View source for Pursuing Cognitive Democracy

<!-- you can have any number of categories here -->
[[Category:Henry Farrell]]
[[Category:Cosma Shalizi]]
[[Category:Democracy]]
[[Category:From Voice to Influence: Understanding Cit- izenship in a Digital Age]]
<!-- 1 URL must be followed by >= 0 Other URL and Old URL and 1 End URL.-->
{{URL | url = http://henryfarrell.net/wp/wp-content/uploads/2013/04/Farrell-paper.pdf}}
<!-- {{Other URL | url = }} -->
<!-- {{Old URL | url = }} -->
{{End URL}}
{{DES | des = Compares failings of markets and hierarchy as problem solving techniques, and shows how democracy can solve problems they cannot.  Pp. 211–231 in [[From Voice to Influence: Understanding Citizenship in a Digital Age]] | show=}}
<!-- insert wiki page text here -->
<!-- DPL has problems with categories that have a single quote in them.  Use these explicit workarounds. -->
<!-- otherwise, we would use {{Links}} and {{Quotes}}  -->
{{List|title=Pursuing Cognitive Democracy|links=true}}
{{Quotations|title=Pursuing Cognitive Democracy|quotes=true}}

{{Text |
The rise of the Internet makes this an especially good time for experi- menting with democratic structures.1 Democracy is uniquely fitted to help people with highly diverse perspectives to come together to solve problems collectively. Democracy can do this better than either markets and hierar- chies, because it brings these diverse understandings into direct contact with each other, allowing forms of learning that are unlikely either through the price mechanism of markets or the hierarchical arrangements of bu- reaucracy.
Yet the means available for communication and information process- ing constrain the possibilities for collective decision making. Bureaucracy was not an option in the Old Stone Age, nor was representative democracy possible without something like cheap printing. We do not yet know the possibilities of Internet-mediated communication for gathering dispersed knowledge, for generating new knowledge, for complex problem solving, or for collective decision making, but we really ought to find out. Democ- racies can, by experimenting, take advantage of novel forms of collective cognition that are facilitated by new media.
We do not believe that new media are a panacea for the problems of modern democracy. Democracy is undergoing a crisis of representation in the United States, where the interests of the poor are systematically ig- nored by the political system, and in other regions such as Western Europe, where crucial economic decisions are made in forums outside the grasp of democratic accountability. Naive accounts of new media suggest that it empowers ordinary individuals against the strong. Instead, new technol- ogies may serve to cement the advantages of large corporations and the state. Neoliberalism is steadily replacing democratic relationships of ac- countability with diffuse and unaccountable networks between politicians and business interests.
Our argument is not that new media leads to better democracy, but that proper democracy leads to better use of the potential of new media. It is only in the context of strong and indeed radical democratic institutions
Uncorrected proofs for review only
212 Chapter Nine
that new media can realize their potential for improving problem solving and decision making. By changing political structures and minimizing power asymmetries we can ensure that new media can be properly em- ployed to encourage “collective cognition” and resolve problems.
Before we return to the relationship between democracy and new me- dia, we need to offer a full outline of a cognitive approach to democracy, an approach that is inspired by older work on political decisionmaking but is also newly relevant because of the possibilities inherent in new me- dia. Much of what we say is synthetic. On the normative side, we build both on the academic literature (Cohen 1986, Anderson 2007 on epistemic democracy; Knight and Johnson 2011 on the pragmatist case for radically egalitarian democracy; Rosenblum 2008 on partisanship; Landemore and Elster 2012 on Collective Wisdom), and on public intellectuals such as Cory Doctorow, Chris Hayes, Steven Berlin Johnson, Clay Shirky, Tom Slee, and the late Aaron Swartz. We steal outrageously, in different ways, from Josiah Ober and Scott Page, and draw much more on Charles Lindblom than our explicit citations suggest. Much of the prior literature focuses on epistemic questions, about whether democracy can reliably determine the truth. Throughout, we emphasize the cognitive benefits of democracy, how it combines multiple mechanisms to build a powerful collective process for solving problems.
We start by explaining one thing which social institutions should do: they find solutions to social problems, which means they have cognitive, information-processing goals. From this viewpoint, we discuss sophisti- cated arguments on behalf of markets (Hayek’s “catallaxy”) and hierarchy (Richard Thaler and Cass Sunstein’s “libertarian paternalism”) and the inadequacies of these arguments. We then lay out our arguments for de- mocracy, emphasizing how democratic procedures have problem-solving capacities that other social forms do not. The penultimate section discusses how democracy can learn from new forms of collective cognition on the Internet, treating these forms not as ideals to be approximated, but as im- perfect experiments whose successes and failures can teach us about the conditions for better decision making; this is part of a broader agenda for cross-disciplinary research involving computational scientists and demo- cratic theorists. Our argument is very much alive to new democratic pos- sibilities enabled by new media, but it also intersects with the normative questions raised in this volume by Howard Gardner and Angel Parham and Danielle Allen as well. Our argument for cognitive democracy reinforces a case for the value to democracy of interestedness.
Uncorrected proofs for review only

Pursuing Cognitive Democracy 213 JUSTIFYING SOCIAL INSTITUTIONS
Three of the most widespread and enduring kinds of macro-institutions are markets, decentralized forms of economic exchange governed by prices; hierarchies, centralized forms of decision making in which inferiors report to superiors and superiors issue authoritative orders to inferiors through more or less elaborate chains of command; and democracies, forms of collec- tive decision making in which individuals are able both to argue with each other over proposals in a nonhierarchical setting and to vote over which proposals are accepted as collectively binding. (This list is not exhaustive.) What might justify such institutions?
The dominant tradition in political theory tends to evaluate macro- institutions such as politics, markets, and hierarchies in terms of justice— whether institutions use procedures, or give results, that are just according to some reasonable normative criterion. Others, more cynically or more modestly, look to institutions’ contributions to stability—whether they produce an acceptable level of social order, reducing violence and provid- ing a modicum of predictability. We start instead with a pragmatist ques- tion: whether these institutions are useful in helping us solve difficult social problems.2
Some political problems are simple; the solutions might not be easy to put into practice, but the problems are easy to analyze. But the most vexing problems are usually ones without any very obvious solutions. How do we change legal rules and social norms in order to mitigate the problems of global warming? How do we regulate financial markets so as to minimize the risk of new crises emerging, and limit the harm of those that happen? How do we best encourage the spread of human rights internationally?3
These problems all share two important features. First, they are social. That is, they are problems which involve the interaction of many human beings, with different interests, desires, needs and perspectives. Second, they are complex problems, in the sense that scholars of complexity under- stand the term. To borrow the definition of Page (2011, p. 25), they involve “diverse entities that interact in a network or contact structure” (italics in the original).4 They are a result of behavior that is difficult to predict, so that consequences to changing behavior are extremely hard to map out in advance. Finding solutions is difficult, and even when we find a solution, it is hard to know whether it is better than others, let alone the best.
Macro-institutions will best be able to tackle these problems if they have two features. First, they should foster a high degree of direct communication
Uncorrected proofs for review only

214 Chapter Nine
between individuals with diverse viewpoints. This kind of intellectual di- versity is crucial to finding good solutions to complex problems. Second, they should provide relative equality among affected actors in decision-making processes, so that socially or politically powerful groups do not block so- cially beneficial changes that conflict with their own particular interests.
We base these contentions on arguments from work on collective prob- lem solving and theories of political power. Both are clarified if we think of the possible solutions to a difficult problem as points on a landscape in which we seek the highest point.5 Difficult problems present many peaks (solutions which are better and “higher” in value than the other points immediately adjacent to them). Such landscapes are rugged—they have some degree of organization, but are not so structured that simple algorithms can quickly find the best solution. There is no guarantee that any particular peak is glob- ally optimal (i.e., the best solution across the entire landscape) rather than locally optimal (the best solution within a smaller subset of the landscape).
Solving a complex problem involves a search across this landscape for the best visible solutions. Individual agents have limited cognitive abili- ties and (usually) limited knowledge of the landscape. Both of these make them likely to get stuck at local optima, which may be much worse than even other local peaks, let alone the global optimum. Less abstractly, people may settle for bad solutions because they do not know better (they cannot perceive other, better solutions). It may even be, as McAfee argues (this volume), that there is no “right” answer in any absolute sense, although it is hard to capture this possibility using the ideas that we invoke here.
Hong and Page (2004) use mathematical models to argue that diversity of viewpoints helps groups find better solutions. The intuition is that different individuals, when confronting a problem, “see” different landscapes—they organize the set of possible solutions in different ways, which are more or less useful in identifying good peaks. Very smart individuals (those with many mental tools) have better organized landscapes than less smart indi- viduals, and so are less likely to get trapped at inferior local optima. How- ever, at the group level, diversity of viewpoints matters a lot. Hong and Page find that “diversity trumps ability.” Groups with high diversity between in- ternal viewpoints are better able to identify optima than groups composed of much smarter individuals with more homogeneous viewpoints. By put- ting their diverse views together, the former are able to identify attractive solutions that would be invisible to more homogeneous groups.6
Hong and Page say little about how individuals share their diverse per- spectives with each other, allowing them to build a common framework.
Uncorrected proofs for review only

Pursuing Cognitive Democracy 215
However, their arguments suggest that actors’ different points of view need to be exposed directly to each other, in order to identify the benefits and drawbacks of different points of view, the ways in which viewpoints can be combined to better advantage, and so on. Such claims are supported by work in sociology and elsewhere (e.g., Burt 1992).
The second issue for collective problem solving is more obvious. Even when groups are able to identify good solutions (relatively high peaks in the solution landscape), they may not be able to reach them. In particular, actors who benefit from the status quo (or who would prefer less generally beneficial solutions) may be able to use political and social power to block movement towards such peaks, and instead compel movement towards solutions that have lower social and greater particular benefits. Research on problem solving typically does not talk about differences in actors’ in- terests, or in their ability to pursue them given, for example, bargaining power, level of rhetorical training, and access to symbolic resources. While different individuals initially perceive different aspects of the landscape, researchers assume that once they are able to communicate with each other, they will all agree on how to rank visible solutions. But actors may have diverse interests as well as diverse understandings of the world (and the two may indeed be systematically linked). They may even be working in such different landscapes, in terms of personal advantage, that one ac- tor’s peak is another’s valley, and vice versa. Moreover, actors may differ in their ability to ensure that their interests are prosecuted. Recent work in political theory (Knight 1992;, Knight and Johnson 2011), economics (Bowles and Naidu 2008), political science (Hacker and Pierson 2010), and sociology details how powerful actors can compel weaker ones to accept solutions that are to the advantage of the former, but which have lower overall social benefits.
Here, relative equality of power can have important consequences. Individuals in settings with relatively equal power relations, are, ceteris paribus, more likely to converge on solutions with broad social benefits, and less likely to converge on solutions that benefit smaller groups of indi- viduals at the expense of the majority. Furthermore, equal power relations may not only make it easier to converge on “good” solutions when they have been found, but may stimulate the search for such solutions. Participating in the search for solutions and in decision making demands resources (at a minimum, time), and if those resources are concentrated in a small set of actors with similar interests and perspectives, the solutions they will find will be fewer and worse than if a wide variety of actors can also search.
Uncorrected proofs for review only

216 Chapter Nine
Since we need macro-institutions to solve complex social problems, and their capacity to do this depends on bringing together people with different perspectives and sharing decision-making power relatively equally among those people, we will look at how the broad logics of different macro- institutions relate to those two criteria. Hierarchy, while a remarkable and enduring social form, suffers from dramatic informational flaws. In response, for decades now many scholars and policy makers have been de- voted to pushing markets as the way to address social problems that are too complex to be solved by top-down authority. As we show below, markets, if they embody substantial power inequalities and homogenize human re- lations, are unlikely to possess the virtues attributed to them, though they surely have more particular benefits under specific circumstances (Shalizi 2012). This prompts us to advocate democracy not for the sake of justice or stability, but as a tool for solving complex problems.
MARKETS AND HIERARCHIES FOR PROBLEM SOLVING
Both markets and hierarchies have their advocates among scholars and public intellectuals, who believe that one or the other (or both) are better ways of solving complex problems than democracy. Advocates of markets usually build on the groundbreaking work of Hayek (1937, 1945), to argue that this form of organization is better at eliciting information and putting it to good work than any more collective form. Advocacy of hierarchy is much older, and, perhaps for that reason, does not have any such unified tradition. However, Thaler and Sunstein (2008) have recently made a so- phisticated case for the benefits of hierarchy. They advocate a combination of top-down mechanism design and institutions designed to guide choices rather than to constrain them—what they call libertarian paternalism—as a way to solve difficult social problems. Hayek’s arguments are not the only case for markets, and Thaler and Sunstein’s are not the only justification for hierarchy. They are, however, among the best such arguments, and hence they provide a good initial way to test the respective power of markets, hierarchies, and democracies to solve complex problems.
MARKETS
Hayek’s account of the informational benefits of markets is groundbreaking (Hayek 1937, 1945). Although it builds on the insights of others (particularly
Uncorrected proofs for review only

Pursuing Cognitive Democracy 217
Michael Polanyi), it is arguably the first real effort to analyze how social in- stitutions work as information processors. Hayek reasons as follows. Much of human knowledge (as Polanyi argues) is practical and cannot be fully articulated (“tacit”). This knowledge is nonetheless crucial to economic life. Hence, if we are to allocate resources well, we must somehow gather this dispersed, fragmentary, informal knowledge and make it useful.
Hayek is explicit that no one person can know all that is required to al- locate resources properly, so there must be a social mechanism for such in- formation processing. He considers three such mechanisms: central plan- ning, planning by monopolistic industries, and decentralized planning by individuals. The first and second of these break down in the face of the need for vast amounts of tacit knowledge, which cannot be conveyed to any cen- tralized authority. Centralized or semicentralized planning are especially poor at dealing with the constant flows of major and minor changes through which an economy (or, as Hayek would prefer, a catallaxy or spontaneous order) approaches balance. To deal with such changes, we need people to make the necessary decisions on the spot—but we also need a way to convey information about changes in the larger economic system to these people. The virtue of the price mechanism, for Hayek, is to compress dif- fuse, even tacit knowledge about specific changes in specific circumstances into a single index, which can help individuals adapt to changes elsewhere. I do not need to grasp the intimate local knowledge of the farmer who sells me tomatoes in order to decide whether to buy them. The farmer needs to know the price of fertilizer, not how it is made, or what it could be used for other than tomatoes, or the other uses of the fertilizers’ ingredients. The information that we need to decide whether to buy tomatoes or fertilizer comes through prices, which may go up or down, depending on the aggre- gate action of many buyers or suppliers, each working with their own tacit understandings.
This insight is both crucial and beautiful,7 yet it has stark limits. It sug- gests that markets will be best at conveying a particular kind of information about a particular kind of underlying fact: the relative scarcity of differ- ent goods. As Stiglitz (2000) argues, market signals about relative scarcity are always distorted, because prices embed information about many other economic factors. More importantly, although information about relative scarcity surely helps markets approach some kind of balance, it is of little help in solving more complicated social problems, which may depend not on allocating existing stocks of goods in a useful way, given people’s dis- persed local knowledge, so much as on discovering new goods or new forms
Uncorrected proofs for review only

218 Chapter Nine
of allocation. More generally, Hayek’s well known detestation for projects with collective goals leads him to systematically discount the ways in which aggregate knowledge might work to solve collective rather than individual problems.8
This is unfortunate. To the extent that markets fulfill Hayek’s criteria and mediate all relevant interactions through the price mechanism, they foreclose other forms of intellectual exchange. In particular, Hayek’s dep- recation of “rationalism” leaves little place for reasoned discourse or the useful exchange of views. In Hayek’s markets, people communicate only through prices. The advantage of prices, for Hayek, is that they inform in- dividuals about what others want (or don’t want) without requiring anyone to know anything about anyone else’s plans or understandings. But there are many useful forms of knowledge that cannot readily be conveyed in this way.
People may learn something about those understandings as a by-product of market inter- actions. In John Stuart Mill’s description:
But the economical advantages of commerce are surpassed in im- portance by those of its effects which are intellectual and moral. It is hardly possible to overrate the value, in the present low state of human improvement, of placing human beings in contact with persons dissim- ilar to themselves, and with modes of thought and action unlike those with which they are familiar. Commerce is now what war once was, the principal source of this contact. (Mill 1909, §III.17.1).
When this learning happens, however, it is incidental, a side effect of buying and selling to best advantage. As markets grow depersonalized, they lead to less of this salutary contact between different modes of life. The resurgence of globalization, the creation of an Internet where people buy and sell with those they will only ever know as account names, the replace- ment of local mores with global standards—all these provide enormous efficiency gains, and allow information about supply and demand to flow more smoothly. Yet they also undermine the Millian benefits of commerce, making it less likely that individuals with different points of view will be exposed to each other’s perspectives. So much, in other words, for the pos- sibilities for “rooted cosmopolitanism,” for which Parham and Allen argue. More tentatively, markets may themselves have a homogenizing impact on differences between individuals and across societies, increasing peace but reducing diversity (Hirschman 1992). Sociologists such as Meyer et al.
Uncorrected proofs for review only

Pursuing Cognitive Democracy 219
(1977) find evidence of increased cultural and social convergence across nations as a result of exposure to common market and political forces.
Finally, it is unclear whether markets tend to reduce or reinforce power inequalities in modern democracies. It is almost certainly true, as Marx argues, that the spread of markets helped undermine such historical forms of domination as feudalism. It is not clear that markets continue to do so. On the one hand, market participation gives people some ability (presuming equal market access, etc.) to break away from abusive relationships. On the other, markets give greater voice and choice to those with more money; if money talks in politics, it bellows across the agora. Nor is this limited to the marketplace. The market facilitates and fosters asymmetries of wealth which in turn may translate, directly or indirectly, into asymmetries of political influence (Lindblom 1977, 1982). Untrammeled markets are asso- ciated with gross income inequalities, which in turn infect politics with a variety of pathologies. Markets are at best indifferent levelers of unequal power relations.
HIERARCHY
Authoritative hierarchy is endorsed (if not presumed) by what are, histori- cally, the oldest, most widespread, and most enduring traditions of political philosophy. In modern times, especially within the last century, it has come to seem normatively objectionable, but it remains an extraordinary polit- ical achievement. States with clear, accountable hierarchies can achieve vast and intricate projects, and businesses use hierarchies to coordinate highly complex chains of production and distribution.9 Accordingly, it is not without its defenses even now. In particular, the recent influential book by Thaler and Sunstein (2008) is a sustained brief for hierarchy in solving complex problems.
Thaler and Sunstein argue that “choice architects,” people who have “re- sponsibility for organizing the context in which people make decisions,” can design institutions so as to spur people to take better choices rather than worse ones. Thaler and Sunstein are self-consciously paternalist, claiming that flawed thinking consistently stops people from doing what is in their best interests. However, they also find direct control of people’s choices morally opprobrious. Libertarian paternalism seeks to guide but not eliminate choice, so that the easiest option is the “best” choice, the one people would make if they only had enough attention and discipline. It provides paternalistic guidance through libertarian means, shaping choice
Uncorrected proofs for review only

220 Chapter Nine
contexts to make it more likely that individuals will make the right choices rather than the wrong ones.
This is, in Thaler and Sunstein’s words, a politics of “nudging” choices rather than dictating them. Although Thaler and Sunstein do not put it this way, it is also a plea for the benefits of hierarchy in organizations, espe- cially in government. Thaler and Sunstein’s “choice architects” are hier- archical superiors, specifically empowered to create broad schemes that will shape the choices of many other individuals. Their power to do this does not flow from, for example, accountability to those whose choices get shaped. Instead, it flows from positions of authority within a firm or a government, which allow the architects to authoritatively craft pension contribution schemes within firms, environmental policy within the gov- ernment, and so on.
Thaler and Sunstein’s recommendations have outraged libertarians, who believe that a nudge is just a well-aimed shove—that individuals’ freedom will be reduced nearly as much by Thaler and Sunstein’s choice architecture as it would be by direct coercion. We are also unenthusiastic about libertarian paternalism, but for different reasons. While we do not talk here about coercion, we have no particular normative objection to it, provided that it is proportionate, directed towards legitimate ends, and constrained by well-functioning democratic controls. Instead, we worry that the kinds of hierarchy that Thaler and Sunstein presume actively in- hibit the unconstrained exchange of views that is essential to solving com- plex problems.
There are, after all, reasons why bureaucracies have few modern de- fenders. Hierarchies need power asymmetries to work. Inferiors take or- ders from superiors, in a chain of command. This is good for pushing orders down the chain, but notoriously poor at transmitting useful information up, especially information that superiors have not anticipated wanting. Furthermore, as scholars from Max Weber onward have emphasized, bu- reaucracies systematically encourage a culture of conformity in order to increase predictability and static efficiency.
Thaler and Sunstein discuss ordinary people’s bad choices at length. However, they have remarkably little to say about how it is that the ar- chitects housed atop the hierarchy can figure out better choices on these individuals’ behalf. Sometimes, Thaler and Sunstein suggest that choice architects can rely on introspection: “Libertarian paternalists would like to set the default by asking what reflective employees in Janet’s position would actually want.” At other times, they imply that choice architects can
Uncorrected proofs for review only

Pursuing Cognitive Democracy 221
use experimental techniques; this is hard for any sort of social problem, involving how people interact with each other,10 and it becomes intractable precisely with complex problems, where the set of candidate solutions is too vast for systematic exploration. Finally, Thaler and Sunstein sometimes argue that choice architects can use results from the social sciences to find optima.
One mechanism of information gathering that Thaler and Sunstein sys- tematically ignore is active feedback from citizens. Although they argue in passing that signals from choice architects can help guide consumers—for example, by giving information about the content of food, or by shaping online interactions to ensure that people are exposed to others’ points of view—they have no place for feedback from the individuals whose choices are being manipulated to help guide the choice architects, let alone con- strain them. As Mettler (2011) has pointed out, Thaler and Sunstein depict citizens as passive consumers who need to be guided to the desired out- comes, rather than active participants in politics.
This also means that Thaler and Sunstein’s proposals don’t take advan- tage of diversity. Choice architects, located within hierarchies which tend to promote conformity, are likely to have a much more limited range of ways of understanding problems than the population whose choices they are seeking to structure. In Page’s terms, these architects may be very “able,” but they will also be very similar to each other in background and training (professionals, in other words), so that as a group they will see a far more limited set of possibilities than a group of randomly selected members of the population (who are likely to have less sophisticated but far more diverse heuristics). Cultural homogeneity among hierarchical elites helps create policy disasters (the “best and brightest” problem). Direct involve- ment of a wider selection of actors with more diverse heuristics would al- leviate this problem.
However, precisely because choice architects rely on hierarchical power to create their architectures, they will have difficulty in eliciting feedback, even if they want to. Inequalities of power notoriously dampen real ex- changes of viewpoints. Hierarchical inferiors within organizations worry about contradicting their bosses. Ordinary members of the public are un- comfortable when asked to contradict experts or officials. Work on group decision making (including Sunstein 2003) is full of examples of how per- ceived power inequalities lead less powerful actors either to stay silent or merely to affirm the views of the more powerful, even when they have independently valuable perspectives or knowledge.
Uncorrected proofs for review only

222 Chapter Nine
In short, libertarian paternalism is flawed not because it restricts peoples’ choices, but because it makes heroic assumptions about choice architects’ ability to figure out what the choices should be, and because it blocks the architects’ channels for learning better. Libertarian paternalism may still have value where people probably do want to save more money or take more exercise but face commitment problems, or when other actors have an incentive to misinform them or structure their choices perversely in the absence of a “good” default. However, it will be far less useful, or even ac- tively pernicious, in complex situations, where many actors with different interests make interdependent choices. Thaler and Sunstein are far more convincing when they discuss how to encourage people to choose appro- priate pension schemes11 than when they suggest that environmental prob- lems are the “outcome of a global choice architecture” that could be usefully rejiggered through voluntaristic mechanisms.
DEMOCRACY AS A WAY TO SOLVE COMPLEX PROBLEMS
Is democracy better at identifying solutions to complex problems? Many— even on the left—doubt this. We believe that they are wrong, and that de- mocracy offers a better way of solving complex problems. Since, as we’ve argued, power asymmetries inhibit problem solving, democracy has a large advantage over both markets and technocratic hierarchy. The fundamental democratic commitment is to equality of power over political decision mak- ing. Real democracies do not attain this any more than real markets deliver perfect competition or real hierarchies deliver an abstractly benevolent interpretation of rules. But just as a commitment to markets is a commit- ment to improving competition, and a commitment to hierarchy (in its pos- itive aspects) is a commitment to greater disinterestedness (as explained in chapter 10 of this book), a commitment to democratic improvements is a commitment to making power relations more equal.
Besides equality, the other requirement for solving complex social prob- lems is exposing diverse points of view to each other; we saw that both markets and hierarchies are bad at this. But democracy is good at fostering such engagement and rendering it productive, because, along with voting, it involves debate. In competitive elections and in more intimate discus- sions, democratic actors argue over which proposals are better or worse, exposing their different perspectives to each other. It is for this reason that, as Knight and Johnson (2011, p. 151) put it, “democratic decision processes
Uncorrected proofs for review only

Pursuing Cognitive Democracy 223
make better use of the distributed knowledge that exists in a society than do their rivals” such as market coordination or judicial decision making. It is the combination of voting and debate that makes it work.
And yet, democratic debate looks ugly. It is partisan, rancorous, and vexatious, and people seem to never change their minds. Not least for this reason, the dominant tradition of American liberalism actually has considerable distaste for the less genteel aspects of democracy. The early- twentieth-century Progressives and their modern heirs deplore partisan- ship and political rivalry, instead preferring technocracy, moderation, and deliberation (Rosenblum 2008). Reinforcing this, earlier epistemic accounts of the benefits of democracy have relied on results such as Con- dorcet’s Jury Theorem, which says that as long as people are even slightly more likely to be right than wrong, and do not seek to influence each other, the more people who vote on a particular question, the more likely it is that they will arrive at the correct answer. Such approaches have assumed that people judge best when they judge in isolation from each other. This in turn implies that democracy would work better still if debate could be suppressed—that debate is not just vulgar; it is a weakness.
Such reactions miss two crucial points. The first is that, as Knight and Johnson (2011) say, politics is a response to the problem of diversity. Actors with differing—indeed conflicting—interests and perceptions find that their fates are bound together, and that they must make the best of this. We are lucky when genuine conflicts over things that really matter to people lead only to ugly words. Second, and more subtly, as Knight and Johnson argue, politics can also seek to harness diversity so as to generate useful knowledge. This does not require impartial deliberation; rather, partial and self-interested debate can have practical benefits. Knight and Johnson suggest that approaches based on diversity (e.g., Page 2007 and Anderson 2007) are better guides here than the earlier epistemic accounts.
We agree, yet Knight and Johnson do not explain how clashes between different actors with different viewpoints result in better decision making. Without some account of the mechanisms engaged, it could be that conflict between perspectives results in worse rather than better problem solving. To make a good case for democracy, we need not only to bring diverse points of view to the table, but also to show that the specific ways in which they are exposed to each other have beneficial consequences for problem solving.
Important micro-level work speaks directly to this, however. Mercier and Sperber (2011) advance an “argumentative” account of reasoning, on which reasoning is intended not to reach right answers, but rather to eval-
Uncorrected proofs for review only

224 Chapter Nine
uate the weaknesses of others’ arguments and come up with good argu- ments to support one’s own position. This explains both why confirmation bias and motivated reasoning are rife, and why the quality of argument is significantly better when actors engage in real debates. Experimentally, individual performance when reasoning in nonargumentative settings is “abysmal,” but is “good” in argumentative settings. This, in turn, means that groups are typically better in solving problems than is the best individ- ual within the group.12 Indeed, where there is both diversity of opinion and competition, confirmation bias can have positive effects in pushing people to evaluate and improve their arguments.
A separate line of research in experimental social psychology (Nemeth et al. 2004; Nemeth and Ormiston 2007; Nemeth 2012) indicates that problem-solving groups produce more solutions, which outsiders assess as better and more innovative, when they contain persistent dissenting mi- norities and are encouraged to engage in, rather than refrain from, mutual criticism (Mercer [2000] reports similar effects in schoolchildren). This makes a great deal of sense, from Mercier and Sperber’s perspective. It also provides more reasons to value the “impure” kinds of dissent described by Tommie Shelby in chapter 3 of this volume.
This research provides microlevel evidence that political argument will improve problem solving, even if we are skeptical about human be- ings’ ability to abstract away from their specific circumstances and inter- ests. Neither a commitment to disinterested deliberation, in its strongest sense, nor even standard rationality is required for argument to help solve problems.
This has clear implications for democracy, which forces actors with very different perspectives to engage with each other’s viewpoints, as was pointed out long ago by Lindblom (1965). Even the most homogeneous- seeming societies contain great diversity of opinion and interests (the two are typically related) within them. In a democracy, no single set of interests or perspectives is likely to prevail on its own. Thus, much of the time, po- litical actors have to build coalitions with others holding dissimilar views, and this requires engagement between these views. This need to form coa- litions will also encourage political actors to find options which are broadly rather than narrowly beneficial. Likewise, actors have to publicly contend with others holding opposed perspectives in order to persuade uncommit- ted others to favor their position, rather than another. As new issues arise, actors have to persuade even their old allies of how their shared perspec- tives should be reinterpreted anew.
Uncorrected proofs for review only

Pursuing Cognitive Democracy 225
More generally, many of the features of democracy that skeptical lib- erals deplore are actually of considerable benefit. Mercier and Sperber’s work provides micro-foundations for arguments about the benefits of po- litical contention, such as those of Knight and Johnson and (differently) John Stuart Mill, and for arguments for the benefits of partisanship, such as Rosenblum (2008)’s sympathetic critique and reconstruction of Mill. Mer- cier and Sperber’s findings suggest that the confirmation bias that political advocates are subject to can have crucial benefits, so long as it is tempered by the ability to evaluate good arguments in context.
Other work suggests that the macrostructures of democracies too can aid problem solving. Lazer and Friedman (2007) suggest that networks which isolate individuals (so that they do not converge too quickly on the same solution) do the same kind of work that confirmation bias does in Mercier and Sperber’s studies—theyt preserves diversity and encourage actors to keep exploring solutions that may not have immediate payoffs.13 This underscores the value of having diverse “expressive communities” in the public sphere, to use Danielle Allen’s language.
This work offers a justification for the organization of democratic life around political parties. Party politics tends to organize debate into in- tense clusters of argument among people (partisans for the one or the other party) who agree in broad outline about how to solve problems, but who disagree vigorously about the specifics. Links between these clusters are sparser than links within them, and are usually mediated by competition. One might see each cluster as engaged in exploring the space of possibilities around a particular set of solutions, maintaining some limited awareness of other searches being performed within other clusters, and sometimes discreetly borrowing from them in order to improve competitiveness, but nonetheless preserving an essential level of diversity (cf. Huckfeldt et al. 2004). Such very general considerations do not justify any specific partisan arrangement, as there may be better (or worse) arrangements available. What it does is highlight how party organization and party competition can have benefits that are hard or impossible to match in a less clustered and more homogeneous social setting. Specifically, it shows how partisan ar- rangements can be better at solving complex problems than are nonparti- san institutions, because they better preserve and better harness diversity.
This also helps us think more clearly about the possibility conditions for highly successful problem solving in democracies. We summarize these here in the most cursory fashion, and hope to expand on this in further work. First, in contrast to existing epistemic accounts, cognitive accounts
Uncorrected proofs for review only

226 Chapter Nine
suggest that individuals need to be able to expose their different points of view to each other, rather than being polled in strict isolation from each other, as required by Condorcet’s Jury Theorem. This goes hand in hand with a different account of problem solving: rather than asking if people can determine whether a given decision will be correct or incorrect, as Condorcet does, it asks when individuals will be able to discover hitherto unperceived solutions within a complex landscape. This connects to the question raised by Parham and Allen about the conditions in which people can successfully interpret “messages from some other position.” Second, individuals need to be at least “weak learners” in the terms of statistical learning theory (Schapire and Freund 2012). Individuals who are funda- mentally obtuse, profoundly blinded by ideology, or whimsically perverse will detract from collective learning rather than help it. Third and related, as Mercier and Sperber suggest, cognitive democracy requires that indi- viduals participating in democratic argument have some core commitment to the truth, even if they disagree strongly about what the truth is. People need not be as disinterested as Gardner (chapter 10 in this volume) would like them to be, but neither should they be so warped by self-interest that they cannot see the truth, or allow themselves to care for it. Fourth, even if people disagree on how to solve a problem, they agree on what the problems are that need to be solved in the first place, and have some minimal common empirical standards (see also McAfee, chapter 12 in this volume).
Clearly, these conditions are falsified in our everyday political experi- ence. When the conditions are not met, cognitive democracy will work less well. In the extreme, it may not work at all. Even so, where actual con- ditions even vaguely approximate the ideal, we contend that democracy will be better able to solve complex problems than either markets or hi- erarchy. First, democracy embodies a commitment to political equality that the other two macro-institutions do not. Clearly, actual democracies achieve political equality more or less imperfectly. Yet if we are right, the better a democracy is at achieving political equality, the better it will be, ceteris paribus, at solving complex problems. Second, democratic argu- ment, which people use either to ally with or to attack those with other points of view, is better suited to exposing different perspectives to each other, and hence capturing the benefits of diversity, than either markets or hierarchies. Notably, we do not make heroic claims about people’s abil- ity to deliberate in some context that is free from faction and self-interest. Instead, even under realistic accounts of how people argue, and some min- imal degree of commitment to the truth, democratic argument will have
Uncorrected proofs for review only

Pursuing Cognitive Democracy 227
cognitive benefits, and indeed can transform private vices (confirmation bias) into public virtues (the preservation of cognitive diversity).14 Demo- cratic structures—such as political parties—that are often deplored turn out to have important cognitive advantages.
DEMOCRATIC EXPERIMENTALISM AND THE INTERNET
We have no reason to think that actually-existing democratic structures are as good as they could be, or even close. If nothing else, the designing of institutions is itself a highly complex problem, where even the most able decision makers have little ability to foresee the consequences of their actions. Even when an institution works well at one time, it does so in a context of other institutions and social and physical conditions which are all constantly changing. Institutional design and reform, then, is always a matter of more or less ambitious “piecemeal social experiments,” to use the phrase of Popper (1960). As emphasized by Popper, and independently by Knight and Johnson, one of the strengths of democracy is its ability to make, monitor, and learn from such experiments.15 (Knight and Johnson particularly emphasize the difficulty markets have in this task.) Democra- cies can, in fact, experiment with their own arrangements. As we indicated at the outset, we do not yet know the possibilities of Internet-mediated com- munication for gathering dispersed knowledge, for generating new knowl- edge, for complex problem solving, or for collective decision making, but we really ought to find out.
In fact, we are already starting to find out. People are building systems to accomplish all of these tasks, in narrower or broader domains, for their own reasons. Wikipedia is, of course, a famous example of allowing lots of more or less anonymous people to concentrate dispersed information about an immense range of subjects, and to do so both cheaply and reason- ably reliably.16 Crucially, it is not unique. News-sharing sites like Digg and Reddit are ways of focusing collective attention and filtering vast quanti- ties of information. Sites like StackExchange have become a vital part of programming practice because they encourage the sharing of know-how about programming, with the same system spreading to many other tech- nical domains. The knowledge being aggregated through such systems is not tacit; rather, it is articulated and discursive, but was once dispersed and is now shared. Similar systems are even being used to develop new knowledge. One mode of this is open-source software development, but it is
Uncorrected proofs for review only

228 Chapter Nine
also being used in experiments like the Polymath Project for doing original mathematics collaboratively.17
More modestly, there are the ubiquitous mailing lists and discussion fo- rums on which people with similar interests discuss them. These forums exist for basically all topics of interest to people with enough resources to get online. They are, largely inadvertently, experiments in developing col- lective understandings, or at least shared and structured disagreements, about these topics.
All such systems have to face tricky architectural problems (Shalizi 2008; Farrell and Schwartzberg 2008). They need ways of making find- ings (or claims) accessible, of keeping discussion productive, and so forth. Often, the participants are otherwise strangers to each other, which is at the least suggestive of the problems of trust and motivation that will face efforts to make mass democracy more participative. This opens up an immense space of possible institutional design, which is still very poorly understood—but it almost certainly presents a rugged search landscape, with an immense number of local maxima and no very obvious path to the true peaks. One of the great aspects of the current moment for cognitive democracy is that it has become comparatively cheap and easy for such experiments to be made online, so that this design space can be explored.
Even the online ventures that do not work are informative. They range from poorly designed sites that never attract (or which actively repel) a user base or produce much of value, to online groupings that are very successful on their own terms, but are cognitively full of fail, such as thriving commu- nities dedicated to conspiracy theories. These are not just random, isolated eccentrics, but highly structured networks engaged in sharing and devel- oping very bad ideas. (See, for instance, Bell et al. [2006] on the networks of those who share delusions that their minds are being controlled by outside forces.) If we want to understand what makes successful online institutions work, and perhaps even draw lessons for institutional design more gener- ally, it will help tremendously to contrast the successes with such failures.18
The other way in which this helps learning is that all these experiments are leaving incredibly detailed records. People who use these sites or sys- tems leave detailed, machine-accessible traces of their interactions with each other—even ones which tell us about what they were thinking. This is an unprecedented flood of detail about experiments with collective cogni- tion, and indeed with all kinds of institutions, and about how well they have served various functions. Not only could we begin to just observe successes and failures, but we can probe the mechanisms behind those outcomes.
Uncorrected proofs for review only

Pursuing Cognitive Democracy 229
This points, we think, to a very clear constructive agenda. To exaggerate a little, it is to see whether the Internet enables modern democracies to make as much use of their citizens’ minds as did ancient Athens, according to the clas- sicist and political theorist Josh Ober (2008). We want to learn from existing online ventures in collective cognition and decision making. We want to treat these ventures as more or less spontaneous experiments,19 and compare the success and failures (and partial successes and failures) to learn about institu- tional mechanisms which work well at harnessing the cognitive diversity of large numbers of people who do not know each other well (or at all), and meet under conditions of relative equality, not hierarchy. If this succeeds, what we learn from this will provide the basis for proposing design principles to plat- form designers, for experimenting with redesigning democratic institutions themselves, and for creating developmental contexts in which young people prepare to meet the conditions of successful democratic problem solving.
We have, implicitly, been viewing institutions through the lens of infor- mation processing. To be explicit, the human actions and interactions which instantiate an institution also implement abstract computations (Hutchins 1995). Especially when designing institutions for collective cognition and decision making, it is important to understand them as computational pro- cesses. This brings us to our concluding suggestions about some of the ways in which social science and computer science can help each other.
Hong and Page’s work provides a particularly clear, if stylized, model of how diverse individual perspectives or heuristics can combine for better problem solving.20 This observation is highly familiar in machine learning, where the large and rapidly-growing class of “ensemble methods” work, explicitly, by combining multiple imperfect models, which helps only be- cause the models are different (Domingos 1999; Schapire and Freund 2012). In some cases it helps exactly to the extent that the models are different (Krogh and Vedelsby 1995). Different ensemble techniques correspond to different assumptions about the capacities of individual learners and about how to combine or communicate their predictions. The latter are typically extremely simplistic, and understanding the possibilities of non-trivial or- ganizations for learning seems like a crucial question for machine learning, for social science, and for the cultivation of effective citizenship.
CONCLUSIONS: COGNITIVE DEMOCRACY
Democracy, we have argued, has a capacity unmatched among other macro- institutions to actually experiment and make use of cognitive diversity in
Uncorrected proofs for review only

230 Chapter Nine
solving complex problems. To realize these potentials, democratic struc- tures must themselves be shaped so that social interaction and cognitive function reinforce each other. Also, participants need to be prepared to expose their different points of view to each other; they need to be at least “weak learners”; and they need some core commitment to the truth, even if they disagree strongly about what the truth is. Participants need not be perfectly disinterested, but neither should they be so warped by self-interest that they cannot see the truth, or allow themselves to care for it; and they do need some shared ideas about what problems need to be solved as well as sharing some minimal common empirical standards. In- deed, the cleverest institutional design in the world will not help unless the resources—social and cultural, but material too—needed for participation are actually broadly shared. This is not, or not just, about being nice or equitable; cognitive diversity is not something we can afford to waste.
We differ from Gardner’s proposal (in this volume) that we should seek to nurture disinterested professional standards on the Internet and elsewhere. Our perspective highlights the value, up to a point, of interestedness—that is of vigorous and even rowdy debate between individuals who have different perspectives and different interests, but roughly equal ability to make their voices heard. Where Gardner wants professionalism, we look to a kind of rambunctious amateurism. That said, there are some points of congruence between our perspectives. As we argue above, if democracy is to thrive, it requires some level of agreement about the problems it confronts (even if it disagrees about how best to solve these problems). This agreement may be difficult to generate within democracy itself. We may need to turn, for instance, to scientific advice, ideally produced by disinterested and nonpo- litical experts, to generate this agreement.
Our views have consequences for how we regard the political effects of the Internet (Farrell 2012). In some ways we agree with, for instance, Kahne, Middaugh, and Allen (chapter 2 in this volume), who see the Inter- net as enabling more participatory politics. However, while we see the In- ternet as a space for experiments and learning, we think that the failures of collective cognition online are as important as the successes. We doubt that the Internet is inherently liberatory; what matters most for our purposes is that it makes collective relationships that would otherwise be opaque more transparent, and hence easier to experiment with and to learn from.
More broadly, we emphasize that both Internet and non-Internet forms of collective cognition will be able to take full advantage of diversity only in conditions of political equality. Without such equality, some voices will
Uncorrected proofs for review only

Pursuing Cognitive Democracy 231
dominate, and others will be diminished or silenced. Like a number of other contributors to this volume, we do not think that the conditions of equal- ity are satisfied by standard arguments for deliberation—argument can and will be driven by self-interest, ideally in its equitable form, and it will take a wide variety of forms. We cannot specify in advance the particular conditions under which equality can best be achieved—they can only be discovered through experimentation. However, the commitment to equal- ity is crucial. Much techno-utopianism suggests, directly or indirectly, that new forms of technologically mediated communication can serve as a sub- stitute for more traditional forms of political (and economic) inequality. Our claim is quite different: that the conditions under which technological utopianism will work properly, by providing valuable experimental input into politics, are conditions of equality. These may or may not be made eas- ier to achieve through technological advances, but they are surely not an inevitable by-product of technological change. This implies that a genuine commitment to democracy and the liberatory potential of technology is a commitment to political radicalism. We embrace this.
}}