second_news.jpg
Home > Alumni > Alumni Magazine > Faculty Books > Why Societies Need Dissent

Why Societies Need Dissent
by Sarah A. Sulkowski, '05

Cass SunsteinIn his recent book, Why Societies Need Dissent, Cass Sunstein argues that those who express disagreement with prevailing views are to be commended, not condemned. According to Sunstein, although dissenters are often portrayed as selfish, they actually provide society with information it needs to make the right decisions. When all members of a group appear to agree, they may encourage each other to move toward extremes to which no individual member alone would be willing to go. The theory has far-reaching implications for organizations of all kinds: the book recommends that Presidents and CEOs appoint advisors to serve as "devil's advocates" and that groups establish procedures for rewarding those who speak out.


Why Societies Need Dissent focuses heavily on the benefits that dissenters provide to others. Why did you choose this approach, and why did you write the book now, at a time when dissent is not particularly encouraged by many of our national leaders?

I wanted to take a broader look at dissent as a political phenomenon, one that would go beyond the legal issues. A lot of people on the faculty have tackled those issues; they're important, but dissent raises even larger issues. I didn't mean to focus the book on the post-9/11 atmosphere; conformity and dissent are pretty universal issues. But one of my goals was to suggest, even in the current climate, that we protect dissenters not for their sake but for ours. We're much more likely to blunder if we impose pressures toward conformity. Whistleblowers are often an organization's best friend.

In discussing affirmative action, you distinguish between courses like law, which may be improved by increased diversity in the classroom, and subjects like math, which may not. But can't affirmative action in education be justified on the grounds that it fosters tolerance and therefore makes dissenters more effective across the board?

Tolerance is better than intolerance, but it's not clear that we get more of the former if there are affirmative action programs to get more Hispanics into graduate school in  math and physics. In fact I don't much explore the connection between affirmative action and tolerance; diversity of opinion is my focus. A note on tolerance: The word suggests that there's something bad out there that people ought to "tolerate" (people, ideas). Respect is a lot better than tolerance, no?

By focusing exclusively on the societal benefits of dissent, does the book run the risk of devaluing more traditional individual-rights-based arguments supporting affirmative action and civil liberties?

Maybe. But we might overemphasize those individual rights arguments. At least we underemphasize the benefits, to organizations and nations, of having dissenting opinions.

The book mentions the Condorcet Jury Theorem, which states that the larger a group is, the more likely it is to answer a question correctly, so long as each group member is more likely than not to know the right answer. This suggests that public opinion polls might be the best way to set policy. But so-called "poll-driven" politics have come under fire from many quarters in recent years. Is this opposition to polls misplaced?

The problem with opinion polls is that we aren't likely to get good or even stable answers from snapshots of nondeliberative opinions. Especially on technical issues, poll results tell us little about what's right, especially because questions can be framed in many ways, and framing affects answers. The problem is that groups and organizations often fail to aggregate the information that their members have. A crucial task is to elicit what people know. If members follow a few early speakers, or leaders, they'll fail to carry out that task. Both the CIA and NASA appear to be recent examples.

The book suggests that compliance with the law might be increased by informing violators of the large percenage of people who comply. But wouldn't such a cascade deprive society of valuable information about the perceived fairness of its laws? 

A cascade in favor of obeying the law isn't the worst thing! If we can produce a cascade against violent crime, we'd really have something to celebrate. Of course civil disobedience has its place, one that is honorable — but very small. So if social pressures can be enlisted in favor of obeying the law, so much the better.

You also note some areas in which dissent does not improve society's understanding. Do you have a theory about when a society can feel comfortable enough about its beliefs to disregard minority opinions?

I don't think there's a theory on that one. Learned Hand said that the spirit of liberty is that spirit which is not too sure that it is right, and that's a pretty good place to start.

How should people choose in their everyday lives between following others and expressing their own opinions?

A lesson of the book, one that I didn't expect to draw, is this: Say what you think. Those who follow others, when they have some contrary information of their own, are actually behaving selfishly. Civility is important, of course, but if you have reason to disagree, you owe it to others to disagree.