and ultimately takes the step themselves. The actual size of that step and the time the movement takes become part of what the first dancer subsequently has to feel.

This becomes even clearer in a giro. One dancer can initiate the turn while the other steps around them. But the outside dancer takes real steps, with a real length and speed. Those steps determine the actual radius and tempo of the giro. The dancer in the centre cannot completely dictate this in advance and has to keep listening to what is actually happening.

At a high level, you regularly see movements that beginners strongly associate with one role being used in both directions. Men or traditional leaders make ochos themselves, move around their partner, or are turned in a planeo by the traditional follower. Sometimes the roles visibly switch, but often something subtler happens: the basic roles remain recognisable while one particular function moves to the other side for a few beats.

Within a movement initiated by the other person, the second dancer also has room to retain a musical voice of their own. A step can be placed softly or sharply. A pivot can have its own phrasing. A free foot can draw the music. A movement can be small and rhythmic or expansive and flowing, as long as it fits what the two bodies are doing together. This is an important distinction: having a voice of your own is not the same as taking a new initiative.

Proposal, uptake and division of tasks

So traditional tango has long contained that second voice. But what we want to investigate is what happens when the conversation becomes genuinely reciprocal. A new initiative begins with a proposal. One dancer uses their body to make a new possibility perceptible. The other feels that proposal and can take it up. Only after that uptake does it become something they are going to do together.

A temporary division of tasks then emerges. In an ocho, the person proposing the movement has different tasks from the person executing the ocho. In a giro, the division is different again. One dancer may be responsible for opening and maintaining a direction or rotation, while the other organises their own balance, pivot, step and actual arrival. These are not higher and lower tasks. They are different parts of the same shared movement.

Much of the intelligence tango has accumulated over its development resides precisely in these divisions of tasks. Not every asymmetry is a problem we need to solve. Two bodies cannot occupy the same place, and many movements work precisely because each dancer temporarily does something different.

And then there are embellishments

Alongside what is necessary to execute the shared movement, there is another kind of freedom: the embellishment, or adorno.

An embellishment is optional. The movement works without it. A free foot can draw something extra, a rhythmic detail can be added, a leg can trace a different line. Such an embellishment often arises directly from something the dancer hears in the music.

The important difference is that the partner does not have to respond to it.

Some things I do are necessary for my part of our shared movement. Other things I do because I want to express my own voice within that movement. Neither automatically means that I am now asking something new of you.

That distinction becomes important as soon as both dancers are allowed to propose new initiatives themselves. Otherwise the dance becomes restless. If every small variation or embellishment is interpreted by the partner as a new invitation, both bodies will constantly respond to messages that were never intended as messages.

Listening well, then, does not mean reacting to everything. Listening well means being able to feel what requires a response and what you can simply allow the other person to do.

That may initially seem like a minor detail, but I think it is one of the conditions for making reciprocal tango not only freer but genuinely pleasant to dance.

A new proposal

The interesting question, therefore, is not how we can make everything symmetrical, but which asymmetries help the movement and which we have retained merely because traditionally the same person is always the leader or follower.

This becomes even more interesting because the abrazo itself is asymmetrical. The two arms have different functions. The arm around the back may need to slide further around during a turn or, conversely, make more room. The torsos are not always positioned exactly opposite one another. As a result, a movement that appears easy to mirror on paper may require a different practical execution on the other side.

Equality therefore does not mean both people have to do everything identically. It means both people have access to the same possibilities and principles, while the actual division of tasks can take account of the positions their bodies really occupy.

Suppose we look at it this way: during a movement, the dancer who did not initiate it can begin proposing something new. How do we begin that conversation?

Imagine A has proposed a giro and B has taken it up. While the giro is happening, B does all sorts of things that are simply part of their task. B may also embellish. A does not need to interpret any of this as a new initiative.

But B can also do something with a different intention. B can make a possibility perceptible for what might happen next. Perhaps a certain rotation is deliberately maintained. Perhaps the orientation of the torso changes in a way that is not required for the current movement. Perhaps a small projection develops in a new direction. Or perhaps B simply wants the existing movement to continue for longer than A originally had in mind.

For the purposes of communication, it does not matter very much whether this extends the previous idea or introduces something entirely new. In both cases the same thing happens: B makes a proposal. That proposal is not yet a movement that A has to execute.

This seems to me an important principle for reciprocal tango: a new initiative is first offered, and only becomes a shared movement after it has been taken up.

That means initiative does not have to be fought over. I do not have to physically pull you in the direction I have imagined. I only need to provide enough information for you to feel that a possibility is there. If you take it up, we can develop it together. If you do not, I can let it disappear again, either by doing nothing or by continuing to move with your proposals.

A proposal can also appear gradually. While you are still finishing a phrase, I can already allow something small to emerge that might become what happens next. As an improvising musician, that feels familiar. Near the end of another musician's solo or phrase, you can cautiously add a few notes. Not to say, stop, it's my turn now, but to make a possible next direction audible. The other person can make room for it, respond to it, or continue for a little longer.

Tango as an ideal basis for dancing freely together

Tango possesses a remarkable combination of qualities. The dance has developed a sophisticated language for weight, axis, direction, rotation and distance. The abrazo provides continuous information about the other person's body. No fixed choreography is required. Stillness is allowed. The music can be interpreted rhythmically, melodically and dynamically in many different ways. Movement can remain small enough to function among dozens of other couples and at the same time become enormously rich technically.

Above all, tango has more than a century of experience with a rather complicated problem: how can two people move together when almost nothing about the next movement has been determined?

That makes tango, for me, not just a beautiful dance but potentially one of the best foundations for dancing freely together.

Contact improvisation explores some of the same territory from a different history and with different possibilities, for example by going much further in shared weight, floorwork and changing orientations. Other partner dances have found other solutions. But tango has something special in its combination of continuous contact, clear physical language, improvisation and applicability to many different kinds of music.

If everyone learns the entire language from the beginning, something emerges that goes beyond simply “being able to lead and follow”. You learn how to make a physical proposal without forcing another person. You learn how to take up a movement without becoming passive. You learn which parts of a shared movement are your responsibility, where you can play freely without asking anything of the other person, and how to communicate that you yourself see a new possibility. It also creates much more freedom for dancers of different experience levels to find their own balance, something that can have a major influence in a more traditional division of roles but here leaves much more room for that balance to emerge naturally.

You also learn to let go of a proposal. That may be just as important as learning to take initiative well. Dancing freely together can only work if neither person needs their idea to win. Tango already has a wonderful tool for this: we can always do nothing for a moment.

An etiquette within and beyond the abrazo

That contact gives us an extraordinarily rich channel of communication. Through the abrazo we feel weight, projection, rotation, direction, tension and relaxation. We feel whether a body is arriving or wants to continue. Often, even before a step, we can already feel which possibility is being prepared. But we also follow each other without an abrazo: we remain oriented towards one another, usually with the whole upper body, sometimes briefly with the face alone.

It therefore seems unnecessary to invent an entirely new system of gestures for reciprocal tango. Preferably not, in fact. A kind of Morse code in which a particular shoulder movement always means “I'm taking over now” would probably make the dance poorer rather than richer.

It is more interesting to investigate what information existing tango technique already contains and how we can make it legible in both directions. One particularly interesting moment is the end of a movement. Often both bodies arrive in a state from which many possibilities are available again. But the dancer who has mainly been executing a received movement can also deliberately choose not to return completely to the most neutral position. A certain torsion may remain, a projection may become perceptible, the connection may indicate that something is still continuing.

That can become an invitation: follow what I see here. For this to work reliably, we need a few simple conventions. A movement that already clearly contains weight or momentum takes precedence over a new proposal that conflicts with it. A proposal begins small enough not to be forced. Whoever proposes something gives the other person room either to take it up or not. Once it has been taken up, both dancers respect the division of tasks belonging to the movement that has emerged. And if two things happen at once and neither dancer is quite sure what the other means, nobody has to win. Slowing down or standing still is still perfectly good tango.

Equality does not require simultaneity

One obvious objection to dancing without a fixed leader is this: if both dancers are allowed to choose, what happens when they want different things at the same time?

The traditional division of roles solves this problem simply. One person is given the default initiative to choose the next shared path. But we are looking for a conversation in which both dancers may make proposals, a conversation on the dance floor in which neither has to dominate.

After all, two people can participate in a conversation as complete equals. Usually only one of them speaks at a time, but that does not make it a long monologue. We have learned all sorts of subtle ways of signalling that we are continuing, finishing, wanting to add something or giving someone else room. Sometimes we briefly speak over each other and one of us naturally makes space.

Jazz works in a similar way. A saxophonist may be playing a solo while the drummer, bassist and pianist are anything but passive. They continuously make decisions within their own tasks and can strongly influence the musical character. Sometimes an idea from the drummer or bassist creates a new direction that the entire group adopts. Nobody needs to establish formally at every bar who the leader is.

That distinction is useful in tango as well. There can be a difference between initiative and direction. I may have considerable influence over the overall direction of the dance for an entire musical phrase while you retain a pronounced voice of your own within it. Meanwhile, you can make a small new proposal without immediately taking over all direction. If I go with it, that direction can gradually shift towards you.

Equality therefore does not require simultaneity, nor does it require bookkeeping in which we check every eight beats whether both dancers have had exactly the same amount of influence. The freedom lies in both being allowed to speak and both having learned how to listen.

Everyone learns the whole language

If we want to move in this direction, it also has consequences for how tango is taught. Everyone could learn the same movement principles from the beginning: walking, pivoting, making forward and backward ochos, moving around the partner, being the centre of a turn, opening and occupying space, making proposals and taking them up.

That does not mean beginners should immediately try to do everything at once. Quite the opposite. To learn communication properly, it is often useful to make an exercise temporarily very asymmetrical.

One dancer, for example, may be given only the task of proposing a weight change, while the other's only task is to feel and take it up. Then they switch. Next, they investigate what is actually divided between the dancers during an ocho: what must the person proposing it do, and what does the person executing it organise themselves? Free embellishment can then be added, with the explicit understanding that this is not a new proposal. Only after that does the dancer executing the movement get the opportunity to offer something new while it is still being performed.

In this way we can learn symmetrically, practise temporarily asymmetrically, and eventually dance freely and asymmetrically. This has another advantage: it becomes much easier for everyone to teach everyone else something.

In traditional teaching, someone who has followed for years may know an extraordinary amount about how an ocho feels and is performed, while having far less experience in offering the rotation. An experienced leader, conversely, may have initiated thousands of ochos without often performing the movement themselves.

If both dancers know both sides, you can literally say: you do what I'm doing for a moment, and I'll stand where you are. Instead of merely explaining a problem, you can let each other feel what is happening.

Knowledge becomes much less dependent on role. The teacher remains important, of course, but no longer needs to be the only person in the room capable of transferring knowledge. Everyone acquires more tools for investigating with any partner why something does or does not work.

If we retain the classical abrazo—and there are certainly good reasons to do so—that also means learning to initiate all movements within that abrazo in mirror image. This creates new possibilities and space. Many existing movements have a strong preferred direction because of the position we traditionally adopt, with hand contact on one side and arm contact on the other.

Understanding the position, not just the figure

From quite early in tango's history, there have been teaching principles in which, instead of learning only how a particular figure goes, we learn to recognise which possibilities arise from a particular position, and which techniques those possibilities consist of.

A sacada then becomes more than a sequence of memorised steps. You learn what is actually happening: one dancer vacates space and, from a particular geometry, the other can occupy it. You can then investigate how many different ways that situation can arise.

The same applies to an ocho. What state makes a pivot possible? What changes if the other dancer moves? What happens if both dancers pivot? Can the same technique be performed in the other direction or by the other body? What happens when, within the traditional abrazo, the dancer on the mirrored side initiates the ocho?

A figure then ceases to be a recipe and becomes a recognisable instance of a more general principle. And it becomes much easier to understand why some possibilities are preferred possibilities. Not everything that is technically possible works equally well. The position of the feet matters, as does the abrazo, whether the arm around the back can slide or not, momentum, differences in height between the dancers, the available space on the floor and, of course, the music.

If you know only the figure, you know one route. If you understand why the figure works from this particular state, you begin to read the landscape—and that is where creativity really begins. Suppose a movement has ended and we are each on our own axis, standing on one supporting leg. I can now move your free foot. The direction does not matter. Under certain circumstances I can even lift it. I can keep moving it until it becomes a pivot, move it with one leg and, after a step, move it back with the other. You can take the initiative and move my foot back to where it came from—something that is already becoming more common in traditional tango.

And what if a misunderstanding occurs? Those are interesting too. If an adorno is repeatedly mistaken for a proposal, perhaps the signal was unclear, and together you can investigate why—or ask another dancer or teacher. Sometimes a misunderstanding goes wrong in such an enjoyable way that it becomes inspiration for something completely new.

A conversation with two voices

This finally brings us back to the image with which we began: two dancers moving in a giro. One of them initiated the movement at some point, but that does not determine everything that follows. The other helps determine its speed and geometry, hears a different line in the music and gives it form. The first dancer listens and adapts. Perhaps, meanwhile, a small proposal for what comes next emerges from that second voice. The other feels it, makes room, and without anyone formally announcing that the roles have now been reversed, initiative shifts. A little later it can move the other way again.

None of this means traditional tango is wrong, or that nobody should want to be a leader or follower anymore. Sometimes it is wonderful to spend an entire tanda mostly following. Sometimes someone has a musical idea you are happy to explore with them for several minutes. Sometimes a clear temporary asymmetry simply works best.

Equality does not lie in requiring both people to take exactly the same amount of initiative all the time. It lies in the possibility that they both can.

The abrazo then becomes not a channel through which one body tells the other what to do, but a place where two people continuously share information about what is happening and what could happen. There can be a proposal, an uptake, a temporary division of tasks. Within that movement, two musical voices can exist without getting in each other's way. Embellishments can remain simply embellishments. A new proposal can begin to appear before the previous idea has completely disappeared.

Tango remains recognisable. There is still structure, there are still differences in tasks, and there are still moments when one person has much more direction than the other. The only difference is that it no longer has to be determined before the dance begins who will occupy which side of those differences throughout. Perhaps we can simply call it the emancipation of both follower and leader, giving each much more room to develop towards what they can and want to do.

For me, that is not a break with tango, but a fairly natural next step in a dance that has always revolved around improvisation, listening and responding to what is actually happening. Tango has already given us an exceptionally rich language for moving together.

Let's take that next step together.

Further viewing and reading

For those who would like to explore some of the ideas behind this article further:

Michael Kimmel — Intersubjectivity at Close Quarters: How Dancers of Tango Argentino Use Imagery for Interaction and Improvisation A particularly interesting analysis of tango as shared bodily improvisation. Kimmel explores how dancers continuously recognise possibilities in one another's posture, movement and position, and how two people can function as a single moving system without choreography. Read the article at De Gruyter

Luna Beller-Tadiar — Queer/Tango/Theory: Gendered Semiosis, Dancing the Binary, and Dancing on Out A recent study of queer tango in which communication between the dancers takes centre stage. Beller-Tadiar discusses mutual listening, initiative, interpretation and role-switching, as well as the interesting transitional territory in which it is no longer self-evident who is leading and who is following. Read the open-access article at Cambridge University Press

David Kaminsky — Social Partner Dance: Body, Sound, and Space A broader book about social partner dance, drawing on tango, salsa, lindy hop and blues, among others. Kaminsky looks at partner dance as an interplay between yourself, your partner, the music and the surrounding space, and also examines why lead/follow is such an effective solution to the coordination problem faced by two improvising dancers. Precisely for that reason, his work provides an interesting counterpoint to the proposal made in this article. View the book at Routledge