Feeling Crowded

An Emergent Model

The theory of emergence as it pertains to aggregates of living beings can be traced back to mathematician John Conway and “The Game of Life.” This game, which Conway developed in 1970, consisted of black and white tokens placed in a grid. Each token could change from black to white depending on the number of adjacent black or white tokens. A black token which bordered two or fewer cells containing black tokens turned into a white token, it “died of loneliness” to use the game’s language, whereas a white token bordered by three or more black token “ressurected,” that is, turned into a black token. When the game was plugged into computer simulations of larger and larger grids, the game yielded ever changing, pulsating, fractal patterns (Yong 2013: 1).

Almost two decades later a computer animator, tiring of the laborious process of having to animate large flocks, herds, schools, or other aggregations of animals on an individual, took up this notion of complex patterns generated from simple sets of rules to program a landmark computer application titled Boids. This program used three simple steering behaviors to simulate the motion of coordinated aggregations of animals: alignment, attraction, and repulsion. Specifically, he achieved remarkably lifelike group behaviors simply by defining units which carried out three simple behaviors in response to other units within a set perceptual radius. These units (a) aligned themselves with the average heading of other, perceptually available units, (b) were attracted to the average position of these perceptible units, and ( c ) steered toward the average position of these other units (Reynolds 2001). Remarkably, Boids, given certain relative intensities of these forces, could simulate a number of collective movements ranging from that of a flock moving along a migratory path to ones resembling circling vultures or ritual circumambulations and others resembling a crowd of shoppers milling about or herds grazing (Yong 2013: 1).

After Reynolds was invited to present his work on Boids at the first transdisciplinary Artificial Life workshop in 1987, the emergence model with respect to crowd behavior began to garner growing interest. Most notably, the Hungarian and German physicists Czirok Vicsek and Dirk Helbing used mathematical techniques derived from their expertise in fluid dynamics to apply the parameters of the Boids model to various collective animal movements, including and especially those of human beings (Yong 2013: 1). Specifically, through the study of video footage and long-exposure photography of crowds in everyday and panic situations, the use of ethnographic informants, and the study of the patterns of footprints in snow, they observed that the passage of crowds was quite similar to that of granular media in forms as variable as hourglasses and sedimented riverbeds. (Helbing, Farkas, Molnár, Vicsek 2001: 7). They ultimately managed to develop simulations in which simple, perceptually limited steering behaviors akin to those used in Boids showed individuals engaging in spontaneous, crowd behaviors observable in everyday life, such as the formation of directional lanes on metropolitan sidewalks, variable bursts of groups from directionally competing crowds groups moving through bottlenecks, self-organized roundabouts at pedestrian intersections, and so on (Helbing, Farkas, Molnár, Vicsek 2001: 12-14).

The proof of principle of these theories of emergence, that is, the relation between simple, microsocial parameters and macrosocial movements, was only achieved in the late 2000s. Familiar with the work of Helbing and Vicsek, biologist Iain Couzins became interested in exploring whether such microsocial parameters could explain the still mysterious “plague of locusts;” no one had yet determined how a disorganized throng of locusts, upon reaching a certain threshold of density, organized itself into a ravenous swarm. Having constructed a large, ring shaped “locust accelerator,” Couzins was perplexed by a mounting number of missing individuals. After reviewing video replay of the locusts, he discovered that locusts were prone to cannibalizing one another from behind. The simple threat, given a threshold density, of a bite delivered to a locusts posterior created a social repulsion between the locusts that, on the macrosocial level, yielded the patterned organization of the swarm. (Yong 2013: 1) Couzins went on to confirm this finding by denervating the locusts’ abdomen, preventing them from sensing threatening bites from behind and consequently suppressing their ability to swarm (Simon 2009).

Since Couzin’s pioneering research at Oxford he has moved on to found the Collective Animal Behavior Lab at Princeton. Among other projects, his lab has observed a species of minnow, the Golden Shiner, which moves by tracking areas of lower light intensity, presumably because such shady places shelter them from predators. The lab used projectors to create a shifting tapestry of shadow and light onto a tank of the shiners and observed that the accuracy with which they tracked the shade increased in proportion to the size of the school. Hypothesizing that tracking shade depended on simple, microsocial variables – the tendency of the fish to swim slower in the shade and faster in the light and the principles of attraction, repulsion, and alignment à la Boids – they managed to develop algorithms which near perfectly modeled the emergent behavior of the fish (NPR 2013).

Interestingly, Andre Gallup of Couzin’s lab has recently begun to conduct analogous experiments on human crowds. Gallup used digital cameras and powerful video cards designed for computer games to automatically determine the direction of individual gazes, crowd sourced employees to monitor the feed, and groups of actors of various sizes to prompt crowds to look at the hidden cameras by themselves looking in that direction. The experiment was a high tech rehash of Stanley Milgram’s 1969 “obedience experiment,” but conducted in a more realistic, in vivo setting. In contrast to Milgram’s work, which largely echoed the conclusions of 19th century crowd theorists in asserting that, at a certain threshold, a near-complete deindividuation and herd mentality comes into being, Gallup’s research showed that the miming of perceptually adjacent individuals’ line of site was present, but increased at a steady rate, that is, there was no clear divide between individual actions and some sort of crowd mentalité. The group also discovered that the position of one onlooker with the respect to another affected the likelihood that the direction of their gazes would align:

Individuals with trajectories leading them to walk behind the visual orientation of the stimulus group exhibited a higher propensity to follow the gaze of the stimulus group. This finding suggests that gaze-following under these conditions is not caused by social pressure or some form of obedience (Gallup, Couzin, et al 2012: 5).

While valuable as a proof of principles, we might observe that this phenomenon was largely anticipated by sociologist Erving Goffman in his 1971 book Relations in Public, where he noted that, when moving through a crowd, “the scanning area is not a circle but an elongated oval, narrow to either side of the individual and longest in front of him, constantly changing in area depending on traffic density (Goffman 1971: 12).” Moreover, the team seems to be in dire need of an anthropologist when they express frustration with respect to their attempts to draw attention using actors engaged in “suspicious behavior,” which resulted in wide variations in behavior depending on gender and other social and economic variables (Gallup, Couzin, et al 2012: 6)

The value of Gallup’s direct application of the emergence model to the study of human crowds remains to be seen. It seems doubtful that methods so reductionist will ever yield more than vague confirmations of long ago intuited principles with respect to crowds. Yet the work of Couzin’s lab has been invaluable in substantiating emergence as a generative principle of patterned aggregations of animals, including human beings. This substantiation is important to our treatment of crowds insofar as it frees us from necessarily seeking out some heteronomic agent or imperative in order to explain their coordinated actions. From the 19th century onwards, the ability of crowds to act as coordinated, even animate systems was explained by positing a geist-like crowd mentality which made them susceptible to particular forms of suggestion, hypnosis, ideology, etc. Even in Goffman, who closely anticipated the techniques employed by the emergentists such as the study of video evidence and footprints in snow, continued to defer to a hypothetical, tacit “traffic code.” This is evident, for example, when he discusses the formation of passing lanes in bidirectional pedestrian traffic, supposing that the dividing line would be somewhere near the center of the sidewalk with the directionality of the lanes mapping loosely onto that of regulated automobile traffic. Helbing and Vicsek, however, have shown that this is not in fact the case:

The conventional interpretation of lane formation assumes that pedestrians tend to walk on the side which is prescribed in vehicular traffic. However, the above model explains lane formation even without assuming a preference for any side. The most relevant point is the higher relative velocity of pedestrians walking in opposite directions. Pedestrians moving against the stream or in areas of mixed directions of motion will have afrequent and strong interactions. In each interaction the encountering pedestrians move a little aside in order to pass each other. This sidewards movement tends to seperate oppositely moving pedestrians which leads to segregation (Helbing, Farkas, Molnár, Vicsek 2001: 13.

Yet all this does not discount the literature dating from the 19th and early 20th centuries. Rather, as we shall discuss in a subsequent section on the principle of mimesis, many of these older theories skirted this autonomy of crowds, only falling back on heteronomic diktats for lack of a model that could ground a more distributed conceptualization of agency. The principle of emergence allows us to revisit and complement the depth and breadth of these theories and opens up new avenues for anthropological inquiry.

In this chapter, we will look into a single one of these simple parameters that, interacting en masse, result in the self-organizing, patterned behaviors of crowds. Specifically, we will be looking into repulsion, that is, the force that holds a crowd’s constituative units apart from one another, allowing for movements, their ability to flow in tactical maneuvers and through confined or complex spaces.

One or Several Units?

Before embarking on our consideration of repulsion, however, we should ask ourselves what units we are positing as mutually repulsive. In the models put forth by Helbing and Vicsek who, like economists, need to artificially simplify their field to a certain extent in order to make it mathematically accessible, the unit in question is clearly the biological individual (or, if we were to take a more critical-Foucauldian route, the individuated subject). Because we are not seeking to further any sort of simulation or predictive model, however, we can afford to be somewhat more empirical and fine grained in our analysis. We might observe, then, that crowds are not entirely composed people who show up alone. While there are many such “singles,” as Goffman terms them, in a crowd, there are at least as many groups of friends, acquaintances, or allies who interact as a unit against outside individuals and groups, that is to say, the crowd at large. In other words the intra-actions of these subgroups operate according to an entirely different field of forces than that of the interactions which they navigate, like lone persons, with respect to the crowd. In the language of Elias Canetti, the active unit in such situations is neither the individual nor the crowd, but the pack.

In introducing the pack, Canetti states that he is “deliberate opposing all the usual concepts of tribe, sib, clan, with a different kind of unit.”

Those well known sociological concepts, important as they are, all stand for something static. The pack, in contrast, is a unit of action, and its manifestations are concrete (Canetti 1962: 94)

What Canetti means here is that the pack designates a constellation of bodies whose headings closely mirror one another and who are constantly moving from the group’s center to the periphery and back again, all the while remaining invisibly tethered (Canetti 1962: 93), Indeed, the pack is more purposeful in its movements that the individual, where more focussed and single-minded routing as much as distracted raptures result in a tendency to meander or wander in a way that perhaps more closely resembles the movements the crowd writ large than those of the pack. Most importantly, however, packs are gated units.

Insofar as packs do grow, growth takes place in discrete quanta, and by agreement of the participants. A pack formed from a second group may come across the first pack and, unless they fight, they may join forces for temporary purposes. But the separate consciousness of the two quanta will be preserved (Canetti 1962: 94).

Packs, in short, are neither as closed as the individual nor as open as the crowd.

Canetti, however, sees these packs as a prototypical form of the crowd. His account of mass society consists of an interesting teleology that inverts the previously proposed relation between “primitive” aggregations and the crowds of industrialized metropoli. The crowd, instead of being a form regression or anachronism in evolutionary time, is the telos of packs: it is the culmination of an always present desire to “become more.” In rituals, packs press in and turn away from the emptiness of the wilderness, simulating the visual density of the crowd while moving as one to simulate its auditory density through the principle of rhythm and resonance (Canetti 1962: 31-34, 95-96). While this reversal is interesting in its own right, to follow its evolutionist claims would be to make Canetti’s theorization of packs useless to our purposes. Accordingly, we might follow Gilles Deleuze and Félix Guattari reading of Canetti’s notion of packs, where the difference between pack and crowd (what Deleuze and Guattari call mass) is not a question of epoch but a question of scale. “There are only multiplicities of multiplicities forming a single assemblage, operating in the same assemblage: packs in masses and masses in packs (Deleuze and Guattari 1987: 34)¹.”

Goffman, in Relations in Public, develops a similar distinction between what he calls singles and withs. While he writes that withs “maintain some kind of ecological proximity,” he tends to conflate the common, constellational movement of these units with circuits of informational, linguistic flow. The withs’ mutal proximity “permits easy conversation and the expulsion of nonmembers who might otherwise want to intercept talk.” He notes that “the members have to exhibit ritual concern when joining and withdrawing” from the pack, yet he assimilates this ritual concern to the hierarchy of the spokesman and the spoken-for, noting that that the constituents of the with can “be expected to be deferred to by outsiders who would make contact with one of the members (Goffman 1971: 19).” This conflation relates to a broader tendency in Goffman to concern himself with an Anglo-american, middle class society that is rmore or less ficticious, or at the very least cannot be found in vivo. In such a society, verbal communication between individuals as prohibitted by a sort of tacit law: everyone is cripplingly “awkward,” to put it colloquially. A brief forray into New York City crowds explodes this assumption, however. Conversation between strangers, while not ubiquitous, is not uncommon. Groups and even individuals are as prone to loud, performative conversations and exclamations, directed towards the proximous crowd sui generis as they are to whisper amongst eachother or mumble to themselves, while packs can move in perfect silence.

While packs, as formulated by Canetti via Deleuze and Guattari, are preferable to Goffman’s withs insofar as they allow us to disengage group motility from informational networking, we must give all these thinkers their due for foregrounding the fact that, as Goffman puts it, “if we look closely at the concept of territoriality, especially the ‘egocentric’ forms, the notion of the individual ceases to have an analytically coherent, single meaning, and several different terms have to be employed in its stead (Goffman 1971: 27).” Goffman, moreover, gives us an additional term that proves useful insofar as it encompasses both individuals and packs as they relate to the crowd; he categorizes them both as “participation units.” A person, moving through the crowd alone constitutes participation unit in relation to the crowd inasmuch as a number of persons moving as a pack. In the following chapters, we will use “subject” (“intersubjective)” and “participation unit” interchangeably, the former as a shorthand for individual(s) moving as one and the latter as an intermittent reminder that subjecthood does not necessarily indicate bodily individuation.

Granules and Fluids

In the stirring opening to Elias Canetti’s opus Crowds and Power, he tells us that “the reversal of the fear of being touched belongs to the nature of crowds (Canetii 1962: 16)” This reversal, the “discharge” in Canetti language, will play an active part in our treatment of attraction and alignment in subsequent chapters. With respect to repulsion, however, the emphasis on the crowd’s startling ability to collapse intersubjective frontiers has long elided the repulsion that nonetheless continues, in alter form, to keep its participant units apart, to act as a sort of lubricant which that preserves the crowd’s serous motility. In other words, we might observe that a complete reversal of the repulsion between a crowd’s participation units would clot or otherwise impede the crowd. This is not to say that that there exists some fixed measure of personal space that constitutes a threshold across which the feelings of repulsion are transgressed or break down. Rather, the density of human bodies is proportional to various phase shifts between metastable states.

As mentioned, varying degrees of repulsion between participation units are necessary to a crowd’s capacity to flow through confined or complex spaces. In panic situations, a fire in a crowded theatre for example, the desired velocity of various participation units tends to supercede the repulsion that holds these units apart. The result, tragically, is what Helbing, Vicsek, et al term “freezing by heating” or the “faster-is-slower” effect . Specifically, the panicked desire for rapidity and concordant breakdown in repulsion results in a sort of solidity or viscosity that inhibits or retards the crowd’s passage through a bottleneck – the principle at play is not unlike that which undergirds architecture and physics of the arch. Again, this is not to say that crowds cannot tolerate any mutual pressing of bodies. As Goffman already observed in Relations in Public,“rules” of personal space, which we will take to mean a degree of repulsive force, vary as crowds grow denser:

It is a central feature of personal space that legitimate claim to it varies greatly according to the accountings available in the setting and that the bases for these will change continuously. Such factors as local population density, purpose of the approacher, fixed seating equipment, character of the social occasin, and so forth, can all influence radically from moment to moment what it is that is seen as an offense. Indeed in human studies it is often best to consider personal space not as a permanently possessed, egocentric claim but as a temporary, situational preserve whose center the individual moves (Goffman 1971: 31).

The breakdown that sometimes follows an incendiary device in a nightclub or the consumerist frenzy of Black Friday at WalMart and which results in the seemingly aberrant phenomena of human trampling is not the function of a particular threshold of a crowd’s density or a transgression of “personal space,” however. Rather, it results from the breakdown of the system of thresholds itself, that is, a transformation of a punctuated equilibrium to a purely scalar gradient of repulsion. Deleuze and Guattari echo this observation when they describe multiplicities as “composed of particles that do not divide without changing in nature, and distances that do not vary without entering another multiplicity that constantly construct and dismantle themselves in the course of their communications, as they cross over into each other at, beyond, or before a certain threshold.” Or, as Helbing and Vicsek put it with respect to bottlenecks, “the faster is slower effect occurs when the sliding friction force changes continuously with the distance rather than being ‘switched on’ at a certain distance (Helbing, Farkas, Molnár, Vicsek 2001: 17).”

Goffman’s nuanced description of personal space largely anticipates the notion of metastable thresholds to various intensities of repulsion between participant units. Yet, as we have observed, his attribution of these degrees of personal space to an operative “traffic code” made up of “informal understandings” is tenuous, even if taken in a metaphorical sense (Goffman 1971, 9). Let us consider an empirical case, one which will carry through the entire body of this chapter, the enclosed, crowded space of a New York subway car (a technology and infrastructure that was and is both materially and theoretically imbricated with this writing and this writer). When a rider absent mindedly puts his or her hand onto a stranger’s when grasping for a pole with which to stabilize him or herself, the reaction is reflex and precognitive, that is, he or she jerks his or her hand away. It is difficult to swallow that such a jerking away could be an instance where the proposition “do not put your hand on another’s in the subway” is acted upon. Neither does such an “infraction” occasion the feelings of guilt or rebelliousness that attend the transgression of a law or rule. Rather, it is as if if some repulsive force springs the two hands apart like two positively charged magnets. This nuance of this eminently quotidien and ephemeral interaction contains the difference between the old understanding of crowds as ideologically directed and the position that we are trying to advance, that they are first and foremost emergent phenomena.

If we ask what constitutes or powers the repulsion that holds between a crowd’s participant units, however, we come to the limits of the models afforded by Craig Reynold’s Boids, Helbing and Vicsek transposition of fluid mechanics, and Couzin’s ethological applications of said models. Here, in seeking to flesh out the nature of intersubjective (insofar as participation units rather than individuals are our subjects) forces of repulsion without resorting to legalistic metaphors we might turn to the work of William Ian Miller regarding disgust.

Mimetic Disgust

In his book The Anatomy of Disgust, Miller endeavors to undo the oral fixation that has overdetermined the apprehension of disgust in the English speaking world; an overdetermination that stems from the particular etymology of the English word, which ties back to the rejection of food, but which is by no means universal. Disgust, Miller argues, is neither essentially rooted in taste nor necessarily tied to interoceptive experiences of nausea: it can be experienced in tandem with olfactory, tactile, visual, and even auditory percepts as well (Miller 1998: 2-8). Miller elucidates how disgust and being disgusting are functions of a particular proximity to the disgusting/disgusted object and spread through principles of contagion or sympathetic magic in a manner that echoes the emotive “madness” described by the late 19th century crowd theorists and mass psychologists. For Gustave Le Bon, “ideas, sentiments, emotions and beliefs possess in crowds a contagious power as intense as that of microbes (Le Bon in Laclau 2005: 24),” and for Miller “disgust names... a strong sense of aversion to something perceived as dangerous because of its powers to contaminate (Miller 1998: 5).” Le Bon’s contemporary, Emile Durkheim, in his discussion of what he termed collective effervescence, advanced a similar notion of emotion subject to ramifying vectors of transmission, but used the metaphor of electrical conduction rather than microbial contamination when he wrote that “the very fact of the concentration acts as an exceptionally powerful stimulant, when they are once come together, a sort of electricity is formed by their collecting (Durkheim in Brighenti 2010: 296).

In the work of the generation of crowd theorists which followed that of the early French mass psychologists and Italian criminal anthropologists we find writings which even more closely skim the surface of crowd emergence. As Laclau notes in On Populist Reason, the more advanced notions of mimesis advanced during this period, while still closely tied to the study of hypnosis and suggestion, were largely altered by the triumph of the X school over the Y school of psychology. Whereas the former had argued that susptibility to hypnosis was itself indicative of pathological psychic state, the latter had argued that suggestion was a universally effective principle. It is this context that William McDougall,could call attention to the fact that “all of us, if our attention is keenly concentrated on the movements of another person, are apt to make, at least in a partial incipient fashion, every moment we observe (McDougall in Mazzarella 2010: 719). Far more impressive than McDougall’s work, however, is that of Gabriel Tarde. While, as Brighenti cautions, we should be cautious about unconditionally celebrating Tarde, who could be as ochlophobic, elitist, and scientistically chauvinistic as his predecessors, one cannot help but be startled at the prescience of some of his principles. Consider the following passage, for example, where Tarde describes phenomena in a manner that, if we filter out its socio-evolutionary overtones, closely maps onto a theory of emergence:

Let us put ourselves before a large object, the starry sky, the sea, a forest, a crowd, a city. From every part of such objects impressions flow that lay siege to the savage man as well as the scientist. But, these multiple and incoherent sensations are understood by the latter as a pattern of logically connected notions, a bunch of explicative formulas. How are these sensations and notions slowly transformed into laws? How could knowledge become more and more scientific? To my mind, it is the extent that we find more similarities or that, having perceived superficial, seeming and deceitful similarities, we discover more real and deeper similarities. More generally, that means that we have moved from similarities and repetitions of complex and confused masses to detailed repetitions, more difficult to grasp but more precise, elementary and infinitely numerous as well as infinitessimal. (301)

Tarde proposes “repetition, opposition, and adaption” as the three “elementary” and “infinitessimal” parameters giving rise to patterned skies and multitudes – three principles strikingly akin alignment, repulsion, and attraction, that is, the steering behaviors undergirding Boids and other simulations of collective animal behavior (300). Ultimately, however, in the absence of a distirubted model of agency, even this second wave of theorists of mimesis and collective behavior were forced to fall back upon an originary, individual source for the mimed behavior. For Tarde, ultimately, “a crowd or sect has no idea other than the one that is blown into it,” even if “the emotion that comes with this idea and diffuses with it, does not remain the same” as it interacts with other affects being routed through the crowd (Tarde in Bringhetti 2010: 302).

Contemporary Anthropology, it seems, has rediscovered and furthered these applications of conductive or contagious Mimesis. Michael Taussig, in Mimesis in Alterity, describes mimesis as an “unstopabble merging” winding its way between and through the agents of colonial encounter, “coursing back between Contact and Imitation, tactility and visual image (176),” reproducing while altering itself paradigmatically, though this mimesis is more macrological than that which we are considering. Analogously, Alessandro Duranti has advocated rereading of Husserl's notion of intersubjectivity. In his article Husserl, intersubjectivity, and anthropology, he argues that the critique of Husserl on the part of ethnographers of the South Pacific, where some groups consider the anticipation of another’s thoughts as absurd or impossible (Duranti 2010: 5), flies in the face of Husserl’s insistence that intersubjectivity, like perception, has a thoroughly non-propositional dimension. He claims that the confusion arises from a mistranslation of compounds with the root weschel, such as wechselverständnis, einverständniss, and wechselverständingung, as “mutual understanding” and goes on to state that weschel is best translated as ‘change, exchange, reverse, succession, rotation (Duranti 2010, 5) and should thus be approached through the lens of Husserl’s notion of platzwechsel or “place exchange (Duranti 2010, 6).” Such place exchange, more akin the pre-cognitive anticipation of the others line of site than an imaginative reconstruction of his or her subject position, Duranti suggests, might be related to “mirror neurons” as studied by the neuroscientist Vittorio Gallese: neurons that have been shown to mimic those active when performing a particular action, even when merely observing an image of that action. Finally, William Mazzarella, in his work on advertising and globalization in India, has advocated rereading the classic crowd theorists, particularly Tarde, and extrapolated a deployment of mimesis that allows an account of the conduction of affect through media networks that circumvents the discourse on embodied versus mediated experience.

Miller’s claim, then, that disgust spreads through a principle of emotional contagion, then, is not as ex nihilio nor anecdotal a statement as might first appear. His notion that disgust and being disgusting spread through proximity is well grounded in a deep routed and transdisciplinary elucidation of the crowd as emotionally conductive via the principle of mimetic exchange. Having abandoned the search for the originary point where disgust is “blown into” the network, we might see the crowd as a sort of vast epidermal surface, a plane of mimetic nodes across which disgust can reproduce itself and circulate at lightning speed.

Repulsive Affects

We run into an obstacle to our transposition of Miller’s elucidation of disgust into our model of crowd dynamics when we consider his insistence that disgust always entails moral or social judgement. The entire book is laced with the refrain that disgust is “no mere feeling” or “hard wired reflex” but rather something closely related to forms of judgement that produces a form of hybrid moral-aesthetic evaluation. Most incompatible with the purposes to which we want to put Miller’s work, he insists that disgust is always a marker of social difference and hierarchy, an emotion that always negatively evaluates what it touches, proclaiming what it touches inferior (Miller 9). To this insistense we might counterpose Goffman’s careful elucidation of the somatic and pre-cognitive modes by which participant units circumvent any putting into play of status or other social positionality. Take his illustration of what he terms a “body gloss,” for example:

By providing this gestural prefigurement and committing himself to what it foretells, the individual makes himself into something that others can read and predict from; by employing this device at proper strategic junctures—ones where his indicated course will be perceived as a promise or a warning or threat but not as a challenge—he becomes something to which they can adapt without loss of self-respect (Goffman 1971: 11).

Indeed, when we pose the question of the disgust of crowds to ourselves in a vernacular manner, it immediately strikes us as melodramatic to state that we are disgusted by a stranger’s touch in a subway car, though such an emotional response is certainly possible given certain circumstances or subjects, such as when a subject is percieved as smelling putrid, or when one is forcefully grabbed or shoved rather than brushed or pressed.

Nor are we helped out of this impasse by defining the disgust at play in crowd dynamics as an attenuation or trace of emotional disgust. First, Miller already addresses such soft forms of disgust and they pertain not to quasi-instinctive avoidances and aversions but rather to the myriad subcultural condemnations and “pet peeves” that are as judgemental as harder forms of disgust, though less acutely felt. Second, there is nothing weak or inconsistent in mutually repellent force of participation units – indeed the continous fluidity with which crowds move through the city’s underground, baring exceptional events or insidents, is a testament to the resilience of this force. Finally, the intersubjective repulsion that allows crowds to flow seems to operate quite independantly of judgemental emotions as described by Miller and others. The reflex jerking away of one’s hand can take place without necessarily interupting the flux of what we usually term emotions. A woman all aglow at having found a new and to her mind superior lover will move through the tunnels and in and out of the subway cars while fluidly carrying out various avoidances and routings, even mumbling an apology when she accidentally trips someone up without interupting either the passage of her fellow riders or her sensations of warm satisfaction. Of course, a despairing man who draws away from the incidental touch of a fellow rider might despair further, taking his recoiling as yet another sign of an anomic “society” that disgusts him, but this requires him to read the reflex as such, to arbitrarily attribute meaning in the Sausserian sense, that is, arbitrary in respect to the the sign’s relation to the signified.

In Massumi’s Autonomy of Affect we find a conveniently two-tiered model of feeling, one which I am hesitant to adopt in its entirety but also one that can furnish the language to tease apart disgust in Miller’s sense from disgust as it pertains to the repulsion at play in the fluidity of crowds. Specifically, Massumi terms these two domains emotion and affect respectively, where the former is the receptor of what he dubs qualification and the latter of what he terms intensity. Part of what is attractive in Massumi’s model is that these two tiers do not map onto Cartesian mind/body dualism: neither domain is mind-locked, both have their own outlets to the body.

Both levels, intensity and qualification, are immediately embodied. Intensity is embodied in purely autonomic reactions mostly directly manifested in the skin–at the surface of the body, at its interface with things. Depth reactions belong more to the form/content (qualification) level, even though they also involve autonomic functions such as heartbeat and breathing... Modulations of heartbeat and breathing mark a reflux of conciousness into the autonomic depths, coterminous with a rise of the autonomic into conciousness. Intensity is...disconnected from meaningful sequencing, from narration, as it is from vital function... spreading over the generalized body surface. (25)

To clarify, Massumi is here making reference to two different autonomic indicators of feelings, heart rate and galvanic skin response (the tightness of the skin), which, to the puzzlement of some psychophysiologists, map differently onto identical stimuli. As crowds close around a participation unit, the skin tightens and becomes alert to a corresponding degree, bristling not only at the stranger’s touch but at his or her variable proximity; when the threshold corresponding to the crowd’s density is transgressed, the effect is electric. When a crowd’s participation’s units space themselves so as to minimize intersubjective friction, their hearts do not flutter unless they are at the same time in love or abjectly disgusted on an emotional level. In short, this affective microdisgust, so to speak, is not necessarily or even usually more than skin deep and does not interrupt the narrative into which experience is continually fed.

Massumi, however, exaggerates when he calls this affective tier a “never-to-be-conscious autonomic remainder [my emphasis].” It would be more accurate to say that this tier does not necessarily become conscious. This is evident in our example of the despairing man for whom the immediate pulling away from the stranger’s touch becomes inscribed into an emotional landscape of anomie. We sometimes find in Massumi and other aficionados of Gilles Deleuze what William Mazzarella, citing, Michael Silverstein describes as a “radical binarization of conceptual mediation and affective immediacy” that is “analytically untenable but also a contingent feature of modern European philosophy... a romantic (and complicit) attachment to a fantasy of immediacy (Mazzarella 2010: 294).” Mazzarella is especially vexed by the vitalist elements in Antonio Negri’s concept of the multitude. Where Antonio Negri deifies (in a literal, Spinozan sense) the multitude over and against the heteronomous crowd as “the elemental flesh that is maddeningly elusive, since it cannot be entirely coralled into the heiarchical organs of a political body (Negri in Mazzarella 2010: 708) ” and when Massumi falls into condemning the mediated and qualified as “without potential, pure entropy, death (Massumi in Mazzarella 2010: 294),” Mazzarella is quick to point out that such rhetoric merely mirrors the dualisms of early crowd theorists such as Taine and Le Bon, where reason and the liberal individual stand in opposition to dividual flesh. Whether or not we concur with Mazzarella’s critique, we might take it as an opportunity to caution that the affects at play in the dynamics of crowds are never pre-social, though the temporality and mechanism of their sociality is different from what we are terming emotion. Against “romantic immediatism,” whether it be a central tennet of Negri or Massumi’s or simply a side effect of some of their flourishes, we might observe that though affect, as the repulsive principle undergirding the fluidity or viscosity of crowds, operates autonomously, it is nonetheless heteronomously inscribed.

That repulsion as a sense of space is nonetheless a socialized sense is evident in the variable expertise at play in each participation units routing through the crowd. In Marc Augé’s In the Metro, he gives a poetic description of the deployment of this spatial sense, on which should be eminently familiar to any frequent subway goer:

The regular traveler on a given line is easily recognized by the elegant and natural economy of his or her way of walking; like an old sailor who calmly descends towards his boat at dawn and appreciates in a glance the billowing waves at the exit of the port, measuring the force of the wind without appearing to touch it, with style, but in a less studied way than a taster sniffing a glass of wine, listening without seeming to heed the waves slapping against the jetty or the clamor of the seagulls gathered on the shore or already scattered over the sea in little avid flocks, the seasoned traveler, especially if he or she is in the prime of life and strongly resists the desire to suddenly burst into the stairs for sheer pleasure, can be recognized in the perfect master of his or her movements: in the corridor leading to the platform, the traveler walks swiftly but without rushing; without letting on, all senses on alert (Augé 2002: 7).

This sensitivity describes how, in general, attunement to intersubjective repulsion is proportional to the efficiency with which a unit can navigate a crowd without interfering with the crowd’s overall fluidity, that is to say, the participation unit can increase its speed without incurring the faster-is-slower effect. For Goffman, the absence of collisions is precisely the determinant of a perception of order, even where one sees only a chaos of swirling bodies:

A condition of order at the junction of crowded city thoroughfares implies primarily the absence of collisions between men or vehicles that interfere with another. Order cannot be said to prevail amoung people going in the same direction at the same pace, because there is no interference... But when all who meet or overtake one another in crowded ways take the time and pains needed to avoid collision, the throng is orderly (Goffman 1971: 6)

Interestingly, our deployment of Goffman here puts him at odds with his own notion of “disattendability,” a term which Miller adopts when he discusses disgust in mass, public settings. Specifically, Goffman and Miller propose that, at least in metropolitan, middle class, mass society, there exists an emotional precept “not to make undue claimson peoples fears or trigger their embarassment and disgust,” to “accord to others civil inattention (Miller 1997: 199).” Though Goffman limits his study to the Ango-american Middle Class, the hand of Georg Simmel, cited as one of Goffman’s greatest influences, is evident here – we might recall Simmel’s seminal essay, The Metropolis and Mental Life, where he weaves together the anaesthesia and hyperaesthsia of the city:

There is perhaps no psychic phenomenon which is so unconditionally reserved to the city as the blasé outlook. It is at first the consequence of those rapidly shifting stimulations of the nerves which are thrown together in all their contrasts from which it seems to us the intensification of metropolitan intellectuality seems to be derived... Just as an immoderately sensuous life makes one blasé because it stimulates the nerves to their utmost reactivity until they finally no longer can produce any reaction at all, so, less harmful stimuli, through the rapidity and contradictoriness of their shifts, force the nerves to make such violent responses... This incapacity to react to new stimulations with the required amount of energy constitutes in fact that blasé attitude... one never feels as lonely and as deserted as in this metropolitan crush of persons (Simmel 1972: 14, 16)

Goffman, of course, does not distinguish between affect and emotion in the manner that we have proposed, assimilates to this Simmelian precept to the tacit “traffic code” of pedestrian mass movement. Any empircal account of the rythms of the New York subway system, however, raises several objections to disattendability as articulated by Miller and the Goffman of Stigma and The Presentation of the Self in Everyday Life.

First, we must observe that, at least in the setting of the subway car, there is no such thing as a monolythic or even hegemonic American Middle Class. What we might imagine to constitute such a class is, if it can be said to exist at all, staged away from the metropole’s core, in the anywhere suburbs of hollywood sitcoms. Moreover, the participation units which might socio-economically approximate such a class are riven by racial affiliation, language and dialect, generation, temperament, subculture, and so on. Second, even if we were to attribute this middle-classness to particular participation units, they can never observed in vitro but only in relation to other units, even at the height of the business district’s rush hour. Third, and most importantly, the subway is rife with clashing emotions. A group of teenagers laugh and jostle one another while a old woman, hunched over his little grocery cart, scowls and sighs audibly. A shy country boy timidly keeps his eyes fixed to the ground, lifting them nervously only to cast them down the moment someone catches him peeking. Flirtatious glances and wry smiles ricochet across the car. Two men, brows furled, maddog one another across the aisle in a duel of the gaze. Even those who intervene technologically in order to remove themselves as much as possible from the scene, donning big dark sunglasses and noise cancelling headphones, betray pleasure as they bob their heads to beats or frown at the “outer” world through the lens of furious power chords.

Yet in spite of this crowded cacophony of emotional states, a common spatial feeling permeates the car. This affect, this intersubjective feeling of repulsion, serves as the lubricant that allows the crowds to floods in and spill out as a sentient liquid. Indeed, perhaps the most artful demonstration of a highly cultivated attunement to intersubjective repulsion are the breakdancing crews which transfer from car to car during periods of intermediate crowd density. Using the cars’ various poles and handholds, these dancers perform various rythmic contortions, flips and spin moves, without ever brushing or crashing into their fellow passengers. They glide through the field of intersubjective repulsions, propelled by these affective intensities inasmuch as they are propelled by centrifugal force, gravity, and their well honed muscle-memories – like a flock of crows playing in airstreams and atmospheric upwellings.

Conclusion

Though, in the following Chapter, we will discuss the implications of treating alignment and attraction as forms affective conduction, the problems and ramifications of treating affect over and against emotion as a principle of crowd self-organization should be already evident. Namely, the first chapter was concerned with techniques for sealing the individual against crowd emotion as destabilizing of the particular balance of such emotions which make up rationality. They do not seal against the field of affective repulsions, attractions, and alignments, that constituative of a crowd’s body, that is, its being as a coherent and collective being, but which do not necessarily threaten or appear as threaten any social, political, economic, emotional regime. Rather, these affects feed into the order and functioning of such regimes inasmuch as they exist apart and potentially against them – consider, for example, that without intersubjective repulsion the subway system would be unable to deliver workers to their workplaces. The techniques of removal or distancing from the crowds are, instead, levelled against crowds as potential bearers of emotion in all its political, moral, and juridical plenitude.

In Chapters four and five, then, we will investigate particular types of crowds where emotions are made to flow along the relays that mirror those of the affects which manifest a crowd’s physicality. Specifically, chapter four will consider the stadium as a particular, ocular technology, one which is self-organized before it becomes and architectural intervention, which imposes some measure of emotional synchrony onto the crowds that are, absent such a technology, affectively organized but emotionally heterogenous. In the fifth Chapter, we will finally arrive at the politicized crowd which has served as the starting point for the vast majority of crowd theorists. We will undertake a brief taxonomy of such crowds, untangling demonstrations, manifestations, revolutions, and so on, and ask to how and to what extent and how each of these types of politicized crowds effect an emotional synchrony. We will also explore a quite fascinating attribute of such crowds, where they intervene not only as synchronous blasts of collective emotion but also into the underlying magnetisms of affect; how the experience of a political crowd affects how its participant units are willing and able to constellate themselves in crowds writ large.