Book Review: Divergent Opinions: Why Community Matters — A Review of Sunstein’s Going to Extremes

Going to Extremes: How Like Minds Unite and Divide, by Cass Sunstein. Oxford University Press: New York 2009. Pp. 171. $21.95

Cass Sunstein argues in his new book Going to Extremes: How Like Minds Unite and Divide that extremism is a phenomenon that is enhanced when people of like minds get together to talk. When we think of people that lie at the extremes of society, our minds are often drawn to reclusive characters. People like John the Baptist living in the wilderness “wearing clothes made of camel hair, eating locusts and wild honey;” (Matt. 3:3-4) or people like Raskolnikov from Fydor Doystoyveski’s Crime and Punishment – a reclusive character who develops a radical and warped sense of morality in response to his perception of society’s values. In reality, people that live on the extremes are rarely alone. They are surrounded by a network of like thinkers who confirm the attitudes, beliefs and interpretations of sensory data that those persons embrace as normal. Extremes are about information. That is, where you get your information from; whether you believe that information to be reliable, and how willing you are to accept information outside of your preferred source.

Going to Extremes is about how, when and why extremes develop in communities. The theme of the book is that “[w]hen people find themselves in groups of like-minded types they are especially likely to move to extremes” (p. 2). Sunstein’s work fits into the genre of human behavioral psychology proposed by James Sidanius and others that views extremists’ cognitive complexity as more complex than moderates. See James Sidanius, Functioning Sociopolitical Ideology Revisted, 6 POLITICAL PSYCHOLOGY 637, 639 (1985). This is in contrast to extremism theory, which largely assumes that political extremists display less-sophisticated cognitive behavior than moderates. About the form of extremism we call terrorism, Sunstein writes at one point,

it is tempting to think that terrorism is a product of extreme poverty, lack of education, or a kind of mental illness. It turns out that all of these thoughts are quite wrong. Most of the time, [terrorists] come from middle-income families. Nor have terrorists lacked education. There is no evidence that they suffer from mental illness…. Alan Krueger argues that terrorism is a form of political protest, and those who lack civil rights and civil liberties not having other means of engaging in protest resort to terrorism. To Krueger’s point, we might add that when civil liberties do not exist citizens have only one prominent source of information – the state – and that source cannot be trusted. (p. 115)

Terrorism then becomes a reaction against information that the extreme positions assume can’t be right. Thus, in Sunstein’s work, the why and how of extremisms (like terrorism) can be associated with how individuals interact in communities – the trust they place in the information received, the confidence they derive from like-minded members, and the authority or submission they respond to as a member of the community.

Sunstein supports his theme with various empirical anecdotes, studies and discussion. Sunstein begins by describing how people and groups gravitate towards extremism by polarizing their positions. (pp. 1-20). Chapter two describes how polarization occurs, considering factors that increase the likelihood of extremities; namely, Sunstein describes the role of assimilation, confidence, rhetorical speech, time, authority, and medium in the facilitation of polarization. (pp. 21-98). Chapter three considers several extremist movements and how those extreme movements have impacted American discourse (pp. 99-125). Chapter four considers how extremism can be prevented by traditionalism (pp. 127-135); systems of checks and balances (pp. 135-140) and the role of deliberative democracy. Sunstein’s final chapter illustrates some good consequences of extremism, namely its propensity for creating dialogue across the political spectrum.

This review makes two points regarding Sunstein’s work. One critique about Sunstein’s work as a whole is that it does not define extremism, but implies that extremism is a relative concept to community and time. Second, Sunstein’s work does a nice job, though scattered at times, of describing how information leads to confidence in communities, largely through anecdotal evidence. I want to focus on just one aspect of information based confidence – the role of technology in making information available and engendering confidence.

First, Sunstein does not offer or propose to offer a definition of extremism. At best, Sunstein implies that extremism is a function of communities. (In Sunstein’s defense, the work is not about defining extremism or how to recognize extremes). Nevertheless, the lack of a definition reduces the conversation to a question of community or political dynamics. I find two distinct problems with this approach. First, without a definitional starting point, the notion of extremism does not account for movements over time. Sunstein implies that extremism can have the effect of moving the middle over time. Sunstein says:

For any nation, second-order diversity may be especially important, certainly in the long run. If many organizations are allowed to exist, and if each of them is made up of like-minded people, the nation will ultimately benefit from the greater range of views and practices that emerge. Inevitably, several of those groups will be extreme, but their very extremism will enrich society’s argument pool and thus promote sensible solutions. (pp. 150-51).

I believe that Sunstein’s point is to say that extremism is a context dependent idea. That is, apart from the community or the time in which the community exists, it is difficult to pinpoint an exact meaning of what it means to be an extremist. If this is in fact Sunstein’s point, I think support can be found in a variety of places.

Second, assuming that Sunstein’s definition is community dependent (which I believe it is), the second definitional problem is how do you define the community in which extremism develops. Or, more problematically, can the community self-select itself. These two definitional problems of boundaries and self-selection are exemplified by looking at two American Jurisprudential Groups of the last thirty years – the Crits and the rise of religiously oriented law schools in the 1990s.

In the 1970’s the Critical Legal Studies movement sat on the left fringe of the American Legal Academy as a result of dissatisfaction with the legal academy’s approach to the law. Yet, the Crits were not necessarily welcomed members of the legal academy. As Neil Duxbury suggests, the Crits sought refuge from the law school culture in annual conferences, meetings and conventions, in which Crit ideology was not only embraced, but exacerbated. Eventually, the once-fringe movement, became an accepted part of the legal academy. Two narratives can be told of the Crit movement. One narrative is that the Crit movement forced the legal academy to move further to the left by its continuation over time, so that as more Crits infiltrated the ranks of the academy, the middle shifted. The second narrative is that the Crits self-selected a move to the middle. In other words, one story of Crit acceptance is that the Crits had to moderate their message in order to be accepted into the broader legal community. The other story is that the academy self-selected the crits as a known fringe part of the broader law school community.

In contrast, the rise of religiously affiliated law schools similarly came out of dissatisfaction with the legal academy. By religiously affiliated law schools, I am referring not to those schools that affiliate with a religious institution, but those institutions that endeavor to teach law from a religious perspective. For example, Regent University, Liberty University, and Ave Maria primarily come to mind. These schools emerged similarly to the Crits – beginning in smaller gatherings and then emerging as whole communities of legal training. Their isolation, though was different from the Crit isolation of the early 1980’s. First, the time was different. The Crits isolated themselves in enclaves for weekends at a time or in smaller less formal meetings when conventions were not possible. Afterwards, they returned to the world in which, no doubt, they themselves became moderated, and the non-crits became Crit-i-fied. The Crits thus found themselves as members of multiple communities – members of the academy during most of the year, and members of a Crit community when the enclave was able to meet (whether there be two or more were joined together). It seems that the opposite may be true for those religiously oriented schools I have named. Most of their time is spent as a member of a particular ideological community, with more limited time spent as a part of a mainstream group (perhaps at conferences and other larger community gatherings of American law schools). At this point, if the definition of extremism is dependent on what constitutes community, whether either of these groups are deemed extremists at any point in time depends largely on perception. Because being an extremist is so tied to community, defining both extremism and community would be useful in understanding what it means to be an extremist. These are questions not addressed by Sunstein’s work.

Second, the best part of Sunstein’s work is the description of information as a catalyst for community movement. Sunstein points out that the “key to extremism in all forms is the exchange of new information. Group polarization often occurs because people are telling one another what they know, and what they know is skewed in a predictable direction.” (p. 21). Sunstein ties extremism to confidence in information; when people are less confident in their information, they tend to moderate their views; however, when multiple trusted sources confirm information, people tend to become more extreme. (p. 23). The rise of the Internet and social networks has aided the ease by which people are able to find communes that instill confidence in one position or another. Importantly, these social online networks are not passive engagements, but rather are types of mediums that require regular investment by the participant. By regularly investing his or her time reading and contributing to the activity, participants become quietly both less moderate and more confident in their extremism. (p. 25)

The internet’s instant access allows people to by-pass what we might term general interest media, and instead focus on specialized media that is directly targeting one particular viewpoint or another. (The same problem could be associated with the rise of cable news directed to individual interests). One problem with this analysis is that it ignores the fact that the general interest media may have been the mainline media simply because there was no other option. Sunstein’s assumption that the general interest media would remain the general interest media in the context of choice may not be right. Nevertheless, his primary point on access is a good one. As people find like-minded people on the internet, communities based on interest as the single unifying principle begin to emerge into a homophilized zone of rhetoric. This is not necessarily out of the ordinary from interactions outside of the internet. People attend churches, mosques and synagogues that they find a theological correlation of beliefs and attitudes. Political parties are largely comprised of people that maintain shared interests in government, whether those interests derive from geographic, moral, financial or constitutional concerns. (p. 83-84) The internet does bring two new forms of interaction to the traditional community. First, individuals can build confidence derived mostly from anonymity. Thus, the internet can effectively allow someone to build a different sense of confidence in a community than they would if that community interacted in person. We saw this effect in December when local community members were shocked to learn a young doctor who wore blue jeans and baseball caps led a double life an Islamic jihadist online; they of course learned this after the doctor carried out a bombing that killed himself and seven CIA agents. The internet for good and for bad allows people to segregate their identities into convenient communes.

Second, the community, in theory, is always meeting online. In the form of forums, blogs and list-serves, people can express their views to the community or read the community’s views anytime that is convenient. Thus, the internet becomes a perpetual community, which may be accessed anytime, and confirm the views that the individual already believes.

Sunstein’s work on extremism is simple and intuitive, but also very thoughtful and artistic in its assessment of extremist behavior. Sustein demonstrates not only the detrimental effects of extremism but also its positives. To wit, Going to Extremes makes a distinctive contribution to the literature of human behavioral economics, and deserves attention by academics, judges and attorneys alike.


Marc Roark is Assistant Professor of Law, University of La Verne College of Law. He would like to thank Brannon Denning, Daniel Solove, Malana Jones, and Chuck Doskow for reading prior drafts and Allen Easley, John Linarelli, and Diane Uchimiya for thoughtful comments and discussion of this material.

You may also like...

2 Responses

  1. Vinca says:

    Cass Sunstein and Adrian Vermeule have proposed a means – “cognitive infiltration” – of addressing what they term the “crippled epistemologies” of people with views they deem extremist.

    Glen Greenwald criticizes their proposal here:

    Sunstein and Vermeule do not explain why so many hold these “extremist” views in a society like ours with civil rights and civil liberties, other than by assuming that government information is inherently trustworthy in an open society.

    Perhaps, but this information and the community of people who receive it can also be viewed in terms of “the trust they place in the information received, the confidence they derive from like-minded members, and the authority or submission they respond to as a member of the community.”

  2. Vinca says:

    Sunstein and Vermeule seem limited by their Ivy League educations.

    “So are you saying that we’re all just, like, really excellent sheep?” Well, I don’t know. But I do know that the life of the mind is lived one mind at a time: one solitary, skeptical, resistant mind at a time.