Whimsley3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

Whimsley
…where Tom Slee writes about technology and politics

Notes on Identity, Institutions, and Uprisings
Table of Contents
Intro uc ion d t Face ook as a “free space” b Con ect ng Iden ity to Ratio al Choice? n i t n Iden ity Cascades t Free Spaces and Screening Insti u ions and Challenges t t Con lu ions c s

Intro uc ion d t
Fin sh ng up what I said I’d fin sh a cou le of months ago, this is a shorter ver ion of a i i i p s paper on “Iden ity, Insti u ions, and Upris ngs” with less math at cs, no ref r nces (see t t t i em i e e the link above) and more opin on t ng. Also, a longer ver ion of what I’m going to say at i ai s The iz ng the Web 2013 in a few days. or i There is a the et al side to the “Face ook Rev u ion” debate about the role of dig al or ic b ol t it tech olo ies in the 2011 “Arab Spring” upris ngs, and it boils down to two ways of look ng n g i i at things: the micro and the macro. On the one hand, we have the ratio al choice, agentn based approach and on the other we have more tra i ional soci og al approaches based dt ol ic on larger-scale social structures. If you look at some of the key char c er s ics of the upris ngs, it looks like a win for the a t i t i tomslee.net/2013/02/503.html 1/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

micro side.

The ies, and North African upris ngs. or i
EVENT MICRO MACRO

Sud en upris ng (cascade) d i Lack of strong oppo i ion movement st Net ork technologies w Score

Y Y Y 3

N N N 0

The sin le most dra atic thing about the “Arab Spring” upris ngs was their unex ected g m i p sud en ess. They fit the “infor a ion cas ade” mod ls devel ped by Timur Kuran, d n m t c e o Suzanne Lohmann and oth rs to describe the equally dra atic and sud en 1989 upris ngs e m d i in East rn Europe. Noth ng on the “macro” side matches the ele ant expla a ion of sud e i g n t den, dis on in us change given by the micro-theorists. c t uo Related to this sud en ess is the lack of a strong oppo i ion move ent before the upris d n st m ing. It’s not that there was no oppo i ion, but there was noth ng of the strength to indi ate st i c a com ng cri is. The cas ade the ies have no need, or even place, for orga i a ions or i s c or nz t move ents: these are pop a ion dynam cs mod ls, with no struc ure big er than the net m ul t i e t g worked indi id al. Mean hile, the lead ng soci og al the ies deal with move v u w i ol ic or ments, orga i a ions, and resource mobi iza ion. Score two to the micro-theorists. nz t l t Finally, we have the role of dig al tech olo ies, which segues nat ally to net ork mod ls it n g ur w e of soci ty. Talk of infor a ion tech olo ies leads equally nat ally to a focus on infor a e m t n g ur m tion dif u ion across net orks, in which increased con ec iv ty low r ng bar i rs to col ab f s w n t i e i re l o a ion, dis us ion, and infor a ion shar ng. And the macro-theorists again seem to have r t c s m t i lit le on their side to cope with these kinds of ideas. t It looks like a shut-out win for the micro-theorists; the lan uage of net orks and infor a g w m tion replaces the lan uage of social move ents and reper oires of per or ance, and with g m t f m that comes the inevitable idea that with a new kind of the ry we are we see ng a new kind o i tomslee.net/2013/02/503.html 2/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

of upris ng, in which self-organizing net orks based on dig al tech olo ies take cen i w it n g tre stage… But you will have real zed by now that this is a setup for me to argue that there’s another i way of look ng at these events, so let’s get to it. i The key suc ess of micro-level the ry is the expla a ion of cas ades, which is a nat ral c o n t c u con e uence of any model that has mul i le equi ib ia. Just because of that suc ess, we s q tp l r c 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 con epts that soci og al c ol ic research has shown to be impor ant, and move them into the realm of ratio al choice t n mod ls. And when we do, we not only get pop a ion dynam cs and cas ades, but we also e ul t i c get expla a ions for sev ral other aspects of dis ent and upris ngs that net orks n t e s i w and information-based the ies can’t deliver. or Is there a down ide? Of course there is. Behind the scenes, it’s often the case that ratio al s n choice the ists like long equa ions while soci l ists love long words. Ratio al choice or t o og n the ists see the soci l ists’ con epts as fuzzy, while the soci l ists see the incen ives of or o og c o og t ratio al choice mod ls as sim lis ic. What I have to offer demands both long equa ions n e p t t and long words, and is open to being crit ism for being simul a e usly sim lis ic ic t n o p t and fuzzy. Ah well.

Face ook as a “free space” b
Let’s start with a ques ion. Zeynep Tufekci is a soci l ist who was in Egypt right after the t o og Jan ry 2011 upris ngs, inter iew ng par ic ants. Here’s what she says: ua i v i t ip

When I was in post­Mubarak Cairo, my hosts kept point ng in amaze ent to var i m i­ ous street cor ers where fierce polit al dis us ions were being held and often n ic c s whis ered, before remem er ng they could now speak up and adjust ng their voice, p b i i “You never saw this. Nobody ever dis ussed pol ics openly, ever.” Then they c it would pause and add, “Well, except online, of course. We all dis ussed pol ics c it online.” tomslee.net/2013/02/503.html 3/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

So the ques ion is that final sen ence. Why is it that, prior to the rev u ion, peo le could t t ol t p dis uss pol ics online but not else here? What made “online” a venue where those dis c it w cus ions could take place? It’s not just ease of com u i a ion, because if you want to com s m nc t mu i ate you can stand on a busy street cor er – as peo le were doing when Tufekci nc n p visited. The key thing is that com u i a ion online was, for some rea on, safe, while com u i a m nc t s m nc tion on the street was not. It’s not just that com u i a ion among like-minded peo le was m nc t p pos i le, but that the “online” spaces were a venue where such com u i a ion did not sb m nc t have the same con e uences. Some ow, the speech was hid en from those in power. It s q h d was a trusted environment. Now while the logic of net orks is a good way to explain easy com u i a ion, it doesn’t w m nc t lend itself to dis us ions of trust. For u ately soci l ists have long been aware of the c s t n o og impor ance of these “free spaces” in which dis ent ng voices can com u i ate. Here are t s i m nc Francesca Pol etta and James Jasper in a 2001 paper: l

Con epts of “sub erged net orks”, “halfway houses”, “free spaces”, c m w “havens”, “sequestered social sites”, and “abeyance struc ures” describe insti u t t ­ tions removed from the phys al and ide og al con rol of those in power, ic o l ic t for exam le the black church before the civil rights move ent and lit r ry cir p m e a ­ cles in com u ist East rn Europe. Such insti u ions… rep e ent a “free space” m n e t t r s in which peo le can develop coun er ege onic ideas and oppo i ional identities. p t h m st

So these notions of “free spaces” have been around for some time and surely fit some hing t about the way that online polit al dis us ion worked in Egypt. Free spaces are insti u ic c s t tions (in a broad sense of the term) that are not out awed, but which appeal to out iders of l s soci ty rather than to those who iden ify with the powers-that-be. They man ge to be e t a trans ar nt to their mem ers while being opaque to officialdom. p e b More gen r lly, fol ow ng Charles Tilly and Sid ey Tar ow, we can think of insti u ions in e a l i n r t t tomslee.net/2013/02/503.html 4/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

author ar an states as being of three kinds. it i

Types of insti u ion in author ar an states t t it i
INSTI U ION T T HIGH STA US T LOW STA US T

Pre cribed s Tol r ted e a For id en b d

Y N N

Y/N Y N

Pre cribed insti u ions are the main tream and estab ish ent insti u ions of soci ty. s t t s l m t t e They may include the edu a ion sys em, orga i a ions like the army, and also things c t t nz t like national cel ra ions. Some of these insti u ions include peo le of all lev ls of eb t t t p e sta us, while some are restricted to high-status indi id ls and families. t v ua Tol r ted insti u ions are legal, but their mem er hip is lim ted to lower-status indi e a t t b s i vid ls. In some coun ries these would include reli ious insti u ions asso i ted with ua t g t t ca minor ty groups, per aps some artis ic and cul ural insti u ions, and work lace orga i h t t t t p ni a ions in coun ries where they can exist out ide offi ial con rol. These are the z t t s c t venues that, accord ng to Pol etta and Jasper, can pro ide spaces for dis ent. Obvi i l v s ously there is a wide range of what insti u ions are tol r ted and what are for id en. t t e a b d North Korea has a lot fewer “tol r ted” insti u ions than 1980 Poland. e a t t For id en insti u ions are those that are not per it ed in author ar an soci ties. b d t t m t it i e Oppo i ion polit al par ies, inde en ent unions, that sort of thing. st ic t p d But how do these insti u ions become “removed from the phys al and ide og al con rol t t ic ol ic t of those in power”? The answer lies in what Pol etta & Jasper call “col ec ive iden ity”. l l t t Tol r ted insti u ions –whether sub ul ures, groups, or what ver – build up their own e a t t c t e prac ices to estab ish autonomy. t l Col ec ive iden ity is “an individual’s cog i ive, moral, and emo ional con ec ion with a l t t nt t n t broader com u ity, cat ory, prac ice, or insti u ion.” It gets expressed in “cul ural m n eg t t t t materials—names, nar a ives, sym ols, ver al styles, rit ls, cloth ng, and so on.” And r t b b ua i these expres ions pro ide boundary-setting rit ls and insti u ions that sep ate chal s v ua t t ar tomslee.net/2013/02/503.html 5/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

lengers from those in power, and so can strengthen inter al solidarity. n Exam les of “free spaces” in author ar an soci ties abound. In his book Exit­Voice p it i e Dynam cs and the Col apse of East Ger any, Steven Pfaff high ights the impor ance of i l m l t some very nar ow insti u ions that he calls “Niche soci ty”. These are “pock ts of pri ate r t t e e v life, around home, car and allot ent” where peo le could voice their dis n hant ent and m p e c m cyn ism. A broader form of dis ent took place in insti u ions of youth cul ure: ic s t t t despite party efforts to estab ish bands and music venues for Ger an youth, many sought l m out more alter a ive forms of music, and clashes took place between fans and police at n t con erts. Music events are not, at least pub icly, polit al events and so while the events c l ic might not be for id en, you would not find party sup ort rs tak ng part. Finally, Pfaff b d p e i notes that “Dis ent could only take place in gaps in the sys em of social con rol that dis i s t t sdents could exploit. In the GDR this prin i ally meant the churches.” Again, churches cp are an exam le of an insti u ion that was legal, but which nat ally sep ated society’s p t t ur ar out iders from those in power. s

Con ect ng Iden ity to Ratio al Choice? n i t n
So now we seem to have two sep ate sets of ideas. On the one hand we have a the ry of ar o upris ngs that makes no use of soci og al con epts. On the other hand, to explain prei ol ic c uprising dis ent we need to look at soci og al ideas such as insti u ions and iden ity. s ol ic t t t Obvi usly there is a bridge that must be built if we are to con ect these seem ngly sep o n i arate the et al islands. Can the gap be bridged? Well yes it can, thanks to the “iden ity or ic t eco om cs” work of George Akerlof and Rachel Kran on, who argue that iden ity pro ides n i t t v a key moti a ion for many social sit ions. They take the con ept of iden ity seri usly, v t uat c t o and sim lify it to fit it into a tractable micro-level model. Iden ity, they say, has three p t parts to it. First is a set of social cat ories: for us, those cat ories are “gov rn ent sup orter” eg eg e m p or “opponent”. Next, each of these iden i ies has a set of attrib tes asso i ted with it. These vary tt u ca from soci ty to soci ty. Eco omic sta us is one, reli ious or eth ic or gen er iden i e e n t g n d tties are others. And finally, each iden ity has a set of norms of behav or: in this case we sim lify the t i p tomslee.net/2013/02/503.html 6/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

options to “con orm” to society’s expec a ions or “dissent”. f t t Indi id ls then have two choices to make. First, they need to adopt an iden ity: Govern v ua t ment or Oppo i ion? Next, if they are oppo i ional they need to decide whether to engage st st in active dis ent or to con orm to offi ial expec a ions. If we arrange the pop a ion s f c t t ul t accord ng to sta us, then those at the lower sta us choose O (O has a higher util ty), and i t t i peo le with higher sta us choose G. Here is a graph that shows a case where the p t switchover appears at the mid-point.

Util ty and iden ity in an author ar an society i t it i

In some times and places, no one gives a hoot which iden ity you adopt, while at other t times and places it can be a mat er of life or death. I’ll call this scal ng of the dif er nce t i f e between O and G the iden ity polar za ion of soci ty, and we’ll be need ng this con ept t i t e i c a lot. Iden ity is one of the two things we need to explain “free spaces” but before we go to the t other, let’s take a short detour. One of the key suc esses of information-driven ratio al c n choice mod ls was the fact that they yield cas ades. Can our identity-driven model also e c give cas ades? Funny you should ask… c

tomslee.net/2013/02/503.html

7/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

Iden ity Cascades t
Here is a cascade.

A cas ade. The yel ow line and the yel ow dot are equi ib ia of the model. c l l l r

If you want to know the gory details, includ ng what the “hege ony” label on the x axis i m means, you have to go and read the paper. But see that there are two equi ib ia here. One l r is a sta le author ar an state, with zero activ ty (the yel ow line at the right) and a high b it i i l gov rn ent hege ony. The other is a state in cri is, with a high level of dis ent (the yel e m m s s low dot where the lines cross). And a small change in soci ty can lead to a sud en dis on e d c tin us switch from one to the other: a cas ade. uo c To gen r te mul i le equi ib ia you need some form of exter al ty: some way in which one e a tp l r n i person’s actions influ nce those around them. This model gen r tes cas ades by assert ng e e a c i that active dis ent increases the iden ity polar za ion of soci ty: the more active dis ent s t i t e s there is, the more it mat ers which side you are on. It’s not so much an infor a ion cas t m t cade as an iden ity cas ade. t c Although this is a ratio al choice model, it does not invoke net orks, and infor a ion is n w m t not cen ral to the argu ent. In most cas ade mod ls the cas ade is gen r ted by two t m c e c e a tomslee.net/2013/02/503.html 8/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

things: active dis ent reveals infor a ion, about the state of the soci ty or about the beliefs s m t e of other peo le. This is the “pref r nce fal i i a ion” argument. p e e sfc t there is safety in num ers: the more peo le protest ng, the safer it is to protest. b p i I’ve crit ized these ideas here, but is there any evi ence to sug est that iden ity does get ic d g t polar zed as a result of dis ent? Anec o ally, there is. Here is a Marx st rad al speak ng i s d t i ic i about Paris, 1968:

I was com letely sur rised by 1968… I had an idea of the rev u ion ry process p p ol t a and it was noth ng like this. I saw stu ents build ng bar i ades, but these were i d i rc peo le who knew noth ng of rev u ion. They were not even polit al. There was p i ol t ic no organ a ion, no planning. is t

In the lead-up to 1968, French stu ents were not rev u ion r es who had fal i ied their d ol t a i sf true pref r nces in order to con orm to society’s expec a ions. What hap ened was that e e f t t p dur ng the riots, iden ity (sta us quo or rad al?) became a cen ral issue, and indi id ls i t t ic t v ua had to decide “which side are you on?”, and many stu ents switched their iden i ies from d tt mildly status-quo to enthu i s ic barricade-builder. sa t A switch in iden ity hap ens when peo le are pulled along by those around them. As Den t p p nis Chong (1991) writes of the US Civil Rights move ent: “friend hip and famil al, reli m s i gious, and pro es ional rela ion hips cre te an array of ongo ng exchanges, oblig ions and f s t s a i at expec a ions that individual.” t t In his book on the fall of the GDR, Steven Pfaff repeat dly invokes the “pref r nce fal i i a e e e sfc tion” model, but he often steps out ide it too. In fact, my biased read ng of it is that he s i some imes resorts to the pref r nce fal i i a ion model because of a lack of alter a ives, t e e sfc t n t not because the evi ence pulls him that way. But when he writes that “By 1989 offi d cial social st ide l gy, along with its clear artic a ion of the nature of injus ice, had i oo ul t t become a threat to the sys em it was meant to legit ate” he is talk ng about a cri is of t im i s tomslee.net/2013/02/503.html 9/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

iden ity. The cri is served to “focus dif use griev nces”, unit ng “a host of dis arate con t s f a i p cerns into ‘moral anger“‘. This is the crys al iza ion of iden i ies into the two polar t l t tt choices: “Which side are you on?” The “iden ity cas ade” model also makes a closer con ec ion between dynam cs and the t c n t i efforts of pro est rs. I’ll return to this later, but one of the things that pro est rs do in t e t e upris ngs is lay claim to the sym ols of national iden ity. Whether it’s Gandhi’s Salt March i b t or GDR pro est rs choos ng the 40th anniver ary of the found ng of the coun ry, strug les t e i s i t g over the mean ng of iden ity become cen ral at times of cri is. If infor a ion rev a ion i t t s m t el t was all that were needed, there would be no role for the dis lays of “wor hi ess, unity, p t n num ers and com it ent” that char c er ze polit al protest. An identity-driven b m m a t i ic approach makes this link clear, within a ratio al choice framework. n (Another nice thing is that within the iden ity cas ade model there is a nat ral cat o iza t c u eg r tion of the kind of events that can pre ip ate a cas ade. A shock to the norms asso i ted c it c ca with oppo i ion, a change in socio-economic con i ions that places more peo le into the st dt p “out ider” cat ory, or a change in state pol cy (per troika) all emerge as trig ers for cas s eg i es g cades. See the paper for more.)

Free Spaces and Screening
With that diver ion over, let’s return to the topic of free spaces. How do we get from the s lan uage of Pol etta and Jasper to the world of ratio al choice? There is a nat ral cor e g l n u r spon ence in the con ept of screen ng: a mech ism that imposes dif er n ial costs for two d c i an f e t dif er nt groups, so that (in a “sep at ng” equi ib ium) one group finds it worth hile to f e ar i l r w pay the cost, while the other does not. Here, the identity-driven costs of being a mem er b of “tol r ted” insti u ion screens out those with the sta us quo (G) iden ity. e a t t t t Just as Akerlof and Kran on sim li ied iden ity so that it could be squeezed into a ratio al t p f t n choice pic ure, so we have to sim lify the idea of an insti u ion. Hence orth, then, an insti t p t t f tu ion I is char c er zed by three things: t a t i Sta us (x): This is the nat ral mem er hip of the insti u ion. We can say that the t u b s t t iden ity of the insti u ion is the opti al iden ity of an indi id al with sta us x t t t m t v u t tomslee.net/2013/02/503.html 10/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

Breadth (δ): Indi id ls with sta us in [x — δ, x + δ] are mem ers of I. The “niche v ua t b soci ty” insti u ions of the GDR have a very nar ow breadth, while events such as e t t r national cel ra ions include all of society. eb t Mem er hip dis rim a ion (m): Some insti u ions do not dis rim ate between the b s c in t t t c in two iden i ies, but some do. A dis rim at ng insti u ion demands a cost of mem er tt c in i t t b ship for indi id ls whose iden ity dif ers from the iden ity of the institution. v ua t f t With this idea, you can build a model in which there is a range of insti u ions that even a t t strong state will not mon or, because the cost of mon or ng is greater than the ben it in it it i ef terms of dis ent that is qui ted. These insti u ions pro ide the free space for dis ent to s e t t v s per ist even under con i ions of strong government. s dt Here are some screen ng institutions i

Screen ng institutions. i

The screen ng insti u ions are those inside the lozenge shapes. Along the x axis is the sta i t t tus, so all these insti u ions are “tol r ted” in that they are entirely within the “out ider” t t e a s low-status zone. The broader the reach of the insti u ion (that’s the “δ” in the graph) the t t less scope there is for these insti u ions. Finally, and it’s beyond what I can explain in this t t part, there is a limit to the size of the “pub ic sphere” that also lim ts the avail ble l i a tomslee.net/2013/02/503.html 11/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

institutions. So what this shows is that the eco omic con ept of screen ng brings to the identity-driven n c i ratio al choice model the idea of free spaces, well estab ished within the soci og al lit r n l ol ic e a ure. To go back to the begin ing of this essay, the exis ence of such spaces is some hing t n t t that the net ork mod ls, with their focus on costs of com u i a ion, don’t seem well w e m nc t equipped to describe. So now we have a sin le the ry that cov rs both upris ngs and preg o e i revolutionary dis ent, instead of two (one micro, one macro). We can now see that the s “free spaces” of online dis ent are sim ar to, and exist for the same rea ons as, other free s il s spaces that have existed in the past. Even in Egypt, the role of the Ultras foot all fans b can fit within this model, the foot all sta ium ter aces pro id ng a “tol r ted” insti u ion b d r v i e a t t within which dis ent could be expressed. The model also argues that the key facet of online s spaces is not their tech o og al nature, but the fact that they were adopted by, and asso i n l ic cated with, the broadly anti-establishment demo raphic of urban youth. Nav at ng the dis g ig i cus ion spaces of the online world is easy if you have friends who are tak ng part: not so s i easy if you are a gov rn ent offi ial try ng to pose as a dis n ran hised youth. The tech e m c i e f c nol gy of social media is epiphe om al. In broad strokes, this is an argu ent I made o n en m some time ago here: it’s only taken two years for me to work it out properly.

Insti u ions and Challenges t t
The final case to look at is when a social move ent chal enges a weak gov rn ent. The m l e m goal is to put the gov rn ent in a “dictator’s dilemma”. The idea that clamp ng down on e m i dis ent has the pos i il ty of draw ng atten ion to it, and per aps fan ing the flames, is an s sb i i t h n old one. Here is a recent statement:

[S]ometimes repres ion inspires more mobi iza ion; and some imes it effec ively s l t t t quashes move ents or pushes them under round. Some imes repres ive forces m g t s are suc ess ul in char c er z ng pro est rs as legit ate tar ets of repres ion, and c f a t i i t e im g s other times they delig timize the State and increase the legit acy of the social i im move ents. m – Cristina Flesher Fom aya and Les ey Wood in l

tomslee.net/2013/02/503.html

12/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

Or, going back a lit le further: t

Cen or hip makes every banned text, bad or good, into an extra r i ary text. s s o dn – Karl Marx

The con rast ng for unes of the GDR protests and the Tien n en Square protests in 1989 t i t a m are the best known exam le of this dual possibility. p When they believe the time is right, social move ents may actively seek to pro oke a cri m v sis (con rary to the “safety in num ers” cost min iza ion that the infor a ion cas ade t b im t m t c the ists tend to favour). Famously, here is Gandhi: or

The func ion of a civil resis ance is to pro oke response and we will con inue to t t v t pro oke until they respond or change the law. – M. K. Gandhi v

We can bring this idea of provo a ion into a micro model if we bring in a uni ary social c t t move ent and invoke an inter e en ency between iden ity polar za ion and gov rn ent m d p d t i t e m coer ion. Again, the math at cs is in the paper. c em i The ques ion we ask is “if you were an orga ized oppo i ion, what insti u ion would you t n st t t tar et, so that a clam own would cause polar za ion?” The idea is that clam own on a g pd i t pd main tream insti u ion would be more likely to polar ze soci ty, by dis urb ng even s t t i e t i the government’s own sup ort rs, than clamp ng down on an “out ider” insti u ion. Again, p e i s t t the oppo i ion has to make a pay ent to appro ri te a main tream insti u ion, because of st m p a s t t mem er hip selec iv ty. They have to pass with the iden ity of a sta us quo sup orter. b s t i t t p They need to appeal to main tream sen i il ies and to estab ish legit acy. Under s s b it l im the right cir um tances, an oppo i ion will pay the cost of provo a ion, because they antic c s st c t i ate that a gov rn ent response will weaken, not strengthen, the government’s level of p e m con rol. Here is a fig re show ng a set of insti u ions that can be used by an oppo i ion to t u i t t st pro oke a crisis. v

tomslee.net/2013/02/503.html

13/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

Insti u ions that may pro oke a crisis t t v

The insti u ions that may pro oke a cri is are those within the cen ral closed shape, t t v s t bounded clock ise by light blue (on top), green, red, and pur le. Some of these insti u w p t tions are “tol r ted” insti u ions to the left of the x* line, with oppo i ional iden i ies; for e a t t st tt these insti u ions there is no mem er hip cost to be paid by the oppo i ion. Oth rs are t t b s st e “pre cribed” insti u ions that have a main tream iden ity. The oppo i ion must pay s t t s t st the price of appro ri t ng these insti u ions: par ic at ng in them in such a way as to pro p ai t t t ip i voke the government. An exam le of this behav our comes again from the GDR upris ngs of 1989, as described by p i i Steven Pfaff. The oppo i ion chose the cel ra ions of the GDR’s for i th anniver ary – a st eb t te s main tream insti u ion – in which to pro oke a response. The gov rn ent did respond, s t t v e m but “its bru al attacks on peace ul pro est rs dur ng the for i th anniver ary … prob ly t f t e i te s ab acti ated what might have oth r ise remained despair ng, but inert, citizens.” v e w i The oppo i ion made explicit attempts to por ray them elves as main tream Ger ans, st t s s m adopt ng the sim le slo an of “Wir sind das volk” (“We are the people”). i p g

“Wir sind das volk” [was] a thin claim, but an uncom li ated “us ver us them” mes p c s ­ tomslee.net/2013/02/503.html 14/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

sage, a claim to polit al iden ity that could bridge lines of class, edu a ion, neigh ic t c t ­ bor ood, and so on. – Steven Pfaff h

In pre i us times, other upris ngs have explic tly cho en main tream or some imes tol r vo i i s s t e ated insti u ions as a means of provo a ion. Gandhi’s use of the Salt March, the Chi ese t t c t n stu ents’ use of the death of Hu Yaobang and Tien n en Square, the Egypt an pro est rs d a m i t e appro ri ion of National Police Day and Tahrir Square all fol ow this pattern. p at l There are claims that dig al tech olo ies at times of cri is can act in this man er. Ethan it n g s n Zuck r an has pop ar zed the idea as a “Cute Cat” the ry: that main tream insti u ions e m ul i o s t t pro ide a venue for dis ent that can ot be shut down with ut polar z ng soci ty. The the v s n o i i e ory here pro ides at best lim ted and con i ional sup ort for the idea. Dig al tech olo ies v i dt p it n g were not used as a mech ism of provo a ion, but played a sup ort ng role. The “cute cat” an c t p i idea has cre ence only if the gov rn ent is not able to silence dis ent in a more selec ive d e m s t man er than shut ing down the entire inter et or phone ser ice within the country. n t n v My favourite exam le is the French “Ban uet Cam aign” of 1848. Repub i ans were cam p q p lc paign ng for uni er al male suf rage against an intran i ent gov rn ent that had banned i v s f sg e m polit al meet ngs. Faced with the prob em of orga iz ng an oppo i ion in such an envi on ic i l n i st r ment, they orga ized ban uets. On the 18th of July in Mâcon, Bur undy, five hun red n q g d tables were set up for three thou and guests with stands for three thou and more, osten i s s sbly as a cel ra ion of local lit r ry star Alphonse de Lamar ine. Lamar ine was not just a eb t e a t t lit r ry star though, he was also a well-known repub i an, and the author ies knew that e a lc it the ban uet was a cover for polit al agi a ion. But the author ies judged that inter er ng q ic t t it f i with the ban uets would inflame the sit ion rather than suc eed in sup ress ng the q uat c p i protest, and so let the ban uet pro eed. With the suc ess of the Mâcon ban uet, the “Cam q c c q pagne des ban uets” was launched, and ban uets were held around the coun ry. This is the q q t high wire act that gov rn ents and oppo i ion walk at times of cri is – when to push e m st s ahead, when to hold back, and what tac ics may be effec ive – and is the kind of dance that t t social move ent stud es have helped to elu i ate. The cam aign con in ed until Feb u ry m i cd p t u r a of the next year, when the gov rn ent decided it had no choice but to esca ate. The ban e m l quets were out awed, a hastily orga ized protest brought peo le into the streets of l n p Paris on Feb u ry 22, a con ronta ion between the Munic al Guard and the r a f t ip tomslee.net/2013/02/503.html 15/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

marchers spilled over into riots, every hing got out of hand, and the King fled t Paris. Within a few weeks gov rn ents were top led in Milan, Venice, Naples, e m p Palermo, Vienna, Prague, Budapest, Krakow, and Berlin. I like to think that the graph above cap ures a lit le of that drama. t t

Con lu ions c s
What I’ve tried to do here is fol ow the Akerlof & Kran on exam le of tak ng the rich soci l t p i o og al con epts of iden ity seri usly, and used it to con truct a ratio al choice model of l ic c t o s n upris ngs that com le ents, rather than com etes with, soci og al mod ls. I’ve added i p m p ol ic e some dynam cs to the approach, and brought in a mod l ing of insti u ions to build on the i el t t notion of col ec ive iden ity as a moti at ng force for protest. l t t v i The results are that the the ry recov rs the key facet of other ratio al choice mod ls of o e n e upris ngs, which is cas ades, but with a dif er nt inter re a ion. Here it is “iden ity cas i c f e p t t t cades” rather than “infor a ion cas ades” that drive the sud en change. Beyond cas ades, m t c d c the the ry shows how screen ng pro ides a mech ism for the exis ence of “free space” o i v an t insti u ions in which dis ent can be sus ained, even in author ar an regimes. Finally, it t t s t it i shows how an orga ized oppo i ion may appro ri te main tream insti u ions with the n st p a s t t explicit intent of pro ok ng a cri is, putting the gov rn ent in a “dictator’s dilemma” in v i s e m which nei her respond ng nor fail ng to respond is a good option. t i i (Writ en in Org ver ion 7.9.3f with Emacs ver ion 23) t s s

Share this:

Share

This entry was posted in Uncategorized on February 25, 2013 [http://tomslee.net/2013/02/503.html] .

8 thoughts on “Notes on Identity, Institutions, and Uprisings”

Metatone tomslee.net/2013/02/503.html 16/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

February 27, 2013 at 2:14 pm

I’m tired and jet agged, so I’ve only skimmed this — but in case I never make it back for l rea oned analy is, I wanted to point up the NECSI research about food prices and fre s s quency of riots. I’d sug est that there is a macro-dimension at play and food prices were it. g They act as a kind of pres ure gra i nt that impels iden ity cas ades perhaps…? s de t c

Tom Slee

Post author

February 27, 2013 at 3:56 pm

Thanks Mr/Ms Tone. The clos st I get to such real-world phe om na is that there is a e n e macro-parameter in the model (not here, but in the linked paper) which can stand in for socio-economic fac ors and which can trig er cas ades. I hadn’t seen the research you men t g c tion (I guess you mean this) but this paper (PDF) by Andrea Teti and Gen aro Ger a io n v s seems to make a sim ar suggestion. il

Pingback: OccupyData and TTW13 | MrLiterati
LFC
March 6, 2013 at 11:24 pm

I like some of the ideas here: that peo le (may) have a fairly sud en “switch in iden ity p d t [which] hap ens when peo le are pulled along by those around them” and that out iders p p s have to “appro ri te main tream insti u ions” to pro oke a govt response which will lead p a s t t v to polar za ion — the lat er point, it seems to me, cd be fleshed out a bit more in the i t t Egypt an case, rather than just a pass ng ref to Tahrir Square and Natl Police Day. Maybe i i you do that in the longer version. The way the argu ent is pack ged (w ref r nce to mod ls and the graphs and so on) I can’t m a e e e tomslee.net/2013/02/503.html 17/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

say appeals to me that much. But I’m undoubt dly not the modal reader you are try ng to e i address. Any ay it’s nice to have this shorter ver ion with ut the math b.c I had looked at w s o the full paper a while ago and decided that I didn’t want to try to strug le through it and g extract the main points. Here the main points are some hat more accessible. w

LFC
March 6, 2013 at 11:28 pm

P.s. There’s an arti le on the 1848 — Arab Spring com ar on in a recent issue of ‘Per pec c p is s tives on Pol ics’ — maybe you’ve seen it. I haven’t read it and am too tired now to find the it citation.

Tom Slee

Post author

March 9, 2013 at 8:25 am

Thanks LFC. The arti le you mean must be this one by Kurt Wey and. I haven’t seen it, and c l unfor u ately it seems to be behind the wall, so I can’t get at it. t n I have had the same reac ion about the mode of pre en a ion from oth rs. I admit that it’s t s t t e not very acces i le. If there’s a point to it, it is that the network-based mod ls that focus sb e on infor a ion rev a ion have been suc ess ul because there is a for al the ry behind m t el t c f m o them which can yet be pop ar zed, and which gives those pop ar za ions cred il ty. In ul i ul i t ib i argu ng against those mod ls, there is noth ng for al avail ble, so I had to do it myself. i e i m a Pop ar z ng: maybe that’s some hing different. ul i i t

LFC
March 9, 2013 at 9:38 am

tomslee.net/2013/02/503.html

18/19

3/18/13

Notes on Identity, Institutions, and Uprisings | Whimsley

Yup, I get the reason(s) for formalization. Btw I have access to the Wey and piece (b.c I’m a mem er of AmPolS i ssn) and I will l b cA email you the PDF, though per aps not till tomor ow (busy w/ other stuff today). h r

Tom Slee

Post author

March 10, 2013 at 12:28 pm

I got a copy from some ne else. Thanks though. o

tomslee.net/2013/02/503.html

19/19