David Skarbek (Brown University) describes his ethnographic work on prison governance as a historical analogy to the emergence of states. Join us on this episode of the Governance Podcast led by John Meadowcroft (King’s College London) for a vibrant discussion on how governance emerges (or doesn’t) in different social landscapes, from prisons and gulags to clans and nation-states.
Subscribe on iTunes and Spotify
Subscribe to the Governance Podcast on iTunes and Spotify today and get all our latest episodes directly in your pocket.
Follow Us
For more information about our upcoming podcasts and events, follow us on facebook, twitter or instagram (@csgskcl).
The Guest
David Skarbek is Associate Professor of Political Science at Brown University. His research examines how extralegal governance institutions form, operate, and evolve. He has published extensively on the informal institutions that govern life in prisons in California and around the globe.
His work has appeared in leading journals in political science, economics, and criminology, including in the American Political Science Review, Journal of Economic Behavior & Organization, Journal of Law, Economics & Organization, and Journal of Criminal Justice.
His book, The Social Order of the Underworld: How Prison Gangs Govern the American Penal System (Oxford University Press), received the American Political Science Association’s 2016 William H. Riker Award for the best book in political economy in the previous three years. It was also awarded the 2014 Best Publication Award from the International Association for the Study of Organized Crime and was shortlisted for the British Sociological Association’s 2014 Ethnography Award.
His work has been featured widely in national and international media outlets, such as the Atlantic, BBC, Business Insider, the Economist, Forbes, the Independent, and the Times.
Skip Ahead
00:38: David, you’re well known for writing a book on prison gangs in California and America called The Social Order of the Underworld. Just to begin, tell us a little bit about that book.
2:01: You mentioned that prison gangs are often organized on racial lines. Why is that the case?
4:10: So race is a convenient way of organizing a large group of people. Is that what you’re arguing?
4:34: Does that mean this has changed over time? So as a prison population got bigger in America, gangs organized upon racial lines have become more important?
7:44: You mentioned that the convict code, if you like, was informal. Would you see gangs as providing more formal governance?
9:15: Would it be fair or is it a stretch to suggest that this is like a prison constitution?
10:53: One thing when you read the book that’s quite striking is there are lots of vivid descriptions of violence that occurs in prison. How do you reconcile that evidence with what you describe as some sort of order?
13:55: I imagine that the question that comes to many people’s minds when it comes to prison gangs, is what would happen if they went to prison? Would they have to join a prison gang, and if the didn’t, what would be the consequences?
15:26: So it’d be fair to say you cannot be a solitary individual, you cannot be a holdout, so to speak.
16:15: Could we then imagine that prisons are close to what we might think of the state of nature in social science?
17:05: This brings us to your latest work in this area, which I think is going to be called the Puzzle of Prison Order. How does it extend your previous work?
20:03: Maybe you can say a little more about English prisons. One senses that they don’t have that kind of gang organization that we observe in California. Why should that be the case?
23:39: One challenge this book takes on is trying to unpack all these different factors, all these different possibilities. So I guess one common sense question would be, looking at California, America, the UK, there is a presence of gangs on the streets. One might assume intuitively that the gangs on the streets are more well organized in California compared to England and Wales. Is that the case, and how does that play into what happens in prisons?
26:08: Another dimension which I think would be of interest is the difference between men’s and women’s prisons. What have you been observing?
29:44: Let me ask a more mischievous question: You’ve looked at prisons around the world and spent many years reading research on this. Is there a country or prison system that is completely opposite to what your theory would predict? For instance, where there is a small prison population but there are lots of gangs?
31:42: So it’s a story ultimately about governance, and much less about the size of prisons.
32:10: One thing that’s striking is, prisons have people with very few resources, they may be predisposed to violence… should this lead us to be hopeful about people’s capacity for self-governance?
35:06: So it’s undoubtedly impressive that prisoners are able to self-organize or self-govern in this way. Thinking of the comparative political economy of this, though, wouldn’t it be better if there was formal governance? Is that safer and less violent?
37:00: Essentially you’re engaging in qualitative research. Maybe the first question here is about the challenges of obtaining that kind of data from prisons around the world and how you go about overcoming that challenge.
40:27: What’s your sense of the challenges of comparing different ethnographic studies?
44:26: So you were trained as an economist originally. How do economists view this sort of methodological approach, and would they be concerned about your ability to give causal answers?
46:04: As a political scientist, you see political science going in the direction of causal identification and experimental results. Should we be concerned about that and is it limiting the types of questions we can ask?
48:18: I assume you’re not going to be working on prisons forever. What other ideas do you have going forward?
Full Transcript
John Meadowcroft
Hello, welcome to the Governance Podcast at the Centre for Study of Governance and Society at King’s College London. My name is John Meadowcroft. I’m a reader in public policy here at King’s. I’m delighted to be joined by David Skarbek. David was a member of the Department of Political Economy here at Kings. He’s presently an associate professor at Brown University in the Political Science department. He spent a year at Berkeley visiting the Centre for Study of Law and Society. It’s great to have you here at King’s, David. You’re well known for writing a book on prison gangs in California and America called The Social Order of the Underworld. Maybe just to begin, why don’t you tell us a little bit about Social Order of the Underworld.
David Skarbek
Thanks, John. I’m delighted to be back. Prisons in California today are ruled, are governed by very hierarchically organised groups. They’re ethnically and racially segregated organisations. These prison gangs have a dominant influence on the everyday life of prisoners in California. And that means that when people go out to the prison yard, where they can sit, who they can associate with, which basketball courts they can use, all of that is decided through a sometimes violent but not not always violent set of negotiations between these prison gangs. So gangs govern social interactions amongst prisoners, and they also play a very important role in governing economic interactions between prisoners. So there’s a significant underground economy amongst prisoners and gangs play a sort of governmental role in regulating how those transactions take place, and often adjudicating problems that arise in the underground economy.
John Meadowcroft
Thanks. You mentioned though, that prison gangs often organised on racial lines, why is that the case? I mean, are these people racists?
David Skarbek
So it’s definitely clear that many prison gang members are racist. And they often use a lot of racist icons and sort of motifs. But what’s interesting about gangs is that they also interact across racial lines. When you talk to prisoners who abide by racially segregated informal rules in prison, they say, I’m not a racist, but this environment is dominated by racial rules and I have to comply with them. And so it seems the case that even though these groups are broken down by racial lines, a lot of the reasons why that’s the case has to do with non sort of hateful, racist ideologies. And so my basic argument is that gangs emerge in large populations of prisoners in large populations of strangers. And these gangs operate in sort of a mutual responsibility system. So if, if a member of a particular group violates some social or economic rule, other prisoners can go to his shot caller, his gang leader, and ask that they hold their own member accountable. So these gangs are organised in such a way to exert tremendous in-group pressure to facilitate social and economic interactions across groups. Now, how does this come to the race issue? Well, these gangs, they dominate in large prison populations where you may not, cannot not know who most other prisoners are. And so by affiliating along racial and ethnic lines, it becomes far easier for strangers who’ve been subject to some social insult or some economic opportunism. It’s far easier for a stranger to know who to go, to turn to, to complain to hold some particular prisoner responsible. So that’s why in the large prison populations today race is so important. And prior to the 1960s, the small prison populations, race was far less important in defining social relationships.
John Meadowcroft
So race seems like a convenient way of organising a large number of people. Is that what it is?
David Skarbek
That’s right. In large populations, if you want to hold strangers accountable, you have to know who to go to to hold them accountable. And race is a very low cost way to sort of look at someone and know pretty well who you need to go to speak with.
John Meadowcroft
Yeah, okay. So does that mean that this has changed over time? So as the prison population has gotten bigger in America, then gangs organised upon racial lines have become more important?
David Skarbek
Yeah. And that’s one of the things that I found fascinating and doing the research for this book is that gangs have this dominant influence in prison life today. But prisons existed in California for more than 100 years and there are no groups like these. So if they were so important today, how could they not even exist before? And so I wanted to understand why is it that there was this radical transformation in the social order in California prisons. And as you know, I think it’s very much tied to the size of the prison population. These gangs are very effective at providing governance, but it takes a lot of resources. They have to invest resources to create the organisation to create rules to govern interactions, but between members of gangs, they have extensive procedures for collecting paperwork, and they have written rules that they distribute to prisoners. So it takes a lot of time and energy and resources to create gangs. And it’s only worth doing that if you can’t rely on other lower cost methods of governing interactions with other prisoners.
And so my argument essentially is that when prison populations were small, prisoners could rely on very decentralised informal mechanisms prior to the 1960s social norms governed interactions between prisoners. There was a sort of understanding that good prisoners, good convicts would not inform on other prisoners, they want to steal from other people that want to lie, cheat, they pay back their debts. And to the extent that prisoners abided by this code, something that they often called the prisoners code or the convict code, the more we adhere to it, the higher status that person would be. And being in good standing meant that a person would have the support of his peers, you’d be less likely to be victimised. In order for that informal system of norms to work well. You have to know people’s social standing. You have to be able to know enough about someone and their reputation that you know where in the pecking order they stand.
Now, if a prisoner violated those norms frequently, his standing would fall and people wouldn’t want to associate with them and they might, you know, be more likely to victimise that person. So in small prison populations, it’s your prisoners who are able to know each other’s reputation. And that reputation then becomes something that you can use to encourage compliance with acceptable social norms. So if prisoners violated these codes, they’d be subject to gossip, and ostracism and shaming from other prisoners. And in some sense, those are sort of ideal mechanisms of social control, because they’re very cheap to produce. They don’t require the investment to create an organisation. And they don’t require a lot of coordination of collective action. So in the period when those informal norms work very well, there’s no reason for them to incur the cost to create gangs. And that’s my argument for why gangs are dominant today in large populations of strangers. But we don’t observe those in smaller prison populations.
John Meadowcroft
Yeah. You mentioned that the convict code, if you like, was informal. So would you see gangs as providing more formal governance?
David Skarbek
Yeah, so today, prisoners can’t just sort of rely in an ad hoc fashion on norms to emerge. The gangs themselves write written rules. In many prisons, they distribute written rules to new prisoners that tell them what’s acceptable behaviour there. Some of these are very, sort of usual or ordinary, you know, for example, don’t throw trash on the prison tear unless it’s being swept. So there’s a concern about reducing sort of negative effects on other prisoners. Some of them are more oriented to gang life in some prisons, prisoners have to work out for a minimum of one hour a day so that they can contribute to the sort of physical intimidation necessary for a gang to maintain its place in the prison hierarchy. And then sometimes these written rules include sort of direct rules related to the illicit economy. So sometimes gangs will in these rules tell prisoners how much of an illegal drug deal they have to pay as a so-called tax to the gang leaders. So the gang spends a lot more time explaining written rules, distributing them to new prisoners, and many gangs require new prisoners to sort of pass a test, often to be able to recite what the rules are before they can interact with them out in the prison yard.
John Meadowcroft
So would it be too much of a stretch to think of this almost like a prison constitution?
David Skarbek
I think it’s not much of a stretch. So gangs write rules that tell their members how to interact with members of other gangs. But the gangs also write more foundational or constitutional style rules that define on what terms prisoners will affiliate within a gang. And so I argue that gangs face a very similar problem that nation states face, which is one that in the federalist papers in America Madison spoke very eloquently about. He said that we need to find ways to empower a state that can enforce our rights without allowing it to use that power to violate our rights, and the gangs face the same problem, they want to create a gang that can enforce the rights of prisoners, but control the power so that the gang doesn’t prey on its own prisoners again. And so when you observe the internal written constitutions of these gangs, they often include, you know, fairly elaborate relative to sort of our priors, fairly elaborate systems of checks and balances, elections of leaders, impeachment procedures, there are built into these gang organisations methods of holding members who have been given power accountable for how they use that power.
John Meadowcroft
You may just have been the first person to compare Madison to a prison gang shot caller, which is a good thing, I think. So, I mean, that’s, that makes a lot of sense to me for sure. One thing when you read the book, often it’s quite striking. There’s lots of quite vivid descriptions of violence that occurs in prison. How do we reconcile that evidence with what you describe that prison gangs keep some sort of order?
David Skarbek
Yeah, that’s a great question. I think that there’s a few ways that I think about it. The first is that gangs have a positive effect on other prisoners in a few important ways. They make prisoners in these large populations a bit safer, and they facilitate access to the illicit economy. And if it weren’t for gangs, I don’t think either of those things would be nearly as true. Now, of course, from a prison warden’s perspective, more flourishing, illicit trade is a bad thing. But there’s also a sort of broader criteria that we, I think, should very reasonably think about when assessing the role of gangs in the society.
So prison gangs have few mechanisms of accountability outside of itself. So if a member of a prison gang isn’t happy with the constitution that they have, there’s very few options for them to turn to other institutions. Just sort of checked that power, gangs have very little interest in equity across prisoners, they sort of intentionally seek out prisoners who they think are undesirable, so such as sex offenders, and sometimes just automatically assault those individuals. And in a sense, you know, we’re giving power to people in prisons or gangs to give power to people in prison, who are most willing to use that power to try to dominate others in their community. And so when we sort of think about a broad array of characteristics for what sorts of institutions are working well or not gangs, sort of check a few boxes, but they sort of leave out a few other ones.
There is sort of an interesting pattern that we observe, both in California but also in other Latin American cases, which is that prison gangs don’t use violence. Prison gangs’ goals aren’t to use violence against people. They use violence as a means to accomplish the end of gaining control of the community. So you typically see a lot of prison gang violence when there’s no there’s no stability or equilibrium in the communities that are there. So when they’re fighting for battle, when they’re fighting for control, we see a lot of gang violence. But when that sort of battle has come to an end, when it’s stabilised, violence drops dramatically. And so one of the alternative explanations for the role of gangs in California is that gangs formed to promote violence. That’s something that prison officials often say they have an agenda of violence. The problem is that that’s inconsistent with the evidence over the last 70 years, prisons have become significantly less violent since the 1970s. At the same time prison gangs have emerged and become incredibly prominent. So if your argument is that gangs exist, because they want to promote violence, by the data itself, it looks like they failed to do that. And that makes me think that that’s not the best explanation for explaining the historical variation in gangs.
John Meadocrowft
I’m not sure one question that comes to many people’s minds when they think about prison gangs is what would happen if they went to prison? So would they have to join a prison gang? If they didn’t, what would be the consequences.
David Skarbek
So in California, everyone has to affiliate with some group that’s going to hold their members accountable for their interactions with others. Now, the gangs are the most dominant groups there, and most prisoners will go and they will affiliate with the racial and gang group at the prison in which they live. But when they leave prison, they don’t have to continue working for the gang. So sort of adjust in prison sort of affiliation. And absolutely all prisoners have to affiliate with some group. In addition to the gangs, there are sometimes other periphery groups that will provide this mutual accountability or mutual responsibility that the gangs provide. So for example, I sort of discovered that religious groups in prison often play the same role. So as long as you’re a member of the evangelicals group, that group is responsible for your actions. And as long as they hold you responsible for it, then you can sort of spend less and be less involved with the sort of gang elements. So in that way the religious groups are providing similar governance functions as the gangs, although presumably less of the criminal and violence functions that the prison gangs are often involved with.
John Meadowcroft
So be fair to say you cannot be a solitary individual and you cannot be a holdout?
David Skarbek
Yeah. I think there are very few sort of lone wolves because when these individuals are there, if they’re not affiliated with the group, first of all, gangs will pressure other gangs to be held to bring the new prisoner in. So prisoners will say, look, you’ve got a new guy in the yard. You need to teach him the rules. You need to show him he can hang with you guys, but he can’t hang with us. And so there’s a pressure to incorporate people into this sort of social order. And secondly, even if that sort of failed to happen, a lone individual would be sort of a very prime target for prisoners who might wish to steal from him or victimise them in some ways.
John Meadowcroft
I mean, this might be a stretch too far but could we then imagine that the prisons are close to what we might think of the state of nature, in social sciences, there’s something that sort of …
David Skarbek
I think that the counterfactual of a sort of Hobbesian anarchy is the context in which these prisoners operate, and that there are vast scopes of interactions that are outside of the reach of the state. And in that sense, they are in a state of nature. And so really, I think of this story, this book as a sort of historical analogy to the emergence of states. So I do think that one effective way to think of gangs is as a type of nation state, and their interactions within the prison yard are very much akin to the international anarchy of the global system of nations.
John Meadowcroft
Okay, this brings us really to your latest work in this area, which I believe you’re writing a book at the moment or nearing completion of a new book, which I think is going to be called The Puzzle of Prison Order. And maybe just as an introduction to this part of the description, maybe you could describe what that book involves and how to extend the previous work.
David Skarbek
Yeah. So my first book was about the history of California over time. And it seemed to answer sort of the role of prisoner demographics and prisoner size for the emergence of gangs. When you look around the world, there are a lot of prisons that don’t have gangs. And I wondered, to what extent can this theory help explain those things? There are also lots of places that have gangs but actually relatively small prison populations. So how do we sort of square understand the relationship between the incredible diversity of prison social borders around the world, so it’s in some ways a more ambitious project, which is trying to say, if you look around the world life in prison looks incredibly different depending on where you are. Is there a way to sort of build on this theory to help explain more of that variation?
John Meadowcroft
I guess the question then is, what’s the answer? Why does prison sort of vary as much as we see?
David Skarbek
So in my new book, I make two arguments. The first is that when the quality of governance provided by officials is very low, then there’s a strong incentive for prisoners to invest time, energy and resources to provide governance to fill in the gap of governance left by ineffective officials. And so in the book, I look at cases of Brazil and Bolivia, where there are incredibly few prisoners, excuse me, incredibly few staff members in prisons. They provide very few resources sometimes, practically no food or clean water. Prisons are very dangerous, often in disrepair, and so prisoners are sort of thrown into prison, but provided none of the governance that officials in California, for example, would provide. And so I argue that they respond in a variety of ways to account for that sort of deprivation of governance. So when officials don’t govern prisoners and then I look at other extreme examples where officials govern incredibly well. And so for that I examine Nordic prison systems, where there are a large number of well trained prison staff, where there are extensive resources. And as a result, I argue we see very little of the prisoners produce governance institutions that in the extreme we see in Latin America, or in a sort of middle range case that we see in California. So that’s a sort of a major part of it. And then we could talk if you want about extending the sort of the importance of prison size and looking at places in England for example, and women’s prisons.
John Meadowcroft
Given that we’re based in London here at King’s, maybe you could say a little bit about English prisons. And one sentence is that they don’t have this sort of gang organisation that we see in California, probably also other countries. So why should that be the case?
David Skarbek
I mean, so my argument is that that should be puzzling. So, there are obviously significant differences between the United States, California and England. But the key features that describe prisons, regardless of where we look are pretty much the same. There are places where we take people who have been charged with or convicted of a crime and we forced them to be confined there. They’re typically forced to interact with other people in the same situation. They tend to come from disadvantaged socio economic backgrounds. They don’t have an exit option, you know, you are forced to interact with people. And so when we think of social scientists about what are some of the most important characteristics that define and describe social interactions, those are some of the most important ones and either by definition or practice, that is a good description of prisons, regardless of where we look.
And so that might make you think, well, we should see the same outcomes there. Likewise, in England, there are many of the same legal traditions and institutions as in California in the United States. Many of the prison practices and criminal justice practices developed in the United States have been adopted through a process of policy transfer in England and in English prisons. And as you know, rightfully, there’s no similar or equivalent organisation in English prisons as the prison gangs in California. And that’s the sort of question with all these similarities, why don’t gangs have such an important influence in English prisons? And my argument essentially comes down to three main factors, three key things that make English prisons substantially different from California prisons. The first is the size of the prison population and the size of prisons. In particular, the average size prison in California is more than 3500 prisoners. And in England and Wales, the average size facility is about 600 to 700 prisoners.
And so I argue that in those small prisons, like in California prior to the 1960s, in small English prisons, it’s a lot easier and more effective to rely on decentralised mechanisms like gossip, shaming and ostracism. The second key difference is that there’s an intentional process of placing prisoners in prisons that are sited close to their home communities in this country. And the intuition is that if you’re close to where you come from your family and other people can visit you, you can maintain healthy positive relationships. Well, one consequence of that is that when you go to prison, you’ll show up there and you’ll meet and reconnect with people that you know when you are out in the community. So when you get to prison, there are people there who you already know. And they will already know you. And you’ll both know that when you both return to your community outside of prison, that you’ll know what happened when you were in prison. And so both of these factors make it very important that you maintain your reputation. And so it increases the ability to wield these sort of reputation mechanisms to facilitate social control. And with these things being so effective, I argue that there’s no need to invest in the sort of high structure of what the gangs providing California…
John Meadowcroft
What seems as such a great challenge that this book takes on is trying to unpack all the different factors or all these different possibilities. So I guess one, if you like, common sense, sort of question would be another difference between California, America and the UK. I guess it’s the presence of gangs on the streets. When my assumption is true to me that the gangs on the streets are more well organised in California. Have you compared to England and Wales, if you like? Is that the case? And how does that play into what happens inside prisons?
David Skarbek
Yeah. And that’s a common explanation, especially in the criminology literature that prison gangs exist in prisons because there are street gangs out in the community. And when their members become incarcerated, they import the organisation and their culture and organisational structure into the prison system. In the United States case, that doesn’t seem to be the case. Many of the states where prison gangs emerged earliest didn’t have a significant street gang presence. And if there’s no street gang presence, then there’s nobody to import the street gangs into the prisons and recreate them there.
In the California case, we know that there were street gangs for at least 50 years before we had prison gangs. And so it’s unclear why it would take so long for these prison gangs to emerge. And then we can drill down even closer into the historical record in California. And it turns out that many of the first prison gangs formed gangs because they were perceived to be more vulnerable because they weren’t sort of sophisticated street criminals like their other peers. They weren’t judged to be sort of sophisticated, they were seen to be sort of more rural and less sophisticated prisoners. So it doesn’t seem to line up very well. If you look at English prisons, there are people who are affiliated with street gangs on the outside who are incarcerated, of course, and, you know, according to, you know, the sort of best ethnographic research that I’m aware of on English street gangs, there are hundreds of street gangs in London and there are, you know, hundreds of street gang members who could if they wanted to go into English prisons and reproduce those organisations there, but we don’t see that happening. And I think that the reason we don’t see that happening is again, because it’s costly to do that. In fact, you know, some of the research being done shows that it’s actually very more likely that when you go to prison, you will assist from street gang activity rather than enhance your gang affiliation activity.
John Meadowcroft
Another dimension, which I think will interest people is the difference between men’s and women’s prisons. So I believe in the book you write a bit about women’s prisons, and how they differ. And maybe you could tell us a little bit about that.
David Skarbek
Yeah, well, I’m going to follow up on the street gang activity, you know, anywhere from you know, the best estimates suggest between a quarter and maybe 40% of street gang members are female. So there are a large number of female street gang members, and there is a significant number of all female street gangs. In California in particular, this is especially true, but when you look at the social order in women’s prisons in California, again, there are no prison gangs. And so it doesn’t seem like the presence of street gangs is enough to get prison gangs. In those facilities, the social order that exists in women’s prisons in California has actually been remarkably stable from the 1950s until the present so more than 60 years, the best studies of women’s prisons have a remarkable consistency and how they’ve organised themselves and I guess it’s probably characterised in a few ways is that they have a similar system to what men had before a sort of convict code system.
The individual reputation of a woman and her interactions in prison is what matters the most. To the extent that there’s social sanctions or punishments, that’s sort of chosen by each woman on her own, whether to conform to the norms or not whether to violate or punish violations of the norms. In between, you know, zero and 70% of the women studied based on the prison. Many women form fictive kin ships. They often call them play families or just family and you. That’s essentially women deciding to take on the roles of the sort of nuclear family. One will be the mom, the other will be the father. They’ll adopt kids. They’ll have sisters and uncles that create aunt and uncle relationships. And you know, these are structured in some way. There are organisations in some ways, but they’re very different from prison gangs. So, membership in these families isn’t permanent. You can get divorced and remarried, you can disavow children, when you leave prison and return to prison. So these things are not permanent, and they’re much more fluid and a large number of women don’t participate in fictive kinships Another interesting feature of women’s prisons is that they’re not as racially and ethnically segregated as California men’s prisons are. Now we don’t have any particular reason to explain that difference. Except I argue that women’s prisons don’t have racial segregation. In the same way, because they still have a relatively small prison population.
So remember that men solidify racial and ethnic distinctions when they have to rely on community responsibility systems among strangers. Because women’s prison populations are very small, they can still rely on reputation mechanisms. And so they don’t need to use race as a sort of indicator of some broader affiliation. So women’s prisons look in these sort of abstract terms a lot more like men’s prisons when they have very small prison populations. And in fact, the size of women’s prisons in California have never reached the height of men’s prisons when men were forced to turn to gangs.
John Meadowcroft
Well, let me now ask a sort of mischievous question, which I know you’ve looked at prisons all around the world, and spent many, many years reading lots of research. Is there a country in a prison system that is completely the opposite to what your theory would predict, like a small prison population, and we see lots of guns within the prison?
David Skarbek
So that’s a good question. I mean, I think that the best cases that specifically are challenged by that argument, there’s other things that are also moving other variables are changing. So, you know, I mentioned there are Brazilian prisons where they’re fairly small populations, but officials provide no governance. So to the extent that gangs are governing the scope of interactions and which presenters must govern in Brazil is significantly larger than it is in California. So because they have no access to official governance, you know, they really have to invest even if they can know each other’s reputations. There’s also, you know, sort of prisons that I’ve not studied that I think are very important that you simply haven’t had time to sort of work on and think about.
So for example, the Soviet gulags are a very interesting and very distinct set of prisons. They’re incredible, like the Latin American prisons, but instead of being neglected by officials, they’re dominant, you know, there’s a dominant control by officials to control prisoners. And they’re sent out to work, you know, brutally long hours and then returned. And so there’s not a flourishing of extra legal governance and gulags because the resources are taken away, but then they’re controlled and forced out to do other work. So there’s no opportunity, there’s no freedom for them to respond to that scarcity by governing themselves. And so I think that, you know, there’s a lot that I don’t know, a lot of open questions about, sort of how to think about things like the gulags or concentration camps, places of ultimate dominance by officials.
John Meadowcroft
So it’s a story about governance, much less about the size of prisons, it’s about the presence or absence of formal governance.
David Skarbek
It’s the presence or absence and then if they’re present, are they providing helpful things or are they just controlling sort of every single move, every crumb of food, every breath of air. So are they present and if they are, what are they doing.
John Meadowcroft
One thing that’s striking is, as you said earlier, prisons are people, if you like, with fewer resources, histories of violence and what’s the word? Things have been done, that … victimisation is what I’m talking about. So should this lead us to be hopeful about people’s capacity for self governance, if even these people are capable of organising themselves to govern themselves?
David Skarbek
Yeah. You know, that’s one of the reasons I was initially interested in this research topic is to understand sort of, to look at a community in which we might think it’s least likely that cooperation would emerge. So in the academic literature, there are some fascinating studies of sort of elite and rich commercial enterprises that they’re able to engage in self governance over commerce. They can’t, they don’t rely on the state. And that’s sort of an interesting finding. But you wonder how much is it that these come from very privileged backgrounds that they’re able to maintain this level of cooperation. And so by turning to the prison, we tend to think that on average, prisoners tend to have less self control, they have less, they’re more impatient, they’re less willing to wait for future benefits. They’re more often perpetrators and victims of violence. So for all these things, you would think that cooperation will be far more difficult there. And so what I think, you know, sort of this research that I’m doing finds is that in small prison populations, those are not sufficient to undermine cooperation, the ability to rely on reputation, even if there’s a sort of bias in the agent type, cooperation can still emerge and work pretty well.
I think that In large prison populations, as those reputation mechanisms become less effective, then it becomes less stable. And this investment in sort of centralised mutual responsibility systems, it actually looks a lot like clan based societies. So if you look historically clan based societies are incredibly common if you look in many places in the world today, clans are a defining feature of social life and economic life. And where clans are most common, they’re common in places that have failed or weak states and that are of sufficient size that you kind of rely on these sort of decentralised mechanisms. So I think that gangs are the sort of clans of the prison society. And so to some extent, this biassed agent type is successful because it’s able to cooperate despite sort of the inherent difficulties on average of the characteristics of the members. But it’s not sufficient. To sort of overcome large populations of strangers and lack of state support.
John Meadowcroft
So it’s undoubtedly impressive that prisons are able to self organise, if you like self governance in this way, we figured out the comparative political economy of this though, would it be better if those formal governance in some sense is that safer? Less violent? All those things?
David Skarbek
Yeah, I think that if we were designing an ideal prison social system, nobody thinks that it should be the gangs that control the everyday life of prisoners. So it’s sort of like given the poor official governance and given the size of the population on the margin, you know, I think that gangs play an important role on the margin, probably a productive role. But no, yeah, I mean, I think the solution to gangs is more and better state produced governance. So I argue that people turn to gangs because they feel vulnerable in a dangerous environment. I argue that prisoners turn to gangs because they want access to resources that they can’t get from officials. So if you want to reduce the demand for prisoners to turn to gangs, it would seem that the answer is make prisons safer, arguably by a better-trained staff and smaller prison populations, and to increase access that they have to resources. So, yeah, I think that the appropriate lesson is to think that if you want to reduce the control of gangs, you need to improve the effectiveness of state-based governance.
John Meadowcroft
That’s a very interesting conclusion. Maybe for some people it might be a slightly challenging conclusion, I suppose. And I wanted to move on a little bit to talk about the basis of the research on which you base your conclusions. Essentially, you’re engaged In qualitative research, maybe the first question though, is to say a little bit about the challenges of obtaining that kind of data from prisons around the world and how you go about overcoming that challenge.
David Skarbek
Yeah, well, maybe if I can start with the sort of first project, the goal was essentially, so there’s never available to me the sort of quantitative data that I have in my mind, I sort of might most hope to have. And so I started to try to collect evidence in as many different ways as I could. So I’d go and collect legal documents like law enforcement documents and indictments that are describing case after case, you know, how these gangs are operating. But also visit prisons in California and talk to correctional officers, would meet with the gang investigations unit, which is the sort of police for gangs in prisons. I talked to prisoners, some sometimes behind bars, but often formerly incarcerated people after release to understand their experiences. So the goal you know, in addition to the secondary literature, of course, and so the goal was to sort of triangulate across these resources to understand how governance or to understand what gangs are doing today and what they did previously.
Historically, I collected data on the demographics of the prison population, which allowed me to track the sort of rise of the population and the rise of gangs. And the Department of Corrections also has a number of reports that were produced in the 50s and 60s and 70s, where they were writing about the problems prisoners faced, you know, and so the officials knew that prisoners were a lot more vulnerable, that things were more chaotic. They recognise the same problems. And so really, the goal was to sort of bring a sort of new institutional economics theory to bear to test and to triangulate across these different evidentiary sources to see if that was a plausible story.
And so I sort of argue in the book that that seems pretty persuasive. This new project is, is in some ways, just an entirely different method, logical approach. So the vast majority, you know, I’m sure 99% or more of prison ethnography is focused on a single prison, some within a single country, but usually it’s a single prison.. And ethnography is an incredibly time and resource intensive method that requires, you know, countless hours observing prisons, talking to prisoners, talking to staff. And that’s necessary in order to get the sort of fine grained, you know, thick, rich, descriptive understanding of the community, that’s an operation there. But for the most part, these studies produce a book or a few articles in a book, and the sort of the questions stop there. And so my argument in this book essentially is that we can use these individual single site studies ethnography as data in a broader comparative analysis.
And so my goal for this book is to sort of make an argument that prison ethnography and maybe prison sociology more generally should take a comparative turn. Just as in political science, comparative politics has been sort of, you know, a very sophisticated field methodologically for many decades, I think we can take a lot of the learning that was done in political science, and use it to inform how to make comparisons and how to compare across, in some ways, different cases, to understand or drop broader theories or understandings about governance.
John Meadowcroft
What’s your sense of the challenges of comparing different ethnographic studies, studies with chronic and geographical variation?
David Skarbek
I think that it must be done carefully. And I think that’s true in comparative politics, too. So prisons as I’ve sort of argued, do share core similarities, which means we should compare them, but there’s a lot of things that are different across them. So some ethnographies are just simply irrelevant to maybe a particular question that I want to ask. So for example, there’s a pretty extensive literature on sort of family relationships for people who are incarcerated. There’s a lot of sort of work in prison and food in prison. And these are interesting and important studies, but they’re sort of not directly relevant to the work that I’m trying, the questions that I’m trying to ask. So I think the goal is to read extensively enough that you can lead out ones that aren’t relevant, and then to start selecting the cases that you study based on some developed theory. So we say that you should select cases based on variation in the explanatory variables, which basically means that if you think that the size of the prison population matters, then you should try to find prisons that are small and prisons that are large, and an extent that they’re similar on other factors as well. Differences in social order can then be explained by differences in explanatory factors like size.
John Meadowcroft
I’ll come on to an explanation in a second but I was curious to also ask, as you spoke to people who also work in prisons or maybe undertake ethnographic research, have you had pushback from people who think that you shouldn’t take other people’s research and and reuse it if you like, in this way?
David Skarbek
Nobody. Actually no, no, I haven’t come across someone like that. And part of my argument is that the work that single site ethnographers are doing is even more valuable than we sort of understand it already because their work doesn’t just explain the case that they like, it helps us explain many other cases. And so I actually see it as an almost inspiring opportunity to do a sort of decentralised, but collaborative research programme, where you can continue to study single sites, but being conversation informally or just, you know, to a lesser extent with prison ethnography is working all around the world. And to me, that’s where we can really accumulate enough studies and enough tests, additional tests, that I think we can really make some advance in sort of learning. You know, across many different cases.
John Meadowcroft
So how does comparative ethnography limit, if you like, or extend the kind of questions we can answer?
David Skarbek
So I think that it expands our ability to understand … So ethnography gives us a deep description of the everyday life of prisoners. And for someone interested in governance, like I am, we need that ethnographic evidence. And so having more ethnography allows us to, I hope, develop a more sophisticated and complete understanding of governance than the sort of initial blocks that I’ve put together. So if we had five times as many ethnographies, I think that we’d be able to tell a lot more how important for example, size versus location is, or I make an argument in my first book that the diversity of the prison population matters. In doing qualitative case studies or ethnographies we can often say we think same size and diversity and location matter. But we can’t always say how much it matters. So if we had a lot more cases, we’d be able to disentangle sort of what is the relationship across all of our different explanatory variables. So for example, we could have prisons that were small and very homogenous prisons that were small and very diverse prisons that were large and very homogenous, and so on. And we’d have the permutations that we’d be able to tease out a lot more about what those relationships actually are.
John Meadowcroft
So you were trained as an economist originally, how would economists view this sort of methodological approach? Would they be concerned about your ability to really give causal answers?
David Skarbek
Yeah, I think that’s a fair, a fair observation. And I think that the sort of standard approach in economics is now focused on causal identification. So statistical tools that allow us to identify how X affects or causes Y I think that that’s an incomplete method for understanding prison social order for a few different reasons. The first is that the data is simply not available, it’s often not kept, and usually confidential. So there’s no data available. And in many cases, it’s difficult to find an identification strategy. So we can only tell us so much, even if it could tell us something about sort of abstract, causal relationships. What I think is really important about this research and about the broader sort of intellectual community that I’m engaged with is that it tells us a lot about causal mechanisms. So we may know that as X changes,Y changes, as the prison population goes up, the gangs emerge, but we don’t know necessarily why that’s the case. So this qualitative work allows us to identify the causal mechanism that explains the why behind a causal effect. So I think it really provides a radically different type. Understanding based on a very different type of evidence, something that I think traditional economic tools and data are, in most cases, not able to speak to.
John Meadowcroft
And as a political scientist, we see physical science going in a similar direction to economics in the past with an emphasis on causal identification on experimental results and so on. Should we be concerned about that? I mean, is it limited to the types of questions that political science can ask?
David Skarbek
I think it is a concern. My sense is that there’s starting to be some pushback. So strict adherence to causal identification, causal effects, limits our ability to ask a large number of important questions, sometimes the big questions, but it certainly – even in smaller, more manageable questions – limits what you can say. And my sense is that there’s a push for causal effects requiring some legitimate evidence about what the causal mechanism actually is. And you see that in the sort of top journals in political science that there may be a nice causal effect. But if it doesn’t have the causal mechanism, then it’s going to be much less likely to be persuasive. So, you know, when we have causal effects, it tells us what happened, but not why it happened. And I think the qualitative evidence, whether it’s ethnography, or interviews, or observations, allow us to know, not just what but why it happened. And I think there’s also going to be returned to theory. So I think that theory and ethnographic and qualitative work is going to allow us to interpret sort of broad relationships between X and Y.
John Meadowcroft
When you say theory there, what’s the theory you mean?
David Skarbek
I mean, formal theory. I think that the careful modelling of relationships, whether that’s game theory or something else, but you know, primarily on game theory in mind, I think is one way to get at what actually is going on in some broader relationship. And I actually think there’s a lot of complementarity between using qualitative evidence to test whether a particular theory is actually describing the process that’s going on, rather than just explaining the sort of start and finish of some causal effect.
John Meadowcroft
Yeah. Well, that makes sense. Well, you’ve done this great work on prisons. And I know you’ve been working on this for a number of years. I’m curious to know what’s next on the agenda. So I assume you’re not going to work in prisons forever. That may be too long. What are the sorts of ideas you have going forward,
David Skarbek
I’m sort of starting two new research projects. One is with an economist at Syracuse named Roger Cobbold, and we’re doing some work on the institutional setting in which criminal forensic science takes place. And it turns out that forensic science is much less scientific than depicted in TV shows and movies, but the institutions, the political institutions in which they operate, are also very imperfect, very deeply flawed, and we essentially are going to argue that the institutions in which forensic science takes place, lack accountability, independence, and replicability. And those are three crucial ingredients for science to weed out errors and learn its craft better. In addition to that, I’m also doing some research on mid 19th century American frontier violence. So this is a time period when many states were not states yet in the sort of Western frontier in America. And so my interest is understanding causes of violence across regions in this period. And when individuals turn to what are called vigilance committees, sort of extra legal justice in order to respond to violence, statelessness and inequality.
John Meadowcroft
Well, those are great topics, and we look forward to reading about them as you write about them. I really look forward to the publication of your new book, I think next summer I believe, so 2020. The Puzzle of Prison Order. And we hope you’ll come back and talk to us again in the future.
David Skarbek
I’d love to thank you. Thank you.
