The discovery of synchrony: By means of the projector as a scientific instrument

Almost any turn of the kaleidoscope of nature may set up in the artist this detached and impassioned vision … and as this harmony becomes clear to the artist, his actual vision becomes distorted by the emphasis of the rhythm which has been set up within him.

The man and the woman move toward each other. They recross their legs while sitting in chairs. They draw these chairs closer by an inch or so before the man asks, ‘So how can I help you?’, leaning back as he says this. Then he leans forward and rises from the chair before sitting down and asking, ‘So how can I help you?’ He does this repeatedly, over and over. And it takes him progressively longer each time. He also repeats the same physical movements—not merely similar, but truly identical movements. His right hand pulls his pant leg up to expose the sock; then the hand moves away from his midriff. The hand stops in midair and begins to travel back. It moves outward again and gets slightly farther, all the way to a pad of paper on which it can lean. While the right hand stays put, the left tugs at a suit coat, then releases the coat and rests on the chair. Again it moves to touch the suit, the chair, the suit, tracing this circuit a total of four times. Finally it settles on the chair arm as before, while the man completes his sentence: ‘So how can I help you?’ But his words are no longer really intelligible; they skitter and stutter in fragments of phonemes. He could speak them a thousand times without getting tired because he speaks them on a filmstrip that is running back and forth.

The hand on the projector controlling these movements releases its hold and the film comes to a stop. Demonstration continues with the man's further movements during four and a half seconds of the woman's response. ‘Yes’, she says, ‘I said what was that, and then I says well, okay what—you know, what can I do about it?’

On the oh in ‘okay’, he turns in his chair.

Over kay his left leg and both his hands rise.

During what the same leg changes direction as the foot begins to move to the opposite knee. Simultaneous with this, the right elbow flexes.

The projector runs each syllable and each set of movements from three to six times for the benefit of viewers: oh—turn, oh—turn, kay—up, kay—up, what—flex, what—flex. And more of the same throughout the entirety of ‘okay what—you know, what can I do about it?’ The movements are in sync with the changing vocal pattern. Speaker and hearer are synchronous with each other. They go forward and backward, forward and backward, until their synchrony is visible and the projector can be turned off.

Revolutions in thought

William S. Condon (1925–2006) spent a lot of time giving demonstrations with the use of film. The instance described above first appeared on British television in a popular science program in 1975 (Bates and Massey, 1975). By then he had been giving them for over ten years. He was especially good at manipulating films to bring out small details of human behavior. Actually, the success of his lectures depended entirely on the skill with which he cranked the film at different speeds. People spoke admiringly of Condon's ‘magic wrist’; others objected, and compared it to hypnosis (Condon, 1982a: 76; Rosenfeld, 1981: 93). But they agreed that he was patient in everything he did. His scientific work consumed countless hours of repetitive viewing of very short segments. He was of the second generation of behavioral microanalysts, and he set a new standard for detail and thoroughness (see Figure 1).

OPEN IN VIEWER

Figure 1.

William S. Condon with film equipment; photographer unknown (Esper, 1978).

Over the years he looked increasingly for only one thing, the phenomenon referred to as interactional synchrony. Synchronous, of course, means at the same time, and synchrony a state of shared rhythmic changes. In the context of behavior, there is basic self-synchrony: the unity of voice and body when one person talks. The voice leads the way with its changing phonetic boundaries and the body falls in line with these sub-second changes. The hands start to move with the onset of oh and they do not change direction until the onset of kay—evidence of what Condon calls ‘a unified human being’ (Bates and Massey, 1975). And the most amazing thing about this self-synchrony is that it is matched by the listener's behavior. The listener also moves, and moves almost as precisely, in one way over oh and another over kay. That is the meaning of interactional synchrony, and no limb in particular is key to its achievement. The listener could move the hands, but could also move the feet, or more likely both together as well as the trunk and head. So too for the speaker, who may be moving parts different from those of the listener. The only thing that matters is that there are changes—of direction or speed, of initiation and conclusion—and that the changes change together when the vocal pattern changes. Condon often spoke of a delicate ‘dance’ in miniature, one happening all the time when people are talking. It is a subtle yet important part of feeling at ease with others.

Though synchrony might be felt at this intuitive level, it cannot be perceived without the aid of instruments. A film of interaction does not alone suffice, for the film must be looked at in a special way. Accounts of Condon's research tend to refer to films viewed either frame by frame or in slow motion. Both are inaccurate; his technique was quite different. He employed some variety of manual projector that could variably scan the film in small sets of frames. Frame-by-frame viewing would not be much help, insofar as single frames do not reveal movement. Two frames side by side might show a displacement but they would also tell us nothing of the nature of the change—whether it is part of an ongoing movement, the beginning of a new one, or the completion of an old. Slow motion would preserve the movement as movement but it would not help determine the change points precisely. The movements would blend together, obscuring sub-phases. Only a machine that runs just three or four frames, at almost full speed, contrasted with the next few frames at almost full speed, could show exactly where the behavioral change points fell. Adam Kendon later called it ‘movement phrase boundary analysis’ because the movements are akin to phrases in language: the phrase has its unity within the larger sentence (1977: 225–40). Such analysis required the manual projector and, as mentioned earlier, great skill in its use. ¹

Skill has become a topic of much interest lately in media studies and the humanities generally. Skill is one aspect of cultural techniques, the chains of operations that create the world we know. A door enacts the difference of inside and outside, and thus enacts the difference of subject and object. Corrals and pens separate animals from people. These are techniques of basic distinctions; one is almost tempted to call them primordial. Less has been said about cultural techniques of scientific practice and scientific instruments. ² These too participate in chains of operations, all highly formalized, repetitive, and recursive. For the operations we perform on things in the world have a tendency to work on and change us in the process. The results of that process cannot be predicted. Historians of science have occasionally said as much, although perhaps with greater deference to the scientist's agency. ³ The cultural-technical concept has little room for agency: its chains of operations precede understanding. They develop in us habits we did not have before. They always appear before the insight or knowledge or whatever we want to call it that we gain with their aid. ‘I hold the fates bound fast in iron chains’, says Tamburlaine, ‘And with my hand turn Fortune's wheel about’ (Marlowe, 1969[1590]: 116). But for ordinary mortals, it is the hand that must adapt to the wheel's demands.

A look at projectors as scientific instruments has much to offer this new field of research. Not only do they animate a series of stills; they animate the hands of their users as well (Cartwright, 2011; Curtis, 2011). And their effect on these users is rarely just sensuous as the mind starts to move within the same circuit. We can show how a theory or even a whole cosmology arises from the handling of definite instruments. Condon's biography is an especially good example because of his long devotion to one technique and theory. The role of the human hand here is vivid indeed. Every time one turns a crank to make pictures move, changes are effected in the observer. It must be turned and turned again for fine-grained analysis. One has to turn it thousands, maybe millions of times. Thus the study of rhythm is itself rhythmic, as we start and stop and start the figures on film. The rhythm is partly set by this type of projector, which in turn will determine the kind of rhythm seen. And the kind of rhythm seen must affect our ideas of life and the living, of normality and pathology. This is what the following study aims to show—that the history of ideas is never just ideal. It would be surprising if our instruments did not somehow leave their mark on the shape of a concept they help us produce.

Beginnings and refinements

Biography, traditionally, proceeds from the assumption that what a person does in life is explained by what came before. Biography appraises the character of a person as it influenced or was influenced by a place and time and group. Among the special burdens of any scientist's biography, especially one framed in terms of techniques, is to show how the encounter with a scientific instrument both answered to and changed the investigator's needs; how it influenced and was influenced by the drift of one's ideas. For instruments are media in which our thought is propagated. Hence Thomas Kuhn spoke of ‘instrumental commitments’ as the bedrock of practice that sustains a world of concepts (2012[1962]: 41).

The concept of synchrony has had a long career, and few discussions then and now have failed to cite Condon. Remarkable, then, is how little we know about him. Even a date of death is difficult to determine by any recourse other than a public records search. Former colleagues still living do not recall much; he may have had a very retiring personality. Like Herodotus or Thucydides, whatever we do know is largely based on what he tells us in his own writings. ⁴ The facts, such as they are, appear to be these. Condon was born in Sligo, a small Pennsylvania town, in 1925; that is, between the wars. He attended Gannon College, where he dabbled in literature and ran a great books club. He had a regular column in the school's student paper, and in one he posed a question that guided him later: ‘How much of life's beauty do we miss because we lack the ability to perceive it?’ (1949: 3). His education was either interrupted or postponed by a three-year term of service in the US Navy. He was probably poor, as he worked as a hotel desk clerk to put himself through college. After getting his bachelors in 1949, he went to the University of Pittsburgh for graduate study. There, he studied English for three years before starting over and changing to philosophy. All in all he spent seven years obtaining his master’s. His PhD took six, and in 1962, at the age of 37, he finally received it. During this time he continued to work; among his positions was psychiatric attendant. From 1950 to 1953, his evenings were spent at the university clinic, where he returned as a part-time lecturer in 1959.

The Western Psychiatric Institute and Clinic, currently Detre Hall, is an 18-story building on the University of Pittsburgh campus. Its director at the time was Henry W. Brosin, who soon became a force in Condon's career. It was Brosin who introduced him to the field of ‘kinesics’, or the study of body motion in a communications framework. The term was Ray Birdwhistell's, a colleague of Brosin with a research appointment at a clinic in Philadelphia. These two had worked together and with Norman McQuown since 1956 on The Natural History of an Interview. The intricate story of this unpublished book, of the working group that wrote it, and of its underground influence need not concern us. It was a model of basic interdisciplinary research in the behavioral sciences after cybernetics (Leeds-Hurwitz and Kendon, 2021). Most relevant is that the project made use of film, formalized techniques of microanalysis, and dragged on intermittently at three institutions. Thus Condon was one of eight recruited in Pittsburgh to use and extend this working group's methods.

Brosin could not instruct them in the use of film, so he sent them to Birdwhistell at the Eastern Pennsylvania Psychiatric Institute. This was in 1962; the training was two weeks long. Condon soon emerged as the star of the group and was commended for his ‘industry and devotion to the topic’. ⁵ He was a mediocre philosopher but an excellent observer of human behavior stored on strips of film. Later, in 1964, he went to see McQuown at the University of Chicago. In this case, he stayed for almost one year to master the technique of sound segmentation. In 1966 and 1967, he published the first results of his synchrony studies. ⁶ They made him a major player in the linguistic-kinesic field; soon everyone was talking about interactional synchrony.

He left Pittsburgh for Boston in 1971, taking a job in the School of Medicine at Boston University. The terms of his appointment are not really clear but his name shows up in catalogs well into the 1990s. He earned a second PhD in counseling psychology in 1982, then went into private practice. The cuts to federal funding that affected nearly everyone in the behavioral sciences had hit him as well. He was also dismayed by the growing skepticism concerning synchrony in the scientific community to which he belonged. Several people tried to reproduce his results, and when they were unable, they tended to discount his work. Their verification studies were not without flaws, but to point out these flaws was not to prove synchrony. ⁷ Nor did Condon have a wealth of disciples to carry on his work and defend his ideas. He seems to have supervised few dissertations. People sat at the projector with him and learned his technique, then went off into the world to do their own thing. By 1982 one could say with some sadness that the theory of synchrony was under ‘a cloud’ (Skupien, 1982: 3). Still, it was evoked in popular articles, and some remembered that a Boston scientist was the one behind it (Douglis, 1987; Quinn, 1982; Ross-Flanigan, 1988). Most of his early colleagues lost track of him entirely, and there was apparently no obituary placed when he died.

In all this the year of 1962 appears decisive. It was then that Condon learned to handle the instrument whose workings he described in nearly all his publications. It was also the year in which he earned his degree for a dissertation titled ‘The Freudian Model of Human Nature’. As we will see, his interest in depth psychology would persist a few years more. One should not overstate the nature of the break effected by the instrument in a career. Its influence on perception and cognition is gradual; the novelty of what it offers may not at first be seen, and Condon could easily assimilate kinesics to psychoanalysis in his dissertation. Over a year later, in 1963, he still sees the body as ‘the behavioral locus of the unconscious’. ⁸ It would take another year of work at the projector to defeat this inheritance of Freudian psychology, and even then the extirpation was not quite complete. At the same time one can say that his use of the projector was an extension of his methods as a historian of ideas. He had apparently spent two years reading Sigmund Freud and taking notes specifically with his topic in mind. But the reading, he tells us, was somewhat haphazard, and he started again in chronological order (Condon, 1962: 3–4). He went so far as to produce a three-volume index with annotated entries for all of Freud's works, and the entries were arranged in year-by-year format so that one could chart more easily the growth of any concept. Thus the writer would be compelled to follow the chronology and discouraged from skipping intermediate phases—as if Condon now had the works of Freud before him reduced to a kind of filmstrip he could reverse and advance at will. He had a temperament well suited to review and repetition. In his later work on synchrony, it became a point of pride to account for every frame in ‘historical’ sequence (Condon, 1976: 286).

Some people invent their instruments; others find them where they can, and Condon's choice of instrument was probably dictated by what was available on campus already. Some of the most avid users of analysis projectors were not scientists at all, but ambitious football coaches. The coach could film the game and review the plays at leisure. A variant of Bell & Howell's Diplomat model was a popular and affordable option for this pursuit. Soon the research clinic was using it as well. Condon liked it very much and continued to use it, even if others spoke of it as just ‘that old thing’ (Birdwhistell, 2021[1980]: 258). It was absolutely central to his research program, so much so that he asked for one while in Chicago. ‘At the moment, and perhaps unwarrantably, I seriously feel that some degree of manual control over the scanning process is required’, he wrote McQuown. ⁹ He rejected other models with only automatic features, no matter how many frame rates they offered. He needed constant variation in the range of frames selected, as well as variation in the speed of frame advancement, in order to detect what he came to call synchrony. The Bell & Howell Model 173BD, also known as the Time and Motion Study Projector, answered to his needs at this point in time.

Many copies of this instrument are still extant; it can be used and examined, a great benefit to the historian. Its supply reel is located above the reel for take-up in a vertical arrangement, with the lens installed between them (see Figure 2). One threads the 16 mm film from the supply arm in a line parallel to the top of the lens, over a roller with sprocket teeth, making a ‘loop’ before passing the projector gate, then another loop and a roller with teeth again. Where it emerges below the lens, the film is tucked in or taped to an empty take-up reel. The projectionist turns on the lamp switch at the base and only then switches the LINE from off to on. The film moves through the gate, past the light, over teeth, apparently continuously but really intermittently; and the result is a moving picture on the plane surface opposite. This is the basic Diplomat structure. The 173BD makes several additions. At the back of the projector is a knob one can turn either clockwise or counterclockwise to increase or decrease the speed. The speeds are on a gradient from eight hundred to 1500 frames per minute, or from about half-normal to standard sound speed. From the supply arm extends a Veeder-style counter that ticks upward incrementally as the filmstrip is advanced. Each number registers an additional frame, and the counter can be set to zero at any time one wishes. Most interesting of all is the hand crank that sits perpendicular to the lens on the projector's right side. To activate the model's manual features, one must first release the clutch on the hand crank itself. Now the image is frozen on the wall or screen opposite, and the projector is so designed that the film will not burn up. This image will change only if and when one grabs the crank and begins to revolve it forward or backward. Each revolution moves four frames of film. The revolution is as slow or swift as one likes. It can advance a single frame and then stop to hold the frame, or it can move the film quickly in three- or four-frame bursts. In the latter case, the image looks more or less as it does when one projects it normally at standard sound speed. This requires much practice and precision, however. Especially hard is not to land between the frames, for then one ends up with only a darkened square. The projector also makes an extreme amount of noise and it generates a smell of heat that soon pervades the room. It is made for silent film, hence sound is treated separately, often with a film sound reader and sometimes an oscilloscope.

OPEN IN VIEWER

Figure 2.

16 mm ‘Time and Motion Study’ Projector, Bell & Howell Model 173BD. Photograph by Faith Holland; from the collection of the author.

Slow motion, as we saw, was fairly useless for Condon's purpose, while his filmstrips would have had their frame numbers printed on them. Only the crank was truly necessary for the study of synchrony, which proceeded step by step in the following way. When a film is shot at normal speed, or 24 frames per second, a set of eight frames is one-third of a second. Even two frames can contain much activity in a film of people talking, gesturing, and shifting. It is easiest to focus on one figure at a time, and on a single bodily area within the chosen figure. Perhaps the speaker's right hand is what first commands our interest. Very well, we focalize this hand and study its changes. We turn the projector's crank for a half-revolution in order to move the film through frames 1 and 2. We must turn it pretty quickly, ‘sweeping’ across the frames. The speaker's hand is moving down toward the speaker's lap. The wrist flexes slightly. The motion, at this point, appears to be smooth—it does not change direction and it does not cease to flex. We add a third frame to our set for sweeping; we go back to 0 and move the crank again. Frames 1 through 3 now flash on the screen before us. There is still no visible change in the figure's right hand, which continues to descend in a straight line. We add a fourth frame to our set for sweeping, turning the crank for a full revolution. Still nothing to report, so we add yet another. And as frames 1 through 5 appear, we think we see a change. The movement of the hand onscreen is no longer smooth: it contains a kind of ‘bobble’ at the tail end. ¹⁰ Something has changed, though it is hard to say what. We may even have to sweep through frame 6 or 7 to verify the change and clarify its nature. Now the speaker's hand goes up instead of down; the wrist has changed slightly from flexing to extending; and we infer that these changes took place on frame 5 (see Figure 3). Only when we added frame 5 to the set did we see a disturbance in the movement's trajectory. The same can be said of movements in the head and eyes, the shoulders, the trunk, the legs, the feet, not just for the speaker but for the hearer as well. We might have to scan the same segment for minutes before all these changes have been detected. We mark them progressively on coordinate paper, probably in a workbook or just a long sheet. When we collate these results with a separate sound analysis, we can see if and when any synchrony was present (Condon, 1970b, 1982b).

OPEN IN VIEWER

Figure 3.

Change of wrist motion from flexion to extension (Condon and Brosin, 1969).

Any unit of synchrony is a cluster of movements that a person sustains together for consecutive frames—sustained in such a way that no bobble appears. And when the bobbles do appear, they fall at phonetic junctures so that the sound is isomorphic with the stream of movement. Thus the unit is defined by the operations that reveal it: sweeping across the filmstrip, and the perception of minimal difference. The question remains of what exactly to call these units; language has so many implications for theory. It is interesting that Condon spoke of ‘contour units’ in his earliest reports to Brosin and others. ¹¹ But a contour is what divides one thing from another, marks it off as separate, and this is not really what he meant to convey. His units of synchrony had two peculiarities. For one, their definition is strictly temporal and does not demand any specific kind of movement. It is enough that there are changes and that the changes change together: the head, arms, and feet all together, or the eyes and right-hand fingers, or maybe just the trunk as it turns across a vowel. Any combination of these can form a unit. And the unit is a function of the scale of the observer. The same piece of behavior—although ‘piece’ is misleading—can participate in many different levels at once. When someone says the word keeping, perhaps the head goes up and right at the onset of kee, down and right slightly while accelerating over pi, then farther right slightly before stopping at the end of ng. But the same three units resolve into two when we move from the phonetic to the syllabic level. For there are only two syllables, kee and ping. The major kinesic pattern is now the movement up on kee and the movement down on ping. And if we take the whole word as our integral unit, we see the head turn left to right for the duration of keeping. ‘Behavior is both discrete-like and continuous simultaneously without contradiction’, Condon declares, and therefore its units are like ‘waves within waves’, each smaller one merging into the bigger that enfolds it (1970a: 22; 1978: 8). The unit is not a thing but a form of organization. It is a moment of a process, a wave in the behavioral sea. The concept of process won out over contour and Condon spoke of ‘process units’ in all his publications (see Figure 4).

OPEN IN VIEWER

Figure 4.

Process units (Condon, 1970b).

‘The assumption that nature is “process” I find to have great heuristic value’, he wrote to a friend. ¹² He was formed in the matrix of John Dewey's thought, on which he based his master's thesis in 1956. Of particular importance was Dewey's aesthetics, which Dewey always traced to the life of the human organism. The chapter of Art as Experience on ‘The Natural History of Form’ offers many parallels to the concept of synchrony. All life is rhythmic, according to Dewey, from the in-out of breathing to the diurnal cycle. Rhythm is order within variation, is the recurrence of relationships despite varied contents. It gives form to every moment of an ongoing process. Art simply builds on those natural rhythms we perceive all around us, and in which we participate: the wind in the trees, shadows that come and go, the flight and locomotion of nonhuman animals. All demonstrate the pattern of tension and release where movements oppose each other; one waxes, the other wanes. ‘Then the operation is reversed, not necessarily in equal periods of time but in some ratio that is felt as orderly.’ That is the process of being alive, the process we experience every day as rhythmic. Dewey names only the obvious rhythms; he admits that his picture is still incomplete. For there are ‘minor coincident changes of expansion and contraction that are going on in every phase and aspect’ of life (Dewey, 1958[1934]: 155). There is no better description of what Condon came to find in ordinary, spontaneous, human conversation. Rhythm and process were the pillars of his theory. And his technique with films allowed him to see just how small and pervasive were these ‘minor coincident changes’.

Condon would eventually oversee production of 16 mm sound films for his own use. He was especially fond of films made at higher frame rates like 48, 72, even 96 per second, which made movement boundaries easier to detect. At first, however, he was given the leftovers from the Brosin group's corpus. The basic data used in The Natural History of an Interview was a 10-minute film by Gregory Bateson. In it were Bateson and the woman named ‘Doris’ as they sat on her sofa in 1956. She had sought Bateson out after hearing him lecture on the subject of communication in unhappy families. She, too, had an unhappy family. She then agreed to let him film her at home with her family and, in one instance, outside the home. All of these he made available to Brosin and the others, who eventually settled on one for full analysis. This scene on a sofa (GB-SU-005) was recorded continuously except for short breaks to reload the camera. Sound was taken separately on magnetic tape and synchronized later, also bridging the gaps in filming. The behavior was unrehearsed, informal, spontaneous, and the style of filming was rather like verité (Engelke, 2021; Watter, 2017). The importance of this film of Doris cannot be denied; it formed the professional vision of early interaction studies; but it should not overshadow the other films of Doris that Condon and others would use just as intensively.

Of these the most important are the therapy and dinner films, also made over the course of 1956. ¹³ The therapy film (GB-SU-008) is set in the office of Doris's doctor. It records an entire half-hour of therapy. Bateson is briefly seen at the film's beginning; he adjusts the camera slightly before saying goodbye. ‘Bye-bye’, Doris tells him with a distinct note of pique. He leaves, and she immediately starts complaining about the camera's presence. The time spent in set-up has cut into her session, and this on a day that she could really use the time. It had been a bad night. Her husband was away, and their son required barbiturates in order to fall asleep. She worried about the dosage; she worried he might die. She needed her husband so badly last night and when he wasn’t there, she brought herself to masturbate. She had never done that before, not ever. She felt herself losing her grip on reality; she wanted to tear down the walls of her house.… And more of the same for 33 minutes as her therapist sits in silence, making only the briefest comments. They face one another across the doctor's desk, Doris in the foreground and he in the middle distance. The camera is placed at a 45-degree angle, and its position is fixed for the film's duration.

The dinner scene (GB-SU-001) begins on a shot of the son as he plays with a fort he has built out of cushions. The camera pans with him over to a table, where he sits between Doris and ‘Larry’, his father. The parents sit in profile; the boy faces the lens directly. The film is six minutes long and contains four ellipses. Larry passes the bread around, then helps the boy cut up the food on his plate. Doris drinks from a stein and looks on in silence. Eventually, she eats, without much enthusiasm. She and her husband talk in low voices. The camera zooms in on the boy with a slice of bread, then it zooms out to include all three again. The man gestures off camera, but his comment is inaudible. The boy nearly finishes a tall glass of milk, puts it on the table, and loudly exhales. He has said very little since the filming began. His mother looks with satisfaction from him to the camera, to which she addresses the following remark: ‘I think you all should come around every night, we never have had a dinnertime like this in months!’ The joke seems to revive her somewhat. A kind of current runs through her; her smile is brilliant; she gives her words emphatic rhythm, and she can barely suppress her laughter. Her husband laughs too, even before she does. But the boy does not laugh; rather, he whimpers.

From 1962 to 1964, Condon and others focused on the therapy film. Typescripts in the archive indicate plans for a book to be authored by Condon and others: E. Joseph Charny and Felix F. Loeb, psychiatrists; and Harvey B. Sarles, an anthropological linguist. It was to be a kind of sequel to The Natural History of an Interview, but only Charny's and Loeb's contributions were published. ¹⁴ Condon's drafts indicate a mind in transition. His ideas are confused and his method unrigorous. ‘Some aspect of behavior which appears to have a high degree of similarity in its repetitive occurrences is tentatively selected’, such as Doris's fist. ‘It is “bracketed” out conceptually and methodologically in order to search for criteria with which it may be correlated.’ ¹⁵ The fist or fist-like gesture occurs alongside almost every reference to Doris's child. As a sign of aggression, it expresses those feelings she cannot avow. Condon's view, though behavioral, is still largely Freudian. It amounts to a rephrasing of Freud's basic thesis that hang-ups are displaced onto the surface of the body. Hence Condon focused on stereotyped gestures and their relation to verbal content, as psychoanalysts tend to do. He gave little thought as yet to the question of units; of what enabled him to speak of this or that as a unit.

Apparently McQuown, on a visit to Pittsburgh, suggested Condon look at a film of three people. ¹⁶ It would pose novel questions not found among dyads. The film, of course, had to show the three together within the same frame for an appreciable duration. Bateson's film of a family dinner met all requirements, and Condon spent the better part of 1964 with it. He looked at it, he claims, for a year and a half, every day for four or five hours a day. Actually, he looked only at the same five-second segment: the time it takes Doris to make her joke about dinner. These numbers sound absurd, but there is no reason to disbelieve him. He wore out 139 copies in this long process of ultra-close reading. ¹⁷ What emerged was a new conception and a powerful technique. The search for the smallest units of behavior consumed him. Whatever interest he had in Doris's problems apparently left him upon completion of the work. More truly, perhaps, the thousands of hours he spent on a fragment made the context irrelevant, dissolved it into bits. Every displacement of every limb of every person would have to be noted in the course of study. The changes had no meaning, but they were not random; they hit every beat of an emergent phonetic pattern.

‘I think you all should come around every night, we never have had a dinnertime like this in months!’ When Doris says ‘around’, for instance, her head moves up slightly at the onset of a. On rou it moves back and further up slightly; and it ends with a downward movement at the start of nd. Other changes happen in the mouth, trunk, elbows, and shoulders, changes that mass together at the points her speech allots them. The speaker's body dances to the tune of her own voice. Nor is that all. The body motion changes in husband and son also seem to coincide with her voice segmentations. This includes the husband's fork going to and from his plate (Condon and Ogston, 1967: 225–30). A professional choreographer could not have timed it better, and in fact would not need to, for the phenomena of synchrony are really involuntary: they are epiphenomena of speech's reception. So Condon argued after 1966. Perhaps this is why the child cannot help dancing even as he whimpers from behind his glass of milk.

‘The order in the material must force itself on the observer’

The concept of synchrony as Condon defined it was stamped by the personalities of user and instrument both. A concept, however, is not yet a theory, which must at least propose an account of causation. An even stronger theory will try to identify the ultimate end or function of the causal chain described. Such and such conditions give rise to some others to further or sustain some even greater process. It was not enough for Condon merely to say that something called synchrony occurred and could be seen. Thus he enlarged its concept, operationally defined, into a theory of human conduct and organic function. Less well developed but equally intriguing is the epistemology of inquiry that emerged alongside it. The form of the instrument, its daily operation, would ultimately contribute to both of these projects.

The coordination of speech and gesture described as self-synchrony is supposed to tell us something about the nature of the organism. It reveals that the human being is a unified organism, a creature patterned rhythmically by a single, central process. The structure of behavior is probably determined by the central nervous system acting as a pacemaker. The psychology of prior decades had already opined as much, and Condon's own dabbling in bioelectric measurement seemed to support the thesis of basic self-synchrony (Condon and Ogston, 1967: 232–4; Lashley, 1951). Interactional synchrony was more controversial, especially in Condon's ‘very radical’ view of it. ¹⁸ One critic outlined two possible objections. The first was a strictly technical objection on the grounds that a listener could not respond so quickly. Since Condon assumes that it is the speaker's speech that programs the changes in listener behavior, he would have to admit a delay in sound processing before the speech could be reflected in the listener's kinesic pattern. If he nonetheless insists that a speaker's and listener's changes characteristically occur together with split-second precision—that their onsets are timed and shared within a single frame of film—he would also have to posit a latency in processing of under ten milliseconds, a sheer impossibility (Rosenfeld, 1981: 77). But this is exactly what Condon did maintain. Interactional synchrony was, for him, evidence of a rapid tracking mechanism in the processing of speech. And it confirmed to the speaker that the listener was listening. Condon used the word ‘entrainment’ to name this tracking process, which would have to be instinctive and wholly unconscious. It even affects the listener's blinking. One can get bored or stop up one's ears, but then one is no longer listening at all.

Most people would admit to their own predictability, insofar as they have traits of character that define them; but they would be appalled to know that determinism infiltrates the very smallest details of nodding and blinking. At the very least it might prevent them from seeing their synchrony as something to endorse without reservations. Perhaps dyssynchrony points to freedom, autonomy, independence—things to be fostered in children and students (Rosenfeld, 1981: 94). Condon also foresaw this second objection. As a philosopher once invested in ideas of free will, he himself had felt ‘revulsion’ when looking at films; they forced him to acknowledge ‘how strictly determined behavior can be’. ¹⁹ The revulsion still echoes in his frequent comparison of people interacting to puppets on strings. At some point, somehow, he overcame his revulsion to see the great beauty of what he discovered. ‘For me all human behavior is now more and more beautiful’, he told a group of colleagues in 1982, because people move together as in a kind of dance without any effort or intention to do so (Condon, 1982a: 67). They display nature's unity, harmony, balance for the benefit of anyone who takes the time to find it.

Their dance, moreover, is evidence of nature's fullness; no vacuum separates speaker from hearer. Space is never empty despite appearance to the contrary, and all is knit together by rhythm and structure. ‘There is a continuity of order in the communicational process’, from the formation of thoughts to their expression as speech. The mouth starts to move and the air around it vibrates. ‘The ear drum of the listener oscillates rapidly and synchronously in relation to those impinging sound waves. Cells take up and carry the order further through isomorphic neural assemblies. The translation begins’ (Condon, 1976: 315). Interactional synchrony is but a part of that translation. The distinctly metaphysical cast of Condon's mind, his faith in a rhythmic principle running through nature, comes to the fore in statements like these. He was, from student days, a follower of the philosophy of F. J. E. Woodbridge, and Woodbridge maintained in An Essay on Nature that we can know nature because we are her children. Nature holds us in her hand, or even at her breast; knowledge is a product of closeness, not distance (1940: 18, 42). The highly gendered language of Woodbridge's book, which Condon will quote from again and again, may well have shaped the future of his research. In the early 1970s, at Boston University, he turned his attention to very young children. The research was sensational because it claimed to find synchrony in children as fresh as 12 hours old (Condon and Sander, 1974). The voice of a mother appeared to give form to the wriggling, squirming movements of her new baby. Superficially, the movements were random. ‘Sweeping’ across the filmstrip told an entirely different story. The synchrony of infants was almost as good as an adult's, and Condon later claimed to find it 20 minutes after birth (Condon, 1979: 139). Perhaps the fetus moved in synchrony with the mother's voice in utero. That would be very useful for language acquisition; it would carve language deeply into emergent flesh. In any case, the child is born into a world for which she or he is well prepared indeed. We swing in the rhythms that precede and enfold us, and are thus in a position to have knowledge of them. ‘There is no “between” in the continuum of order’ (Condon, 1980: 56).

But the discovery of synchrony was still slow and painful, despite the continuity of knower and known. Two years at the projector did not let Condon see it. Only two more with a five-second filmstrip allowed the fact of synchrony to show itself to him. When he saw it, however, he could never unsee it, and over the course of his career he saw little else. Film after film showed much the same thing. It became a permanent part of his conscious perception. ‘Retrospectively’, he writes, ‘what seemed to be involved was that in order to see the patterns in the material, changes had to occur in the observer’. This is his conclusion after 20 years of research. But even this is not strong enough to really express what happens in the course of inquiry, according to Condon. Rather, he continues, ‘order in the material gets through to the observer and this seems to involve a change in the observer's perspective. The order in the material must force itself on the observer’ (Condon, 1982b: 20–1).

The order in the material must force itself on the observer. That is a truly remarkable claim. There are various ways to interpret such a statement, starting with the one that Condon preferred. He, the investigator, is a part of nature. So are his instruments; so are his concepts. All of this is subject to the same laws of nature as his object of study, which is human behavior. How could his inquiry not lead to truth eventually, provided he is open to what nature has to give? Thus he looks and looks and looks at pieces of nature—pieces of time and space recorded on film. He recalls that the pieces are artifacts of study that lead him, nonetheless, to the structures that predate him. He feels his way back to the bosom of nature. He trusts her to teach him if he is patient. Her natural rhythms sink into his being as a mother's voice patterned him once upon a time. So little does he do now besides be receptive that the passive and not the active is his preferred form of syntax. He did not discover synchrony; synchrony ‘was discovered’ (Condon, 1982b: 43, 52). It was discovered by someone it forced itself on, and whose perspective it changed in long hours of viewing.

There is another interpretation, which is ours and not Condon's, one based on a concept of manual enskilment (Pálsson, 1994). It delves somewhat deeper into the kinds of operations that first disclosed synchrony by mechanical means. Condon had said that any act of communication involves a translation of behavioral order. Speech vibrates the air, then the hairs of the inner ear, on through the central and peripheral nervous systems. A like process should apply to scientific observation. The recognition of synchrony would be an effect of a series of translations of behavioral order. There was order in the behavior when it was filmed, and this order has been preserved in photochemical transformations. It is transformed again within the projector, from which it goes outward to the receiving surface. It reflects off the screen and reaches the eye. Cells in the eye are activated; translation begins anew. The result may not be knowledge in 10 milliseconds, but knowledge will come to those who wait and see. Or rather to those who wait and crank, for the manual dexterity of the projectionist is no less important than visual acuity. So it is with Condon and his magic wrist. Only by sweeping again and again can the bounds of a process unit be rightly determined: first across two frames, then across three, across however many frames until a bobble appears. And when it does appear, he knows it is due to an incremental change in the performance of his hand. Likewise, he knows a bounded unit by the motions of his hand; he turns the crank swiftly and sees a smooth, sustained event. The order in the material that forced itself on the observer would force itself partly through the medium of the hand. One must move and change in synchrony with the subjects observed, and do so in concert with the machinery of observation. The translation of order from behavior to film, film to observer must include a translation of muscular order. It is suggestive that at 24 frames per second, process units are never longer than one turn of the manual crank.

There are yet other ways that order can force itself on the observer. It can do so in the process called organ projection, by which our machines become models for thinking (Kapp, 2018[1877]: 27–33). It was easier for William Harvey than it was for Paracelsus to imagine the circulation of blood through the body, for Harvey had the image of the water pump before him while ‘Paracelsus got his water from wells’ (Pachter, 1951: 39). Human organs and functions are reified in machinery, are projected unconsciously into the form of tools; and these technical objects then provide knowledge of the organs or functions that they resemble. Thus a projector shows an image, but is also itself an ‘image’: its very operation is suggestive of ideas. It impresses itself on the sensibilities of those who spend hundreds of thousands of hours by its side. Its two reels move forward, and move each other forward, in the tightest synchrony one can imagine. Sprockets grip the film at regular intervals to advance it both into and out of the projector gate. Entrainment is apparent as one reel turns the other by means of a filmstrip passing over both. There is no vacuum between them but rather a constant traffic that vibrates and hums. From supply reel to take-up, the projector performs its translation of order. Moreover, its continuity is divided by the action of the shutter. Thus it has continuous and discrete-like aspects, and is clearly self-synchronous in all its parts. And if we shift our gestalt a little and see the reels as separate, it becomes the living image of interactional synchrony. Nothing can occur in one of the reels without corresponding changes in the movement of the other. A projector, in any case, gives constant confirmation of the entrainment process in its most inexorable form. Every day Condon saw how his own movement made other people move on a screen some feet away. They moved as quickly or slowly as his hand allowed them, and they stopped where he stopped, on the same few frames of film.

Our discussion is incomplete without mention of improvements Condon made to this projector over the years. The Bell & Howell model, good as it was, offered some resistance to really fine-grained work. Frequently, it tore the filmstrip's perforations due to backlash or recoil in manual transport. Condon abandoned the built-in supply arm and erected a new one to the projector's left. He also describes a ‘trough’ to help guide the filmstrip from this freestanding holder to the projector gate. ‘This eliminates pressure on the film when going forward’ (1970b: 52). He made more improvements in the 1970s. Actually, the original projector disappeared completely and was replaced by another in a new arrangement (see Figures 5 and 6). In a picture from 1975, we see two reel arms or towers on a shared wooden base (‘Research in Progress’, 1975: 25). The reels face each other in a straight line like the wheels of a car when seen from one side. Both towers are equipped with an ordinary household doorknob in lieu of the standard crank they come with from the factory. Apparently the doorknobs gave better control at the one-, two-, three-, or four-frame level. By means of these knobs, one advanced the filmstrip, first through a film viewer and probably a film sound reader. The latter was certainly present by the mid-1980s (Crook, 1990: 25–8). It was as if the projector had been carved at the joints and its various members laid out on the table surface. Yet the delicate dance of synchrony, the unity of sound and image, was preserved despite the separateness of these components. Actually, the arrangement would have emphasized what Condon always knew, that sound and image in a film are ultimately separate. And not only separate, but also out of phase, because the optical sound is printed some frames ahead. Only the projector can restore the original unity of the event as the filmstrip runs through it. Hence the sound and image readers can catch the strip at different points and still create a sync-sound image on the screen before us. Every filmstrip seemed to suffer from a basic fault or defect that the projector corrected in its operation. This also had an analog in Condon's later thinking, which finally had to reckon with human pathology.

OPEN IN VIEWER

Figure 5.

William S. Condon with film equipment, Boston University. Photograph by Bradford F. Herzog (‘Research in Progress’, 1975).

OPEN IN VIEWER

Figure 6.

Diagram of equipment for microanalysis. Courtesy of Jere Lawrence Crook III (Crook, 1990).

Order and disorder

After completing his studies of synchrony in infants, Condon focused increasingly on the study of disability. In 1974, he received a grant specifically to make and microanalyze films of autistic children (Condon, 1975; Dietz, 1974). He had, for a long time, compared films of ‘normals’ to films of other people with various ailments—depression, schizophrenia, Parkinson's, Huntington's—all of which manifested signs of dyssynchrony. One side of the body might lag behind the other, or perhaps it was just one finger that missed the beat of speech. In any case the lack of basic self-synchrony also meant the foreclosure of interactional synchrony. Such were the behavioral manifestations of pathology. They were not, however, Condon's major preoccupation, and they did not force adjustments at the level of theory. Rather, they sat awkwardly alongside the theory of synchrony. Yet anyone could see that the failure to synchronize in millions of people would have to be acknowledged and accounted for eventually. In an earlier era, such as the medieval, those signs of disorder might be taken for signs of evil. The task of the modern scientist is to reduce all disorder to another form of order, and this is what Condon tried to do from 1970 onward.

The early 1970s were a time of transition in the concept and therefore the treatment of autism. The psychoanalytic interpretations so popular in the prior decade were giving way to more strictly neurological explanations (Silverman, 2012: 93–124). One can see this shift in miniature in Condon's own work. The first autistic child whose films he looked at closely was the three-year-old Amy in 1967 (Adler, 1968: 54–6). She appeared to show more interest in objects than people and was more likely to move in synchrony with the sounds made by objects (see Figure 7). She ‘displayed a range of bizarre gestures … representative of retained bits and pieces of the child's relation to her mother. For example, the mother would hold the child tightly by the forearm so that the child's wrist dangled limply. Later, when not with the mother, the child would hold her arm at the same angle and let the wrist hang limply.’ It was ‘as if a part of the body belonged to someone else’ (Condon and Brosin, 1969: 833–4). The cold and controlling mother, whose coldness and control had deformed the child, was a major motif in the popular writings on autism (Zeavin, 2021: 66–73). Scientifically, though, it was dead on arrival and it disappeared entirely from Condon's later writing. He had come to view autism as a sensory processing problem with particular difficulties in the area of speech reception.

OPEN IN VIEWER

Figure 7.

Basic synchrony of ‘Amy’ in response to sounds of tapping; note, however, the dyssynchrony in the frames that follow (Condon and Brosin, 1969).

There is, nonetheless, a continuity of language in the basic description of autistic behavior. Its metaphors, similes, and figures of speech are most often drawn from the world of machinery. ‘A crude comparison might be that of a finely-tuned automobile that moves smoothly or swiftly ahead’, writes Condon. ‘Conversely, a badly tuned or out-of-tune automobile jerks and bucks in fits and starts when the driver steps on the accelerator’ (1982b: 103–4). So it is with autistic subjects, whose bodies move not smoothly but often in ‘jumps and jerks’ (1984: 49). The robotic and uncanny quality of autistics—their appearance to observers as defective machines—has been remarked on ever since the condition was identified. Bruno Bettelheim's account of ‘Joey: A “Mechanical Boy”’ is only the most extreme version of this mechanical discourse (1959; 1967: 233–339). Hans Asperger spoke in 1944 of the ‘intelligent automata’ he saw at his clinic. They moved rigidly and stiffly, were clumsy and abrupt, ‘never able to swing in the rhythm of the group’. They hardly ever laughed or smiled. Nearly everything they did was ‘ugly and angular’, expressive of ‘disharmony’ (Asperger, 1991[1944]: 57–8, 80). It would be easy to dismiss these views as merely intolerant. It is harder to ignore the world's best-known autistic when she compares her behavior to ‘a tripping circuit breaker’ (Grandin, 2020[1995]: 34). Again we sense the power of organ projection to determine the limits of self-understanding.

Public discourse on autism is hotly contested. Generalization is difficult, but one thing seems certain: that today's non-autistic person is likely to see autistics as more ‘quirky’ than defective in any fundamental way (Bumiller, 2008). There are, for one, many more of them than before. The expansion of the category into a spectrum has allowed it to encompass an ever-growing rank and file (Eyal et al., 2010). Idle chatter with friends often leads to speculation about who among one's peers may well be ‘on the spectrum’, and many proudly claim this label for themselves. For we have different ideas about autism now: special capacities, neurodiversity, an especially creative use of language and design (Evans, 2017: 430–3). But deficit was the only model available to Condon when he set about his work in 1974. Once more the projector decided the issue. It both disclosed the problem and modeled the solution. With its aid Condon saw, apparently correctly, that one aspect of autism is a delayed response to sound. The seemingly random quality of autistic ‘jumps and jerks’ resolved into order of a new kind. They were aspects of synchrony with a delay. Each movement was in fact the response to an outside stimulus with a pattern of delay specific to every child. Not only did these subjects appear to respond too late; they responded too strongly and too many times, as if the sound input continued to reverberate and to be perceived long after it had ceased. The naked eye could see it happen; it was happening all the time. ‘It occurs even in relation to the sound of their own voice’ (Condon, 1985: 145). He refers to a film of an autistic teen exposed to certain sounds that are followed by silence. In this case the sounds are recorded double-taps played from a speaker or a nearby tape recorder. The teenage girl shows dyssynchrony every single time. At first she moves her left side at the onset of sound; so her left side is synchronous and therefore normal. She sits quietly with her hand to her mouth for a moment. Then her head jerks several times, as if looking for the source of sound. ‘The very surprising thing is the precision’ (ibid.: 149). The first jerk occurs 27 frames after the double-tap. The second occurs another 20 frames later. If one looked at every instance of the girl's response to sound, one would see that she always showed these responses. She showed them not just in a general way but in the same intervals, with single-frame precision. One could even start predicting on which numbered frames she would show the behavior, so regular was it. Twenty-seven frames after onset of sound, sure enough, one sees the startled figure begin to turn her head. Twenty more frames and she turns it again, almost two seconds since the sound had occurred. The disorder was really just another form of order and this order in the material must force itself on the observer (see Figure 8).

OPEN IN VIEWER

Figure 8.

Delayed entrainment with multiple-orienting response. Original footage not extant (Condon, 1985).

There comes a time, however, when the order in the observer must force itself on the material. One wants to correct what nature has impaired: to smooth out the wrinkles in her all-embracing fabric. ‘He hopes to design instruments to compensate for the bad phasing’, a journalist wrote of Condon in 1978. ‘Another possibility is to train autistic children to ignore the extra sounds’ (Esper, 1978). Whether he pursued either option is unknown at present. He published very little in his later years as he devoted more attention to his counseling practice. A sentimental memoir by one of his patients’ mothers reports that he focused on the autistic use of metaphor (Ziegler, 2010: 35, 44). Film and film analysis were irrelevant to this pursuit. Still, he exhibited the old machinery at conferences occasionally and expressed his approval when someone wanted to learn to use it. ‘There is so much that could be done’, he wrote when he retired. ‘I have never understood why this stuff has never caught on’. ²⁰ He looked forward to a time when film would be used more widely. Early signs of dyssynchrony could be picked up right away; different disorders would show different patterns. The method was time-consuming and therefore impractical, but it could also be expedited if one knew what to look for. Advances in computing and pattern recognition seemed poised to make that happen if anyone was interested. He was deeply sympathetic to the needs of the autistic child, whom he once called ‘a ghost in his own distorted world’ (Dietz, 1974). That world was full of friction, of constant mechanical backlash, and everything should be done to facilitate transport through it. And if the prostheses ultimately came to nothing, the projector gave proof of concept in its daily operation. It too could be fixed and made to run smoothly. Tensions, contradictions resolved themselves in it. Here human beings were cut up in frames; here the frames blended until there were no frames to see. Here sounds were printed some frames ahead of pictures. Here they were also brought back into phase. The film flowed through a setting just designed for its needs. New reel stands, a trough, doorknobs for handles, a device to read the picture, and another to read the sound—everything clicks together when we now turn the film. Everything is synchronized within the whir of gears. What machines have disassembled, they put back together, in the fullness of behavior of the moving human figure. They can even correct for defects in the original.

The ultimate proof lay in a trick of printing by which the original soundtrack was delayed by some interval. That is, the optical sound would be moved out of phase with the series of images from which it was always separable. New relations of sound and image could thus be established, relations that depart from the structure of real events. In the film mentioned earlier, Condon saw the girl entraining to discrete acoustic stimuli with two delayed responses. Just over one second after sound was emitted, she jerked her head to find it, then again a second later. She repeated this behavior with each successive sound. An observer can watch and record these delays, tally their incidence, and produce a graph or histogram to sum up the results. But a more compelling proof, compelling because immediate, was to alter the film to reflect the girl's experience. So the film was reprinted with the soundtrack moved back exactly 27 frames along the entire filmstrip. Then the soundtrack was printed once more on the filmstrip with an additional delay of 20 more frames. ‘If the hypothesis was correct’, if the ratio of delay was constant, ‘it should be possible to show [her] moving and looking around in precise synchrony’ (Condon, 1982b: 161). For the girl is really synchronous in her own way. She entrains to the sounds as her autistic brain receives them. Now a double-tap is heard on the reprinted film, and her figure moves in lockstep with that percussion. The sound occurs again, and she moves just so again. And so on and so forth every time we hear the sound. The reprinted film is of course an illusion; those sounds were not there at those times in the world. Or rather they were, but only for the child and not for the observer. Yet if one reduced her latency of sound and speech reception, she would appear entirely synchronous as she does here. It was only a trick of film, but was it not suggestive? If only real life could imitate the film. The young girl would ‘dance’ instead of feeling pulled apart. She too would join in the dance of life itself. She would learn to move in synchrony with the movements of others in such a way that those others would feel in synchrony with her. No longer would the sight so distress an observer. One could look upon it and deem it to be good. There, on the screen, disorder is reduced to order, jerking to dancing, ugliness to beauty. A wrinkle in nature's fabric has been made smooth. The order in the projector that forced itself on the observer is summoned up and transferred back onto the material. Thus is unity achieved in the continuum of order: the synchronous sound film has become an ideal of unity. And the more one approaches its technical perfection, the more one appears as ‘a unified human being’.