Fin­ish­ing up what I said I’d fin­ish a cou­ple of months ago, this is a shorter ver­sion of a paper on “Iden­tity, Insti­tu­tions, and Upris­ings” with less math­e­mat­ics, no ref­er­ences (see the link above) and more opin­ion­at­ing. Also, a longer ver­sion of what I’m going to say at The­o­riz­ing the Web 2013 in a few days.

There is a the­o­ret­i­cal side to the “Face­book Rev­o­lu­tion” debate about the role of dig­i­tal tech­nolo­gies in the 2011 “Arab Spring” upris­ings, and it boils down to two ways of look­ing at things: the micro and the macro. On the one hand, we have the ratio­nal choice, agent-based approach and on the other we have more tra­di­tional soci­o­log­i­cal approaches based on larger-scale social structures.

If you look at some of the key char­ac­ter­is­tics of the upris­ings, it looks like a win for the micro side.

The sin­gle most dra­matic thing about the “Arab Spring” upris­ings was their unex­pected sud­den­ness. They fit the “infor­ma­tion cas­cade” mod­els devel­oped by Timur Kuran, Suzanne Lohmann and oth­ers to describe the equally dra­matic and sud­den 1989 upris­ings in East­ern Europe. Noth­ing on the “macro” side matches the ele­gant expla­na­tion of sud­den, dis­con­tin­u­ous change given by the micro-theorists.

Related to this sud­den­ness is the lack of a strong oppo­si­tion move­ment before the upris­ing. It’s not that there was no oppo­si­tion, but there was noth­ing of the strength to indi­cate a com­ing cri­sis. The cas­cade the­o­ries have no need, or even place, for orga­ni­za­tions or move­ments: these are pop­u­la­tion dynam­ics mod­els, with no struc­ture big­ger than the net­worked indi­vid­ual. Mean­while, the lead­ing soci­o­log­i­cal the­o­ries deal with move­ments, orga­ni­za­tions, and resource mobi­liza­tion. Score two to the micro-theorists.

Finally, we have the role of dig­i­tal tech­nolo­gies, which segues nat­u­rally to net­work mod­els of soci­ety. Talk of infor­ma­tion tech­nolo­gies leads equally nat­u­rally to a focus on infor­ma­tion dif­fu­sion across net­works, in which increased con­nec­tiv­ity low­er­ing bar­ri­ers to col­lab­o­ra­tion, dis­cus­sion, and infor­ma­tion shar­ing. And the macro-theorists again seem to have lit­tle on their side to cope with these kinds of ideas.

It looks like a shut-out win for the micro-theorists; the lan­guage of net­works and infor­ma­tion replaces the lan­guage of social move­ments and reper­toires of per­for­mance, and with that comes the inevitable idea that with a new kind of the­ory we are we see­ing a new kind of upris­ing, in which self-organizing net­works based on dig­i­tal tech­nolo­gies take cen­tre stage…

But you will have real­ized by now that this is a setup for me to argue that there’s another way of look­ing at these events, so let’s get to it.

The key suc­cess of micro-level the­ory is the expla­na­tion of cas­cades, which is a nat­ural con­se­quence of any model that has mul­ti­ple equi­lib­ria. Just because of that suc­cess, we don’t need to go whole hog and take on board the ideas of information-driven and network-sustained change. I want to argue that we can take the con­cepts that soci­o­log­i­cal research has shown to be impor­tant, and move them into the realm of ratio­nal choice mod­els. And when we do, we not only get pop­u­la­tion dynam­ics and cas­cades, but we also get expla­na­tions for sev­eral other aspects of dis­sent and upris­ings that net­works and information-based the­o­ries can’t deliver.

Is there a down­side? Of course there is. Behind the scenes, it’s often the case that ratio­nal choice the­o­rists like long equa­tions while soci­ol­o­gists love long words. Ratio­nal choice the­o­rists see the soci­ol­o­gists’ con­cepts as fuzzy, while the soci­ol­o­gists see the incen­tives of ratio­nal choice mod­els as sim­plis­tic. What I have to offer demands both long equa­tions and long words, and is open to being crit­i­cism for being simul­ta­ne­ously sim­plis­tic and fuzzy. Ah well.

Let’s start with a ques­tion. Zeynep Tufekci is a soci­ol­o­gist who was in Egypt right after the Jan­u­ary 2011 upris­ings, inter­view­ing par­tic­i­pants. Here’s what she says:

When I was in post-Mubarak Cairo, my hosts kept point­ing in amaze­ment to var­i­ous street cor­ners where fierce polit­i­cal dis­cus­sions were being held and often whis­pered, before remem­ber­ing they could now speak up and adjust­ing their voice, “You never saw this. Nobody ever dis­cussed pol­i­tics openly, ever.” Then they would pause and add, “Well, except online, of course. We all dis­cussed pol­i­tics online.”

So the ques­tion is that final sen­tence. Why is it that, prior to the rev­o­lu­tion, peo­ple could dis­cuss pol­i­tics online but not else­where? What made “online” a venue where those dis­cus­sions could take place? It’s not just ease of com­mu­ni­ca­tion, because if you want to com­mu­ni­cate you can stand on a busy street cor­ner – as peo­ple were doing when Tufekci visited.

The key thing is that com­mu­ni­ca­tion online was, for some rea­son, safe, while com­mu­ni­ca­tion on the street was not. It’s not just that com­mu­ni­ca­tion among like-minded peo­ple was pos­si­ble, but that the “online” spaces were a venue where such com­mu­ni­ca­tion did not have the same  con­se­quences. Some­how, the speech was hid­den from those in power. It was a trusted environment.

Now while the logic of net­works is a good way to explain easy com­mu­ni­ca­tion, it doesn’t lend itself to dis­cus­sions of trust. For­tu­nately soci­ol­o­gists have long been aware of the impor­tance of these “free spaces” in which dis­sent­ing voices can com­mu­ni­cate. Here are Francesca Pol­letta and James Jasper in a 2001 paper:

Con­cepts of “sub­merged net­works”, “halfway houses”, “free spaces”, “havens”, “sequestered social sites”, and “abeyance struc­tures” describe insti­tu­tions removed from the phys­i­cal and ide­o­log­i­cal con­trol of those in power, for exam­ple the black church before the civil rights move­ment and lit­er­ary cir­cles in com­mu­nist East­ern Europe. Such insti­tu­tions… rep­re­sent a “free space” in which peo­ple can develop coun­ter­hege­monic ideas and oppo­si­tional identities.

So these notions of “free spaces” have been around for some time and surely fit some­thing about the way that online polit­i­cal dis­cus­sion worked in Egypt. Free spaces are insti­tu­tions (in a broad sense of the term) that are not out­lawed, but which appeal to out­siders of soci­ety rather than to those who iden­tify with the powers-that-be. They man­age to be trans­par­ent to their mem­bers while being opaque to officialdom.

More gen­er­ally, fol­low­ing Charles Tilly and Sid­ney Tar­row, we can think of insti­tu­tions in  author­i­tar­ian states as being of three kinds.

• Pre­scribed insti­tu­tions are the main­stream and estab­lish­ment insti­tu­tions of soci­ety. They may include the edu­ca­tion sys­tem, orga­ni­za­tions like the army, and also things like national cel­e­bra­tions. Some of these insti­tu­tions include peo­ple of all lev­els of sta­tus, while some are restricted to high-status indi­vid­u­als and families.
• Tol­er­ated insti­tu­tions are legal, but their mem­ber­ship is lim­ited to lower-status indi­vid­u­als. In some coun­tries these would include reli­gious insti­tu­tions asso­ci­ated with minor­ity groups, per­haps some artis­tic and cul­tural insti­tu­tions, and work­place orga­ni­za­tions in coun­tries where they can exist out­side offi­cial con­trol. These are the venues that, accord­ing to Pol­letta and Jasper, can pro­vide spaces for dis­sent. Obvi­ously there is a wide range of what insti­tu­tions are tol­er­ated and what are for­bid­den. North Korea has a lot fewer “tol­er­ated” insti­tu­tions than 1980 Poland.
• For­bid­den insti­tu­tions are those that are not per­mit­ted in author­i­tar­ian soci­eties. Oppo­si­tion polit­i­cal par­ties, inde­pen­dent unions, that sort of thing.

But how do these insti­tu­tions become “removed from the phys­i­cal and ide­o­log­i­cal con­trol of those in power”? The answer lies in what Pol­letta & Jasper call “col­lec­tive iden­tity”. Tol­er­ated insti­tu­tions –whether sub­cul­tures, groups, or what­ever – build up their own prac­tices to estab­lish autonomy.

Col­lec­tive iden­tity is “an individual’s cog­ni­tive, moral, and emo­tional con­nec­tion with a broader com­mu­nity, cat­e­gory, prac­tice, or insti­tu­tion.” It gets expressed in “cul­tural materials—names, nar­ra­tives, sym­bols, ver­bal styles, rit­u­als, cloth­ing, and so on.” And these expres­sions pro­vide boundary-setting rit­u­als and insti­tu­tions that sep­a­rate chal­lengers from those in power, and so can strengthen inter­nal solidarity.

Exam­ples of “free spaces” in author­i­tar­ian soci­eties abound. In his book Exit-Voice Dynam­ics and the Col­lapse of East Ger­many, Steven Pfaff high­lights the impor­tance of some very nar­row insti­tu­tions that he calls “Niche soci­ety”. These are “pock­ets of pri­vate life, around home, car and allot­ment” where peo­ple could voice their dis­en­chant­ment and cyn­i­cism. A broader form of dis­sent took place in insti­tu­tions of youth cul­ture: despite party efforts to estab­lish bands and music venues for Ger­man youth, many sought out more alter­na­tive forms of music, and clashes took place  between fans and police at con­certs. Music events are not, at least pub­licly, polit­i­cal events and so while the events might not be for­bid­den, you would not find party sup­port­ers tak­ing part. Finally, Pfaff notes that “Dis­sent could only take place in gaps in the sys­tem of social con­trol that dis­si­dents could exploit. In the GDR this prin­ci­pally meant the churches.” Again, churches are an exam­ple of an insti­tu­tion that was legal, but which nat­u­rally sep­a­rated society’s out­siders from those in power.

So now we seem to have two sep­a­rate sets of ideas. On the one hand we have a the­ory of upris­ings that makes no use of soci­o­log­i­cal con­cepts. On the other hand, to explain pre-uprising dis­sent we need to look at soci­o­log­i­cal ideas such as insti­tu­tions and iden­tity. Obvi­ously there is a bridge that must be built if we are to con­nect these seem­ingly sep­a­rate the­o­ret­i­cal islands. Can the gap be bridged? Well yes it can, thanks to the “iden­tity eco­nom­ics” work of George Akerlof and Rachel Kran­ton, who argue that iden­tity pro­vides a key moti­va­tion for many social sit­u­a­tions. They  take the con­cept of iden­tity seri­ously, and sim­plify it to fit it into a tractable micro-level model. Iden­tity, they say, has three parts to it.

• First is a set of social cat­e­gories: for us, those cat­e­gories are “gov­ern­ment sup­porter” or “opponent”.
• Next, each of these iden­ti­ties has a set of attrib­utes asso­ci­ated with it. These vary from soci­ety to soci­ety. Eco­nomic sta­tus is one, reli­gious or eth­nic or gen­der iden­ti­ties are others.
• And finally, each iden­tity has a set of norms of behav­ior: in this case we sim­plify the options to “con­form” to society’s expec­ta­tions or “dissent”.

Indi­vid­u­als then have two choices to make. First, they need to adopt an iden­tity: Govern­ment or Oppo­si­tion? Next, if they are oppo­si­tional they need to decide whether to engage in active dis­sent or to con­form to offi­cial expec­ta­tions. If we arrange the pop­u­la­tion accord­ing to sta­tus, then those at the lower sta­tus choose O (O has a higher util­ity), and peo­ple with higher sta­tus choose G. Here is a graph that shows a case where the switchover appears at the mid-point.

Util­ity and iden­tity in an author­i­tar­ian society

Iden­tity is one of the two things we need to explain “free spaces” but before we go to the other, let’s take a short detour. One of the key suc­cesses of information-driven ratio­nal choice mod­els was the fact that they yield cas­cades. Can our identity-driven model also give cas­cades? Funny you should ask…

Here is a cascade.

A cas­cade. The yel­low line and the yel­low dot are equi­lib­ria of the model.

If you want to know the gory details, includ­ing what the “hege­mony” label on the x axis means, you have to go and read the paper. But see that there are two equi­lib­ria here. One is a sta­ble author­i­tar­ian state, with zero activ­ity (the yel­low line at the right) and a high gov­ern­ment hege­mony. The other is a state in cri­sis, with a high level of dis­sent (the yel­low dot where the lines cross). And a small change in soci­ety can lead to a sud­den  dis­con­tin­u­ous switch from one to the other: a cas­cade.

To gen­er­ate mul­ti­ple equi­lib­ria you need some form of exter­nal­ity: some way in which one person’s actions influ­ence those around them. This model gen­er­ates cas­cades by assert­ing that active dis­sent increases the iden­tity polar­iza­tion of soci­ety: the more active dis­sent there is, the more it mat­ters which side you are on. It’s not so much an infor­ma­tion cas­cade as an iden­tity cas­cade.

Although this is a ratio­nal choice model, it does not invoke net­works, and infor­ma­tion is not cen­tral to the argu­ment. In most cas­cade mod­els the cas­cade is gen­er­ated by two things:

• active dis­sent reveals infor­ma­tion, about the state of the soci­ety or about the beliefs of other peo­ple. This is the “pref­er­ence fal­si­fi­ca­tion” argument.
• there is safety in num­bers: the more peo­ple protest­ing, the safer it is to protest.

I’ve crit­i­cized these ideas here, but is there any evi­dence to sug­gest that iden­tity does get polar­ized as a result of dis­sent? Anec­do­tally, there is. Here is a Marx­ist rad­i­cal speak­ing about Paris, 1968:

I was com­pletely sur­prised by 1968… I had an idea of the rev­o­lu­tion­ary process and it was noth­ing like this. I saw stu­dents build­ing bar­ri­cades, but these were peo­ple who knew noth­ing of rev­o­lu­tion. They were not even polit­i­cal. There was no organ­i­sa­tion, no planning.

In the lead-up to 1968, French stu­dents were not rev­o­lu­tion­ar­ies who had fal­si­fied their true pref­er­ences in order to con­form to society’s expec­ta­tions. What hap­pened was that dur­ing the riots, iden­tity (sta­tus quo or rad­i­cal?) became a cen­tral issue, and indi­vid­u­als had to decide “which side are you on?”, and many stu­dents switched their iden­ti­ties from mildly status-quo to enthu­si­as­tic barricade-builder.

A switch in iden­tity hap­pens when peo­ple are pulled along by those around them. As Den­nis Chong (1991) writes of the US Civil Rights move­ment: “friend­ship and famil­ial, reli­gious, and pro­fes­sional rela­tion­ships cre­ate an array of ongo­ing exchanges, oblig­a­tions and expec­ta­tions that individual.”

In his book on the fall of the GDR, Steven Pfaff repeat­edly invokes the “pref­er­ence fal­si­fi­ca­tion” model, but he often steps out­side it too. In fact, my biased read­ing of it is that he some­times resorts to the pref­er­ence fal­si­fi­ca­tion model because of a lack of alter­na­tives, not because the evi­dence pulls him that way. But when he writes that “By 1989 offi­cial social­ist ide­ol­ogy, along with its clear artic­u­la­tion of the nature of injus­tice, had become a threat to the sys­tem it was meant to legit­i­mate” he is talk­ing about a cri­sis of iden­tity. The cri­sis served to “focus dif­fuse griev­ances”, unit­ing “a host of dis­parate con­cerns into ‘moral anger“‘. This is the crys­tal­liza­tion of iden­ti­ties into the two polar choices: “Which side are you on?”

The “iden­tity cas­cade” model also makes a closer con­nec­tion between dynam­ics and the efforts of pro­test­ers. I’ll return to this later, but one of the things that pro­test­ers do in upris­ings is lay claim to the sym­bols of national iden­tity. Whether it’s Gandhi’s Salt March or GDR pro­test­ers choos­ing the 40th anniver­sary of the found­ing of the coun­try, strug­gles over the mean­ing of iden­tity become cen­tral at times of cri­sis. If infor­ma­tion rev­e­la­tion was all that were needed, there would be no role for the dis­plays of “wor­thi­ness, unity, num­bers and com­mit­ment” that char­ac­ter­ize polit­i­cal protest. An identity-driven approach makes this link clear, within a ratio­nal choice framework.

(Another nice thing is that within the iden­tity cas­cade model there is a nat­ural cat­e­go­riza­tion of the kind of events that can pre­cip­i­tate a cas­cade. A shock to the norms asso­ci­ated with oppo­si­tion, a change in socio-economic con­di­tions that places more peo­ple into the “out­sider” cat­e­gory, or a change in state pol­icy (per­e­stroika) all emerge as trig­gers for cas­cades. See the paper for more.)

With that diver­sion over, let’s return to the topic of free spaces. How do we get from the lan­guage of Pol­letta and Jasper to the world of ratio­nal choice? There is a nat­ural cor­re­spon­dence in the con­cept of screen­ing: a mech­a­nism that imposes dif­fer­en­tial costs for two dif­fer­ent groups, so that (in a “sep­a­rat­ing” equi­lib­rium) one group finds it worth­while to pay the cost, while the other does not. Here, the identity-driven costs of being a mem­ber of “tol­er­ated” insti­tu­tion screens out those with the sta­tus quo (G) iden­tity.

Just as Akerlof and Kran­ton sim­pli­fied iden­tity so that it could be squeezed into a ratio­nal choice pic­ture, so we have to sim­plify the idea of an insti­tu­tion. Hence­forth, then, an insti­tu­tion I is char­ac­ter­ized by three things:

• Sta­tus (x): This is the nat­ural mem­ber­ship of the insti­tu­tion. We can say that the iden­tity of the insti­tu­tion is the opti­mal iden­tity of an indi­vid­ual with sta­tus x
• Breadth (δ): Indi­vid­u­als with sta­tus in [x — δ, x + δ] are mem­bers of I. The “niche soci­ety” insti­tu­tions of the GDR have a very nar­row breadth, while events such as national cel­e­bra­tions include all of society.
• Mem­ber­ship dis­crim­i­na­tion (m): Some insti­tu­tions do not dis­crim­i­nate between the two iden­ti­ties, but some do. A dis­crim­i­nat­ing insti­tu­tion demands a cost of mem­ber­ship for indi­vid­u­als whose iden­tity dif­fers from the iden­tity of the institution.

With this idea, you can build a model in which there is a range of insti­tu­tions that even a strong state will not mon­i­tor, because the cost of mon­i­tor­ing is greater than the ben­e­fit in terms of dis­sent that is qui­eted. These insti­tu­tions pro­vide the free space for dis­sent to per­sist even under con­di­tions of strong government.

Here are some screen­ing institutions

Screen­ing institutions.

The screen­ing insti­tu­tions are those inside the lozenge shapes. Along the x axis is the sta­tus, so all these insti­tu­tions are “tol­er­ated” in that they are entirely within the “out­sider” low-status zone. The broader the reach of the insti­tu­tion (that’s the “δ” in the graph) the less scope there is for these insti­tu­tions. Finally, and it’s beyond what I can explain in this part, there is a limit to the size of the “pub­lic sphere” that also lim­its the avail­able institutions.

So what this shows is that the eco­nomic con­cept of screen­ing brings to the identity-driven ratio­nal choice model the idea of free spaces, well estab­lished within the soci­o­log­i­cal lit­er­a­ture. To go back to the begin­ning of this essay, the exis­tence of such spaces is some­thing that the net­work mod­els, with their focus on costs of com­mu­ni­ca­tion, don’t seem well equipped to describe. So now we have a sin­gle the­ory that cov­ers both upris­ings and pre-revolutionary dis­sent, instead of two (one micro, one macro). We can now see that the “free spaces” of online dis­sent are sim­i­lar to, and exist for the same rea­sons as, other free spaces that have existed in the past. Even in Egypt, the role of the Ultras foot­ball fans can fit within this model, the foot­ball sta­dium ter­races pro­vid­ing a “tol­er­ated” insti­tu­tion within which dis­sent could be expressed. The model also argues that the key facet of online spaces is not their tech­no­log­i­cal nature, but the fact that they were adopted by, and asso­ci­ated with, the broadly anti-establishment demo­graphic of urban youth. Nav­i­gat­ing the dis­cus­sion spaces of the online world is easy if you have friends who are tak­ing part: not so easy if you are a gov­ern­ment offi­cial try­ing to pose as a dis­en­fran­chised youth. The tech­nol­ogy of social media is epiphe­nom­e­nal. In broad strokes, this is an argu­ment I made some time ago here: it’s only taken two years for me to work it out properly.

Insti­tu­tions and Challenges

The final case to look at is when a social move­ment chal­lenges a weak gov­ern­ment. The goal is to put the gov­ern­ment in a “dictator’s dilemma”. The idea that clamp­ing down on dis­sent has the pos­si­bil­ity of draw­ing atten­tion to it, and per­haps fan­ning the flames, is an old one. Here is a recent statement:

[S]ometimes repres­sion inspires more mobi­liza­tion; and some­times it effec­tively quashes move­ments or pushes them under­ground. Some­times repres­sive forces are suc­cess­ful in char­ac­ter­iz­ing pro­test­ers as legit­i­mate tar­gets of repres­sion, and other times they delig­itimize the State and increase the legit­i­macy of the social move­ments.
– Cristina Flesher Fom­i­naya and Les­ley Wood

Or, going back a lit­tle further:

Cen­sor­ship makes every banned text, bad or good, into an extra­or­di­nary text.
– Karl Marx

The con­trast­ing for­tunes of the GDR protests and the Tien­an­men Square protests in 1989 are the best known exam­ple of this dual possibility.

When they believe the time is right, social move­ments may actively seek to pro­voke a cri­sis (con­trary to the “safety in num­bers” cost min­i­miza­tion that the infor­ma­tion cas­cade the­o­rists tend to favour). Famously, here is Gandhi:

The func­tion of a civil resis­tance is to pro­voke response and we will con­tinue to pro­voke until they respond or change the law. – M. K. Gandhi

We can bring this idea of provo­ca­tion into a micro model if we bring in a uni­tary social move­ment and invoke an inter­de­pen­dency between iden­tity polar­iza­tion and gov­ern­ment coer­cion. Again, the math­e­mat­ics is in the paper.

The ques­tion we ask is “if you were an orga­nized oppo­si­tion, what insti­tu­tion would you tar­get, so that a clam­p­down would cause polar­iza­tion?” The idea is that clam­p­down on a main­stream insti­tu­tion would be more likely to polar­ize soci­ety, by dis­turb­ing even the government’s own sup­port­ers, than clamp­ing down on an “out­sider” insti­tu­tion. Again, the oppo­si­tion has to make a pay­ment to appro­pri­ate a main­stream insti­tu­tion, because of mem­ber­ship selec­tiv­ity. They have to pass with the iden­tity of a sta­tus quo sup­porter. They need to appeal to main­stream sen­si­bil­i­ties and to estab­lish legit­i­macy. Under the right cir­cum­stances, an oppo­si­tion will pay the cost of provo­ca­tion, because they antic­i­pate that a gov­ern­ment response will weaken, not strengthen, the government’s level of con­trol. Here is a fig­ure show­ing a set of insti­tu­tions that can be used by an oppo­si­tion to pro­voke a crisis.

Insti­tu­tions that may pro­voke a crisis

The insti­tu­tions that may pro­voke a cri­sis are those within the cen­tral closed shape, bounded clock­wise by light blue (on top), green, red, and pur­ple. Some of these insti­tu­tions are “tol­er­ated” insti­tu­tions to the left of the x* line, with oppo­si­tional iden­ti­ties; for these insti­tu­tions there is no mem­ber­ship cost to be paid by the oppo­si­tion. Oth­ers are “pre­scribed” insti­tu­tions that have a main­stream iden­tity. The oppo­si­tion must pay the price of appro­pri­at­ing these insti­tu­tions: par­tic­i­pat­ing in them in such a way as to pro­voke the government.

An exam­ple of this behav­iour comes again from the GDR upris­ings of 1989, as described by Steven Pfaff. The oppo­si­tion chose the cel­e­bra­tions of the GDR’s for­ti­eth anniver­sary – a main­stream insti­tu­tion – in which to pro­voke a response. The gov­ern­ment did respond, but “its bru­tal attacks on peace­ful pro­test­ers dur­ing the for­ti­eth anniver­sary … prob­a­bly acti­vated what might have oth­er­wise remained despair­ing, but inert, citizens.”

The oppo­si­tion made explicit attempts to por­tray them­selves as main­stream Ger­mans, adopt­ing the sim­ple slo­gan of “Wir sind das volk” (“We are the people”).

“Wir sind das volk” [was] a thin claim, but an uncom­pli­cated “us ver­sus them” mes­sage, a claim to polit­i­cal iden­tity that could bridge lines of class, edu­ca­tion, neigh­bor­hood, and so on. – Steven Pfaff

In pre­vi­ous times, other upris­ings have explic­itly cho­sen main­stream or some­times tol­er­ated insti­tu­tions as a means of provo­ca­tion. Gandhi’s use of the Salt March, the Chi­nese stu­dents’ use of the death of Hu Yaobang and Tien­an­men Square, the Egypt­ian pro­test­ers appro­pri­a­tion of National Police Day and Tahrir Square all fol­low this pattern.

There are claims that dig­i­tal tech­nolo­gies at times of cri­sis can act in this man­ner. Ethan Zuck­er­man has pop­u­lar­ized the idea as a “Cute Cat” the­ory: that main­stream insti­tu­tions pro­vide a venue for dis­sent that can­not be shut down with­out polar­iz­ing soci­ety. The the­ory here pro­vides at best lim­ited and con­di­tional sup­port for the idea. Dig­i­tal tech­nolo­gies were not used as a mech­a­nism of provo­ca­tion, but played a sup­port­ing role. The “cute cat” idea has cre­dence only if the gov­ern­ment is not able to silence dis­sent in a more selec­tive man­ner than shut­ting down the entire inter­net or phone ser­vice within the country.

My favourite exam­ple is the French “Ban­quet Cam­paign” of 1848. Repub­li­cans were cam­paign­ing for uni­ver­sal male suf­frage against an intran­si­gent gov­ern­ment that had banned polit­i­cal meet­ings. Faced with the prob­lem of orga­niz­ing an oppo­si­tion in such an envi­ron­ment, they orga­nized ban­quets. On the 18th of July in Mâcon, Bur­gundy, five hun­dred tables were set up for three thou­sand guests with stands for three thou­sand more, osten­si­bly as a cel­e­bra­tion of local lit­er­ary star Alphonse de Lamar­tine. Lamar­tine was not just a lit­er­ary star though, he was also a well-known repub­li­can, and the author­i­ties knew that the ban­quet was a cover for polit­i­cal agi­ta­tion. But the author­i­ties judged that inter­fer­ing with the ban­quets would inflame the sit­u­a­tion rather than suc­ceed in sup­press­ing the protest, and so let the ban­quet pro­ceed. With the suc­cess of the Mâcon ban­quet, the “Cam­pagne des ban­quets” was launched, and ban­quets were held around the coun­try. This is the high wire act that gov­ern­ments and oppo­si­tion walk at times of cri­sis – when to push ahead, when to hold back, and what tac­tics may be effec­tive – and is the kind of dance that social move­ment stud­ies have helped to elu­ci­date. The cam­paign con­tin­ued until Feb­ru­ary of the next year, when the gov­ern­ment decided it had no choice but to esca­late. The ban­quets were out­lawed, a hastily orga­nized protest brought peo­ple into the streets of Paris on Feb­ru­ary 22, a con­fronta­tion between the Munic­i­pal Guard and the marchers spilled over into riots, every­thing got out of hand, and the King fled Paris. Within a few weeks gov­ern­ments were top­pled in Milan, Venice, Naples, Palermo, Vienna, Prague, Budapest, Krakow, and Berlin. I like to think that the graph above cap­tures a lit­tle of that drama.

Con­clu­sions

What I’ve tried to do here is fol­low the Akerlof & Kran­ton exam­ple of tak­ing the rich soci­o­log­i­cal con­cepts of iden­tity seri­ously, and used it to con­struct a ratio­nal choice model of upris­ings that com­ple­ments, rather than com­petes with, soci­o­log­i­cal mod­els. I’ve added some dynam­ics to the approach, and brought in a mod­el­ling of insti­tu­tions to build on the notion of col­lec­tive iden­tity as a moti­vat­ing force for protest.

The results are that the the­ory recov­ers the key facet of other ratio­nal choice mod­els of upris­ings, which is cas­cades, but with a dif­fer­ent inter­pre­ta­tion. Here it is “iden­tity cas­cades” rather than “infor­ma­tion cas­cades” that drive the sud­den change. Beyond cas­cades, the the­ory shows how screen­ing pro­vides a mech­a­nism for the exis­tence of “free space” insti­tu­tions in which dis­sent can be sus­tained, even in author­i­tar­ian regimes. Finally, it shows how an orga­nized oppo­si­tion may appro­pri­ate main­stream insti­tu­tions with the explicit intent of pro­vok­ing a cri­sis, putting the gov­ern­ment in a “dictator’s dilemma” in which nei­ther respond­ing nor fail­ing to respond is a good option.