time-bubbles of nationalism: dynamics of solidarity ritual in lived time

Time-bubbles of nationalism:dynamics of solidarity ritual

in lived time*

RANDALL COLLINS

University of Pennsylvania, USA

ABSTRACT. The growth of modern nationalism can be attributed to structuralcauses, especially the growth of the strong bureaucratic state that penetrates society,creating cultural uniformity and national identity. But structurally based nationalismneed not be very intense, or constant; even when institutionalised in periodic formalrituals, it can be routine, low in emotion – even boring. We need to explain suddenupsurges in popular nationalism, but also their persistence and fading in medium-length periods of time. Nationalist surges are connected with geopolitical rises and fallsin the power-prestige of states: strong and expanding states absorb smaller particular-istic identities into a prestigious whole; weaker and defeated states suffer delegitimationof the dominant nationality and fragment in sudden upsurges of localising nationali-ties. Passing from macro-patterns to micro-sociological mechanisms, conflict produc-ing solidarity is a key mechanism: dramatic events focus widespread attention andassemble crowds into spontaneous ‘natural rituals’ – mass-participation interactionrituals, as distinct from formal rituals. Evidence from public assemblies and the displayof national symbols following the terrorist attacks of 11 September 2001 (9/11) showsan intense period of three months, then gradual return to normal internal divisions byaround six months. Spontaneous rituals of national solidarity are produced not only byexternal conflict but by internal uprisings, where an emotional upsurge of nationalidentity is used to legitimate insurgent crowds and discredit regimes. Although partici-pants experience momentary feelings of historic shifts, conflict-mobilised national soli-darity lives in a 3–6-month time-bubble, and needs to institutionalise its successesrapidly to have long-term effects.

KEYWORDS: Arab Spring 2011; interaction ritual; national symbols; popularnationalism; post-9/11; time-bubbles

Introduction

Frequently, nationalism is explained in terms of macro-structural causes.Ernest Gellner, Anthony Smith and others have debated its relationship to

* Editor’s note: this is a revised version of the ASEN / Nations and Nationalism Ernest GellnerNationalism Lecture, delivered at the London School of Economics and Political Science, 4 April2011.

bs_bs_banner

ENASJ OURNAL OF THE ASSOCIATION

FOR THE STUDY OF ETHNICITYAND NATIONALISM

NATIONS ANDNATIONALISM

Nations and Nationalism •• (••), 2012, 1–15.DOI: 10.1111/j.1469-8129.2011.00530.x

© ASEN/Blackwell Publishing Ltd 2012

modernity and to pre-modern structures. A version of the historical transfor-mation developed by Gellner protégés such as Michael Mann, as well as byCharles Tilly and others, may be called the theory of state penetration intosociety. I would put it in Weberian terms: patrimonial structures based onalliances among armed households were displaced by the bureaucratic organi-sation of the state, beginning with a permanent military and its centralisedlogistics, generating an expanded apparatus of tax extraction and the penetra-tion of state agencies and laws that bring individuals into direct relation withthe state.

Among the important consequences of state penetration was the inventionof the social movement. In parcellised medieval society, protests could only belocal and mostly ephemeral but state penetration created both the means oflarge-scale political mobilisation and a target to aim at. By fostering commu-nications, transportation, education and a dynamic economy, centralisingregimes provided the material means for group mobilisation. By penetratingthe walls of patrimonial households to inscribe persons on the rolls of the state,not as group members but as individuals, such regimes opened up new possi-bilities for identity-formation. By centralising the seat of government, theyprovided a unifying target at which petition campaigns and revolutions alikecould aim. Accordingly, as Mann (1993) and Tilly (1995, 2004) have empha-sised, all kinds of social movements mobilised at the same time: classes, nation-alisms, as well as a variety of reform movements ranging from anti-slavery toanti-vivisectionism. The social movement was a mold from which a variety ofthings could be poured. Nationalism was one of its products, just as classmovements like socialism and conservatism were likewise products of theinvention of the social movement.

I would like to extend another angle of vision: the micro-sociology ofnationalism. Nationalism may be defined, ideal-typically, as an intensely feltbond of solidarity. But as Jon Fox and Cynthia Miller-Idriss (2008) andothers have noted, in the day-to-day experience of modern people, nationalidentity is largely a matter of routine. Most of the time, nationalism is low-strength; latent, perhaps, but far from the thoughts and feelings of mostpeople. The institutionalised celebrations and monuments of nationalism arefor the most part a backdrop whose meaning is hardly reflected upon, likethe statue of the victorious general in the park splattered with pigeon drop-pings. National holidays – to speak of the ones I have observed, like the USFourth of July, Mexico’s Cinco de Mayo or France’s Bastille Day – areannually scheduled occasions for eating and drinking, fireworks displays, orjust a day off from work, and invoke little sense of national solidarity. (Ihave never seen people on such occasions encountering strangers, or evenacquaintances, and warmly addressing them as fellow countrymen.) It is adanger of symbolic analysis to presume that the analyst can identify themeaning of a symbol without examining what participants are actuallythinking and feeling at the moment. Symbols are alive only to the extent thatthey are the focus of shared emotional attention: symbols can be living, dead

2 Randall Collins


or lukewarm. What I would like to add is a mechanism determining theirintensity or lack thereof.

Even in societies where nationalism is institutionalised, it is not necessarilyvery strong or constant. Structural conditions are insufficient to explain theintensity of nationalism – the moments when it is indeed a living bond ofsolidarity, uniting strangers into a vast brotherhood – and when it is on theperiphery of people’s consciousness, or even non-existent in their lives. Weneed a dynamic theory of nationalism as a process of surges in time.

Hence my title: ‘Time-bubbles of nationalism’. These are capsules of col-lectively experienced time, on the whole rather sudden in onset, lasting for awhile, then declining back to banal normality. I could have used the metaphorof a balloon, getting puffed up with Durkheimian collective effervescence,floating high in the air in a mood of widespread social enthusiasm, thengradually coming down to earth as the gas slowly seeps out. Social scienceshave not been very good at explaining the temporal dimension, the time-patterns of social processes. The most common metaphor, the cycle, is notstrictly applicable to most important events of social life, especially socialconflicts and mass enthusiasms. The metaphor of the cycle, which may begraphed as a wave, is inaccurate because most real-life social events do notshow the symmetrical patterns of an ideal-typical sine wave: the ups anddowns do not recur at regular intervals, nor is the amplitude of the peaks andtroughs generally so regular. The metaphor of the balloon gives a betterpicture of the asymmetry of such collective time-dynamics, and of its capsuleshape: the balloon is inflated quickly, stays high for a while, then graduallyfloats down; its beginning is more sudden and dramatic than its ending. Abetter metaphor would be a rocket zooming into the sky, exploding in aneye-catching blast, then dispersing into fallout fading slowing from vivid todrab. But putting all of that into the title of my article would have been moremystifying than illuminating.

Let me give an empirical example. The day after the 11 September 2001(9/11) attacks on the twin towers in New York and the Pentagon, I realised thiswas an opportunity to study the long-asserted principle that external conflictbrings internal solidarity. How long would solidarity last? The first two dayswere a time of shock, people acting bewildered about what to do collectively.On the third day, American flags started to appear. Within a few days, theywere sold everywhere, posted on cars and windows, to some extent worn onclothing. All public gatherings – such as athletic events and concerts – beganwith huge flag displays, generally accompanied by the other newly consecratedsymbols of heroism, ranks of police officers and firefighters representing thosekilled in the twin towers. I surveyed the number of cars on the streets thatdisplayed flags, as well as the number of flags on buildings and windows,repeating observations of the same places at least weekly for a year (Collins2004a). The time-pattern that emerged was the following.

The peak level of displaying national flags was reached quite rapidly, withinthe first two weeks. It stayed at that peak for three months. This was also an

Time-bubbles of nationalism 3


intense period of national solidarity in other respects: political debate largelydisappeared and the popularity of the president (George W. Bush) reached 90per cent (the highest on record), a rise of about 40 percentage points from justprior to 9/11 and far above the levels to which it would fall later in Bush’sadministration. Around three months, political argument resumed and articlesbegan to appear asking ‘Is it ok to take our flags down now?’ The level of flagdisplay began to fall off, reaching a moderate level around six months, withbrief upward blips at the one-year anniversary and thereafter. On the whole,we can describe the pattern of national solidarity as exploding upward, rocket-like, in a week or two, remaining at its balloon-like ceiling for three months,then gradually declining towards normality by six months. I have referred tothis as the three-month solidarity-and-hysteria zone, and adduced compara-tive evidence that it is within such three-month bubbles of extreme collectiveattention upon a common identity and a shared danger that both precipitousventures and violent atrocities are most likely to happen.

Passing by the details of the 9/11 response and similar events, I want to raisea set of theoretical issues. The shape of mass solidarity in time, the shape of thetime-bubble – the rocket-like ascent, the plateau-like ceiling, the slow dissipa-tion – is similar for the wider class of such events, although much remains tobe established by comparison of cases. What varies is how long such eventstake. Here, it appears from cases that I will describe shortly, the big variationis the length of the high plateau. This calls for theoretical explanation: tenta-tively, I will argue that societies with a high degree of state penetration, andthose where symbolic mobilisation is easily perceived throughout the society,sustain plateaus of national solidarity in the three-month range; but societiesthat are more fragmented and less state-penetrated sustain the plateau for amonth or less.

A further theoretical issue concerns what happens after a collective mobi-lisation has gone through its peak and dissipated. My hypothesis is that arefractory period follows, such that no similar mass enthusiasm can be gener-ated for a period of time thereafter. To risk another metaphor, the nerves ofthe social animal are innervated after this orgy of attention upon the collec-tivity; once the air is out of the balloon, it can’t get it up again until after a timeof rest. The Thermidorian Reaction during the French Revolution is a famousexample of such exhaustion from sustained crisis. I will discuss the implica-tions of this point presently.

I offer the metaphor of a time-bubble to emphasise that the most intenseevents of social life are capsules in time. Their coming is not expectable in anyprecise way, certainly nothing like the periodicity of sine waves; it may be along time before another rocket goes up, another balloon is filled with similarlyuplifting enthusiasm. That is why such moments in time have the emotionalcharacter of high drama; both tragic and joyous surprise. There is another anddeeper theoretical reason for this sense of uniquely high experience, besidestheir rarity and unpredictability. What happens inside the bubble is felt to bequalitatively different from ordinary life outside. Not only is life much more

4 Randall Collins


intense at those times, but it has a different structure of social attention. It israre for us to be in a situation in which we can be reasonably sure that mostother people are paying attention to the same thing, and feeling the sameemotion as ourselves. When one is in a mass demonstration, crowds stretchingout as far as the eye can see, and even more people beyond them, all thinkingthe same thought and expressing the same emotion – ‘The regime must go!’,‘The people will triumph!’ – this gives a sense of the solidarity of the whole thatDurkheim called collective consciousness. At such moments it is not anabstraction or a myth, but what people feel as the highest reality. The momentdoes not last, but they are not thinking that now. It will not last because ittakes special circumstances to get such a high degree of simultaneous focus ofattention. But while it is there, its power of shaping emotions and formingsymbols is unsurpassed.

A mass demonstration or focused crowd of this sort is a large-scale versionof what I have called an interaction ritual. This concept attempts to capture ageneric dimension of ritualism, not only in formal and traditionalised ceremo-nies but running through all the activities of social life. Elsewhere, I havespelled out the ingredients and consequences of interaction rituals of varyingstrength (Collins 2004b). The ingredients are: first, assembling people bodily inthe same place, so that they are in full multi-modal intercommunication;second, focusing their attention upon the same thing and becoming mutuallyaware of each other’s focus, thereby generating a sense of intersubjectivity; andthird, feeling and expressing the same emotion. Interaction rituals can succeedor fail, can be intense or mediocre; if the ingredients pass a threshold, mutualfocus and shared emotion feed back into each other, driving them upwards tohigh levels of rhythmic entrainment that Durkheim called collective efferves-cence. At these high levels, what the group focuses upon becomes symbolic,representing membership in the group as well as depicting its boundaries andenemies. Individuals are filled with emotional energy, the feelings of confi-dence and enthusiasm that motivate them to acts of heroism and sacrifice.They are filled with a sense of morality, the palpable experience of good and itsfight against evil. Nationalism is such a symbol.

It is through these symbols, in subsequent days, that people can recall thefeelings of solidarity and morality that pumped them up at the peak moments ofcollective experience. The group cannot stay assembled forever; indeed, it isdifficult for them to stay together for as long as a few weeks. The collectiveemotion fades after the assembly disperses. The symbols keep it alive to a degreebut, as Durkheim recognised, periodic reassemblies and periodic rituals areneeded to rejuvenate the symbols with emotion. I have added an argumentabout a refractory period: this holds that after an intense period of massmobilisation, people cannot experience the same intensity again for some time;they necessarily have to come down. Thus symbols must vary over time,between moments of high-intensity significance and milder levels, even banality.

My discussion of the mass demonstration as an interaction ritual is on themicro-sociological level. Let us recombine it now with the macro-historical



level with which we started, the long-term transformations of state penetrationand the invention of the social movement. The modern social movementbecame possible because it was provided with a target, the central organisationof the state; this created a physical place to assemble and a mutual focus ofattention. Whether a movement wants to petition for redress of grievances, oractually to overthrow the regime and take power for itself, it focuses attentionupon the state and elevates it into an image of omnipotence; an all-encompassing will or agency, which at the moment of maximal struggle is nowendowed with the ability to carry out the will of the crowd, if it only would.This is a somewhat mythical view of the power of the state, ignoring allpractical contingencies of organisation and implementation, but it has veryimportant ideological effects. The unifying focus on the state as the target ofmovement mobilisation makes it easy, indeed natural, to invoke the unity ofthe people. Whatever the specific demands of the movement, it is we – theunified Durkheimian collectivity – who are demanding it. At the moment ofmobilisation, the sovereignty of the people is not a philosophical abstractionbut a felt experience. For this reason, mass movements engaged in stateconfrontation tend towards a nationalist tone because this is an easy identityto evoke at the moment when seemingly everyone is assembled and united intheir demands. It is rituals of this intensity that make primordialism plausible,at least for their own participants.

A recent example, which I have studied from videos, photos, news reportsand eye-witness accounts, is the two-and-a-half-weeks protest at the Wiscon-sin state capitol in the USA (during February and March 2011). The issue wasthe plan of the Republican-dominated legislature to pass a bill eliminatingmost of the organising rights of public-sector employees’ unions. Because thestate capitol building is very close to the state university, it was easy for theTeaching Assistants’ Union to mobilise a mass protest that occupied the statebuilding and attempted to block the legislation. The protest attracted a largenumber of supporters, including even police officers – a fact that delighted theprotestors. Although it was eventually defeated on the legislation, the protestgenerated a mood of euphoria, with chants of ‘This is what democracy lookslike!’ and ‘We are the people!’ ringing through the halls and galleries of thebuilding. In fact, in the election three months previously, the Democrats andunion supporters had lost control of every branch of the legislature and thegovernorship so it cannot literally be the case that the protestors were thewhole or even the majority of the people. My point is that the micro-structured experience of a mass demonstration itself generates the feeling oftotality. One is aware there are enemies outside the collectivity, but they aredefined away; not part of the people but an alien quality to be combated,much like theological views of the omnipotence of God and the shadow-reality of the devil. Thus the Wisconsin demonstrators displayed themselvessurrounded by symbols, American flags, references to democracy and ‘thepeople’. Although the issue was specifically the rights of labour unions, thediscourse and symbolism was not largely about unions: this was played down,

6 Randall Collins


a particularistic interest that had to be reinterpreted in the context an all-encompassing unity.

I have been arguing that the shape of the social movement is what createsnationalism and that this operates through a micro-sociological mechanism,the mass interaction ritual mobilised at the place of the centralising state.Historically, we can see this same process of micro-foundations of movementmobilisation in the history of nationalism, specifically in the shift fromnineteenth-century elite nationalism to the mass nationalism of the twentiethcentury and today. It has been widely argued that nineteenth-century nation-alism, particularly in less modernised states, was a matter of a small number ofintellectuals. Micro-sociologically, their chief tactic was the banquet. A fewdozen gentlemen gathered in private mansions, restaurants or hotels to dine;they made speeches, drank toasts, passed resolutions. It was a banquetingcampaign of this kind that led to the French Revolution in Paris in February1848, and that carried the nationalist revolution to Germany and elsewhere. Itwas a banqueting campaign that mobilised reformers in Russia in 1904, in theopening provided by defeat in the war with Japan, and prepared the way forthe attempted revolution of 1905. I am drawing here on research by StefanKlusemann (2010), who shows that the restricted settings of indoor banquets,extending the polite rituals of upper-class sociability, gave a very limited classcharacter to nationalist mobilisation. A widening came about when the labourunions in St Petersburg demanded entry to the banquets and eventually trans-ferred such meetings to larger, more public buildings, thus setting off themovement for councils – soviets. It was the shift in the micro-organisation ofparticipation in a social movement that generated the key ideological shifts.Radicalisation is a practical process and not merely an ideational one.

To summarise the key points so far: social movements are facilitated by thestructural changes of modernisation through state penetration; movements attheir peak mobilise into huge interaction rituals directed at the central state;the micro-sociological experience at such moments is the feeling of universalsolidarity in the crowd, which can be attached to a variety of ideologies but isespecially compatible with ideas of popular sovereignty and the nation. Andthese peak moments are time-bubbles, lasting for a few weeks or months thenfading, and are subject to a refractory period during which they cannot bemobilised at that same intensity.

I want now to continue with the dynamics of such mobilisation as theystretch out in time. A key point is the difference between the pristine experienceof the Durkheimian collectivity, when mass solidarity discovers itself in alarge-scale mobilisation, and the change that comes in subsequent mobilisa-tions. There is a refractory period of some length after intense mobilisationdissipates. Under some circumstances, the demonstrators can go home; thestruggle may go on, but in a different and milder form. After their eighteendays occupying the Wisconsin state capitol, the crowds dispersed; the struggleshifted to lawsuits in the courts and electoral campaigns to recall officials. Butsometimes crisis can follow crisis and renewed mobilisation may be forced



upon a population, however exhausted. In 1917 Russia, there were two greatpeaks of revolutionary mobilisation (in February and November), separatedby nine months during which the collective identity of the people was, if not inabeyance, at least fragmented and incapable of mobilising as a unitary ornear-unity force. But after the November revolution, all was not settledforever; in fact there were renewed crises, with civil wars beginning already inDecember 1917 and continuing through 1920, a period of three years.

But these subsequent mass arousals have a different emotional atmosphere;less filled with the sense of shared solidarity and glory, more soldieringthrough. The peak Durkheimian moment, the fusion of the individual and thecollective, is replaced by doing one’s duty, going on under the pressure of anorganisation; leaders stop being symbols of ourselves and become authorities,not part of the shared flow but a separate force impinging from outside.1 Thereare numerous instances where revolutions or attempted revolutions followeach other in a close sequence; on the whole, the later ones have both adifferent tone and a different micro-sociological structure. The French Revo-lution of February 1848 was a quick and almost bloodless transition topopular rule. But by the time of the June insurrection, four months later,collective solidarity had split into factional conflict and a great deal of vio-lence. National unity gave way to class conflict.

On the micro-level, interaction rituals can succeed or fail. The key variableis maintaining a mutual focus of attention, which thereby channels sharedemotions into a feeling of solidarity and common identity. More precisely, theamount of mutual focus of attention is a continuum. At high levels there isunity of one group, the people as a whole, foregrounded in the mass mobili-sation we see all around us. There is just one goal: to win out against the enemywe are all confronting. At low levels of the continuum people do not assembleat all, do not pay attention to much in common, but are taken up with themyriad private concerns of everyday life. In between these levels there can bemass mobilisations, but these are split into factions: they can’t agree on goals;they go off into different collective actions; their discourse is splintered andoff-message.

Accordingly, there are two kinds of revolutionary or peak movement strug-gle. One is the Durkheimian collective consciousness, stretching as far as theeye can see; and beyond that, the sentiment that the whole world is watching.Such revolutions are often surprisingly rapid and easy. The fastest were the1848 French Revolutions: three days in February for the government to fall;five days in June for insurrection and counter-mobilisation to put it down(Tocqueville 1987). The key mechanism is the tipping point or bandwagoneffect. There is a tense period of confrontation between regime and massuprising; there may be sporadic efforts at dispersing the crowds, a brief andhesitant use of force by authorities; but if the crowd holds on, sustained by thecourage of mass solidarity, the regime tends to split. Elites go over to theopposition; soldiers and police waver and change sides; leaders feel isolatedand lose their nerve. This is the Durkheimian collective consciousness as a

8 Randall Collins


process in time; not a permanent entity but a forest fire, a hot centre thatnothing can resist. To shift the metaphor – again – the mass demonstrationinteraction ritual is like an emotional magnet: it pulls in not only those who arein the crowd, on the spot, but those who are in the opposition, tied to it by theinvisible line of force that is in fact the common focus of attention shared bythe regime and those who want to overthrow it.

Once the struggle is won, that common focus of attention is lost. The regimethat was the target of protest is gone. The subsequent regime is in a differentmicro-sociological situation; that is why the post-revolutionary situation isnecessarily different from those days of high euphoria and high danger whenthe revolutionary coalition was at its height.

The second type of revolt is one in which the unity of the revolutionists isnever achieved. Instead of one central focus of mobilisation, there are many:dissidents against the regime are also dissidents among themselves. Such revo-lutions are more difficult. They are more violent, often turning into civil wars.The regime may fall, but it is not replaced by anything of similar unity. Suchrevolts are characterised not by overarching Durkheimian collective con-sciousness but by smaller pockets of ritual solidarity, mutually hostile. Groupconsciousness is split. The first type of revolt, the Durkheimian bandwagon,is maximally suited to produce moments of pure idealistic nationalism. Thesecond type is less nationalist; that is, nationalism may be one of the ideologi-cal strands put forward during the movement but it is contested and uncon-vincing, not the dominant identity; in the aftermath a self-consciousnationalist movement may be prominent, but in the special form of a reactivenationalist militancy, struggling against other mobilised movements that arethe enemy of nationalism. The White and Red factions of the Russian CivilWar are an example, but there are many others.

It is dangerous to analyse events still in progress, but in the wave ofMiddle-Eastern revolts of 2011, we can point to the Egyptian revolution as anarchetype of Durkheimian bandwagon revolutions and the Libyan and Yemenrevolts as cases of the fragmented type, with the Tunisian revolution combin-ing elements of both. Let me review a few salient points, chiefly their timedynamics, micro-interactional patterns and the situational evocation ofnationalist symbols.2

The Egyptian revolution covered eighteen days of mass demonstrations,from 25 January 2011 (when the first big crowd gathered in Cairo’s TahrirSquare) to the fall of Mubarak on 11 February. Actually there was a prelimi-nary period of a week, beginning on 17 January (the day of the one-monthremembrance of the self-immolation that set off the Tunisian protests, when aman attempted to set himself on fire in front of the Egyptian parliament andfive others made similar attempts, all stopped by the police). This week-longcalm before the storm was given over to organising, so what we count as dayone (25 January) had pre-arranged widespread support from many political,civil and ad hoc groups. The date was chosen to take advantage of a nationalholiday in honour of the police. Hijacking a national holiday is not an unusual



tactic in authoritarian regimes that allow no other way to assemble, turning abanal mechanism of national solidarity into a mobilising device for revolt. Asimilar tactic was used to set off the Tiananmen Square democracy demon-strations in 1989, using the occasion of a state funeral for a CCP leader.

The police were also ready and violence quickly escalated, beginning withtear gas and water cannons against crowds wielding rocks and firebombs. Forthe first four days, violence took place at demonstrations not only in Cairo butAlexandria, Suez and several other cities. An important swing occurred on dayfive: the army was called out, but they refused to intervene between police andprotestors; by day seven, the army was announcing it would not hurt theprotestors. At the same time, the violence in other cities fizzled out: attentionwas centred on Tahrir Square in Cairo, where the crowds now reached250,000, while in Alexandria they had fallen to hundreds. As of day five,Mubarak began offering concessions, but these were not accepted.

By day eight, the Tahrir Square protest had grown to a million and hadbecome mostly peaceful. There was one more important episode of violence,the so-called Battle of the Camel, when Mubarak supporters invaded thesquare, causing many injuries and several deaths, while the army attemptedto separate the sides. This was day nine; from day ten onwards, there was nonotable violence in Cairo and most of the rest of the country. A sign of thetipping point was visible already on day sixteen, when a small demonstrationin a provincial town was attacked by the police, killing several; but the fol-lowing day, the police were arrested. Mubarak, on the defensive, offeredmore concessions; these were rejected as unacceptable, and he finally resignedon day eighteen. This led to one last big day of rejoicing in Tahrir Square(with imitations in other cities), before the square was cleared. In the follow-ing days, there were demonstrations in various places but these were nolonger big heterogeneous crowds; instead, separate occupational groupspressed their specific demands. The honeymoon was over, replaced byinterest-group politics.

The period of violent contestation was brief, almost entirely in the first fourdays, and spread among many different sites around the country. After that,the Durkheimian collective consciousness became an overwhelming centre ofattention: virtually all other action stopped to focus on Tahrir Square. (Thechief exception is after the last big violent episode, the Battle of the Camel onday nine; this was followed by a brief upsurge of violence and looting inAlexandria.) The massive assembly, unmoved by violent attack, soon brokethe will of the regime; the army had shifted to the crowd’s side, and ended upin charge of the transitional regime. The Durkheimian solidarity of the assem-bled crowd was so great that the revolutionaries’ self-image – and their imagein the eyes of their international supporters – was as a non-violent, virtuallybloodless revolution. This was not strictly true: altogether, about 700 peoplewere killed and 6000 wounded (most of them in the first four days, whenfighting took place in many different cities, before Tahrir Square establishedits primacy in the attention space).

10 Randall Collins


Violence disperses, solidarity concentrates. We see this also in photos, as Ihave shown in a comparative study of violent behaviour (Collins 2008): whendemonstrations are peaceful, they are compact crowds; when violence is hap-pening, people are running around, widely spaced apart.

It was in the gatherings at Tahrir Square that expressions of Egyptiannational unity were most in evidence. Many women took part, some in fullabaya, others flaunting Western styles, even smoking and kissing. Christianand Muslim clerics ostentatiously appeared together, expressing religiousunity. These sectarian and gender divisions soon reappeared after the euphoricmass demonstrations had broken up. Some women were sexually molested andassaulted in the vast crowd on victory celebration day (day nineteen). WhenEgyptian women assembled again in Tahrir Square for International Women’sDay (8 March, three-and-a-half weeks after the downfall of Mubarak – i.e. dayforty-three) they were heckled and threatened (Philadelphia Inquirer, 18 March2011). On the following day, Muslim–Christian violence broke out again in theCairo suburbs, killing thirteen and wounding 140. The ostentatious waving ofEgyptian flags and signs referring to the unity of the people of Egypt waslargely confined to the high points of collective assembly, inside the time-bubble of nationalism. For the Egyptian revolution, the time-bubble waseighteen days (or twenty-six days if we include the previous week of initiation).

For Tunisia, we see a mixture of violent conflicts dispersed around thecountry and the eventual formation of a peaceful revolutionary crowd in thecapital city. The revolution in Tunisia took four weeks in December andJanuary 2010–11, with an additional two weeks of demonstrations after thepresident fled while protestors pressed for further changes. The first weekbegan with two spectacular public suicides in an outlying town, spreading toprotests in other small towns within a 30-mile radius of the capital. Becausethere was news censorship, these first protests probably spread by word ofmouth, like traditional protest movements such as the rural arson campaign inEngland in 1830 called the Captain Swing rebellion (Tilly 2003: 178–87;Collins 2008: 248). The first killings of demonstrators by the police happenedon day eight, as demonstrators became increasingly destructive (burning tyres,police cars and public buildings). In the second week, protests spread to thethree second-rank cities of the country, as the national labour union becameinvolved. In the capital, two days of relatively small and peaceful protests wereorganised by the lawyers’ association. Bigger and more modern organisationwas now taking place. Apparently this is when the ultra-modern Internet socialmedia began to make a buzz – not organising the revolt on the Internet alone,but adding to existing forms of organisational mobilisation on the ground.3

During days eighteen to twenty-seven, violent protests occurred in stillother provincial towns, especially in the remote western region near the Alge-rian border. Here violence became increasingly two-sided, as police escalatedfrom tear gas and water cannons to live ammunition; crowds threw stones andpetrol bombs, and snipers fired from rooftops. Simultaneously, in the Medi-terranean coastal cities where labour and professions were well organised,



more peaceful tactics were used, including protest marches, strikes and blog-ging. The combination of street-fighting in the small towns and the provinces,plus more peaceful, politically sophisticated demonstrations in the urbancentres, created an omnipresent sense of crisis. On days twenty-nine and thirty,the tipping point came: the president attempted to impose a state of emer-gency, the military staged a coup and the government fell. Thereafter, almostall the action was in the capital city itself, as demonstrations kept up pressurefor a more radical change in the transitional government. At this point thedemonstrations were on the whole more peaceful, and the police were eager toaccommodate their demands.

We see a contrast between the tactics and levels of violence in the two kindsof sites of confrontation. In the small towns where the protests began, and inthe remote provincial regions where it escalated to guns, bombs and arson, itis likely that the police had only small detachments. The fact that the violencespread from one town to another – rarely in the same place more than a fewdays (like the pattern of the so-called Paris banlieu riots of 2005 (Collins 2008:492)) – meant that the police were always heavily outnumbered and on thedefensive; when their forces were beefed up by reinforcements, protests movedsomewhere else. In the big cities, with their Westernised cosmopolitan style,protests had more centralised organisation behind them; violence wasrestrained on both sides; and the protestors generated a better sense of legiti-macy and responsibility, creating a magnetic attraction for government offi-cials who would move across the line in a tipping point for regime change.These big coastal cities are the main tourist route, touted in the West for theirbeaches, ancient ruins, colourful but tamed Islamicism and cafe culture. Onthe whole, the Tunisian revolution did not make much use of nationalistslogans and symbols. The revolution was made possible by the temporaryunity against the regime, as a common enemy, between the Arab underclass inthe hinterlands and the liberal cosmopolitans of the tourist zone. The con-trasting directions of orientation between the two components of the teamtended to keep nationalism out of the centre of attention.

In Libya and Yemen, revolutionary struggle was much more dispersed andprolonged, both in time and space. This pattern in itself is enough to limit thestrength of Durkheimian collective consciousness, and to prevent resolution ofthe conflict by a tipping-point abdication of power and flight to the winningcoalition. In Libya the revolt broke out almost immediately after the success ofthe Egyptian revolution, taking it as a sign that the unravelling of authoritar-ian regimes would proceed across the Arab world like a thread pulled from aworn-out sweater. It might also be regarded as the transmission of Durkhe-imian collective consciousness vicariously, by the mass media and by the newelectronic social media. The nationalism of such collective high points ispropagated as a kind of generic enthusiasm for the nationalist tactic; not forany nation in particular but as a device for generating the experience of thepeople united in action. In Libya, the spread of popular enthusiasm went onfor about eight days (beginning on 16 February); Gaddafi’s forces did not

12 Randall Collins


crumble, but took the time to mobilise their superior weaponry. At first,military repulses of the rebels were attributed to an anti-national force, foreignmercenaries; during the second week, as rebels began to be pushed back,African guest workers in rebel-held territory were attacked in a wave ofxenophobic nationalism (Los Angeles Times, 5 March 2011). Hysteria aboutalien mercenaries settled down in the third week, when it became apparent thatGaddafi had tribal support of his own, an apparatus of clients and a militaryforce that was much better organised than the almost completely spontaneousrebel army. There were no equivalents of Tahrir Square: even in rebel strong-holds like Benghazi, quick success meant there was no need for prolongedassembly, and the revolt soon turned into a civil war spread out across 500miles of battle sites.

In Yemen, there were indeed versions of crowd solidarity assembling heroi-cally against the regime, especially in the capital city, but struggles took placein many cities, many of them already strongholds of rival factions. Althoughthe Tunisian revolution provided a bit of a catalyst for a new round, it merelyadded to long-standing and multi-sided conflicts: a fifty-year history of civilwar, partition and reunification, and ongoing wars against secessionists, reli-gious and ideological enemies. Although there were instances of heterogeneouscrowds assembling in temporary unity against the regime of President Saleh,there was no unifying axis of conflict and little likelihood that the departure ofSaleh would be a tipping point to anything other than further multi-sidedstruggle.

Let me read from an Associated Press report from the capital of Yemen on22 March 2011, when protests had been going on for about forty days:

Protesters massed by the tens of thousands in the downtown Sana’a plaza they havedubbed ‘Change Square’. Crowds ululated, chanted and painted each other’s faces inthe red, white and black colours of the national flag. Conservative tribesmen broughttheir wives to the protest,4 and the women brought their children, all basking in acarnival atmosphere.

Similarly, one can view photos of supporters of the Libyan rebels with flags ofthe old monarchist pre-Gaddafi regime, or with their faces painted with thatparticular brand of nationalist colours. Similar uses of national symbols can beseen in revolts all over the Middle East. It is striking how many of these photosare of women. I read this as a strategic use of nationalism: women in the Arabworld are taking advantage of mass political mobilisation to take part inpublic life; unity of the people is of special concern for them, because they havebeen most excluded from the arena in which ‘people-hood’ is enacted. Perhapsshared memories of participation in these revolts will help – a little. It needsmore structural transformation for women to be accepted not as militantnationalists but as banal, taken-for-granted citizens.

The overwhelming impression one takes away from such cases is the illu-sory quality of nationalist time-bubbles if the structural conditions fornational unity, the state penetration and economic and cultural integration



articulated by Gellner, Mann and Tilly are not present. This is certainly thecase with revolts like Yemen and Libya. Will the Egyptian revolution, with itsgreater success in assembling a massive collective consciousness, have moreintegrating consequences?

To be sure, there is also a calmer, everyday nationalism than the impas-sioned displays of revolutionaries at the height of mass mobilisation – whatMichael Billig (1995) has dubbed banal nationalism. But national identity isimportant in ongoing social relations precisely when it is not the defaultsetting, the unmarked case, but the foregrounded gestalt that immediatelystrikes the eye. The national identities of self and other are present mostinescapably when there is a conflict or at least latent hostility founded onpalpable differences in privilege: British, French or Dutch colonialists in theirday; or American soldiers abroad now, irrevocably marked off by their intru-siveness, hence gathered together in consciousness of their nationality.National identity is most salient in everyday life when it is divisive. Wars, too,of course heighten nationalism, although chiefly in the first three months or soafter breaking out. Banal nationalism may not be much, but on the whole itreflects the conditions of everyday life better than the contentious mobilisa-tions and violence that are the platforms on which stronger national identitiesare staged.

I will end with a methodological exhortation. Nationalism is visible; it ismeasurable. It is a varying quantity in the interaction rituals of everyday life.As the ASEN conference of this year displays, many scholars are engaged inclose observation. What I would like to urge is more detailed attention to thedynamics that generate locally successful or unsuccessful ritual solidarity:what conditions exist for assembling or disassembling crowds; how much dothey create a mutual focus of attention, and what breaks up attention; whatkinds of emotions; what degree of intensification through rhythmic entrain-ment; how long solidarity rituals can be sustained. Our theoretical frontier isto explain their trajectories in time.

Notes

1 The mixtures of ideological discourse and banal realities are illustrated nicely by the writings ofIsaac Babel (2002) based on his experiences with Red Army troops during the Russian Civil War.2 Sources: BBC, AP and al-Jazeera reports.3 According to the BBC, 34 per cent of the population of 10 million was Internet users.Most of these were probably in the coastal cities (half the population lives in the impoverishedcountryside).4 Actually the text says ‘bought their wives’, but I think this is a Freudian slip.

References

Babel, I. 2002. ‘The Red Cavalry stories (1924–26)’ in The Collected Stories of Isaac Babel. NewYork: Norton.

14 Randall Collins


Billig, M. 1995. Banal Nationalism. London: Sage Publications.Collins, R. 2004a. ‘Rituals of solidarity and security in the wake of terrorist attack’, Sociological

Theory 22: 53–87.Collins, R. 2004b. Interaction Ritual Chains. Princeton, NJ: Princeton University Press.Collins, R. 2008. Violence: a Micro-Sociological Theory. Princeton, NJ: Princeton University

Press.Fox, J. and Miller-Idriss, C. 2008. ‘Everyday nationhood’, Ethnicities 8: 536–76.Klusemann, S. 2010. After state breakdown: dynamics of multi-party conflict, violence, and para-

military mobilization in Russia 1904–1920, Germany 1918–1934, and Japan 1853–1877. Ph.D.dissertation, University of Pennsylvania.

Mann, M. 1993. The Sources of Social Power. Vol. 2. The Rise of Classes and Nation-states,1760–1914. Cambridge: Cambridge University Press.

Tilly, C. 1995. Popular Contention in Great Britain, 1758–1834. Cambridge, MA: HarvardUniversity Press.

Tilly, C. 2003. The Politics of Collective Violence. Cambridge: Cambridge University Press.Tilly, C. 2004. Social Movements, 1768–2004. London: Paradigm Publishers.de Tocqueville, A. 1987. Recollections of the French Revolution of 1848. New Brunswick, NJ:

Transaction Publishers.



time-bubbles of nationalism: dynamics of solidarity ritual in lived time

Documents