Jesper Kallestrup argues that groups can have knowledge that their members may not

What, if anything, do groups of people know? Not as mere collections of individuals, each one knowing something in his or her own right, but as groups? Can we talk about groups knowing something, and if so, do groups have knowledge over and above whatever knowledge their members have? Most accounts of what it takes to know something traditionally assume that only individuals are capable of having knowledge. But the emergence of social epistemology has changed this.

Before we turn to group knowledge, first we need to consider the question of what makes a collection of individuals a group. First, groups, as I use the term here, have organizational structures by which individual members are assigned particular roles with accompanying job descriptions. Second, groups are such that their (founding) members jointly set up common goals or aims, and agree on how to proceed in order to meet them. Both the ends, and the means by which they are achieved, are captured by a charter. This charter can be broadly understood to include formally enshrined laws or regulations, as well as systems of rules, principles, or norms to which the group implicitly adheres, as evidenced by its behaviour. Third, groups are characterized by joint intentions, namely, individual intentions joined together when shared by their members. Each member intends that together with every other member, they enact a joint performance and come to a group attitude. For example, the players on a sports team jointly intend, as a group, to win a cup final, and the content of that intention involves the players performing some action together, for example, play a certain attacking formation.

Contrast, for example, a sports team or a university committee, on the one hand, with the inhabitants of Edinburgh or Syrian asylum seekers, on the other. The three abovementioned constraints are only satisfied by the sports team and the university committee, and thus only they form groups. When these constraints are met, a collection of individuals unites in forming what is called a ‘rational agent’ in its own right. These constraints are the glue that joins individuals together as a single agent with a cohesive mind of its own. True, the inhabitants of Edinburgh or Syrian asylum seekers may share mutual knowledge—they may all know something to be true—and common knowledge—not only does each person know something, each person knows everyone else in the group knows it too. But though the inhabitants of Edinburgh or Syrian asylum seekers are collections of individuals who have certain features in common, they fail to satisfy the three constraints: each individual does not have a specific role to play within the group, they do not share common goals, and they do not share a common intention. Thus, they do not constitute a group in our sense, and therefore group knowledge is not attributable to them.

Equipped with this conception of groups, our first question is: why think groups can enjoy knowledge at all? An important observation concerns ordinary parlance. As any quick Google search will verify, we speak unproblematically about clubs, boards, firms, churches, organizations, armies, governments, and so on as having knowledge. The ubiquity and diversity of such talk presents a reason to take its content at face value. Were we invariably mistaken about in this sort of talk, a drastic revision of how we talk about knowing would be called for. Of course, we also say ‘the flowers know when to blossom’, or ‘my computer knows me better than my friends’, but such talk is loose or metaphorical. Flowers and computers are neither responsible for what they do, nor do they deserve praise for getting things right. In contrast, we ordinarily assign credit to groups when they overcome obstacles to achieve remarkable knowledge, and we hold them to account for failing to act on a body of knowledge they possess. In fact, not only do we ascribe to groups the kind of low-grade knowledge that certain animals or young children are often thought to possess, we also ascribe to groups fairly sophisticated knowledge, such as being able to engage in reflection upon their own knowledge, as when groups scrutinize the credentials of their evidence or draw complex inferences between distinct pieces of knowledge.

If we accept that groups can acquire knowledge, the second question is whether such knowledge is anything more than just the knowledge of their members: does a group have knowledge if and only if some (or most, or all) of its individual members have that knowledge? Let’s review two examples that suggest group knowledge and individual knowledge can diverge. The first example involves criminal proceedings in a court of law, for which the standards of evidence include that hearsay normally be excluded and that the standard of proof be beyond reasonable doubt. These special standards govern how the criminal court must operate, but they need not be applicable to any individual court member. Thus, Frederick Schmitt imagines cases of the following kind:

A defendant is on trial for the crime of careless driving. The prosecution adduces evidence from the police report, as well as eyewitnesses testifying in court that the defendant was indeed driving the van that hit the victim. But the jury finds the evidence not beyond a reasonable doubt, and hence insufficient to validate criminal conviction. All the members of the jury have hearsay evidence from a reliable source that the defendant caused the accident. But the judge instructs the jurors to ignore this evidence as it fails to meet the conditions for being ruled admissible in a criminal court.

Since the evidence available to the jurors individually is unavailable to them jointly as a jury, this example shows that a group can lack knowledge, even when that knowledge is possessed by all its members. The individual jury members have reliable evidence, sufficient for them to know that the defendant is guilty; but the jury, when functioning in accordance with the appropriate standards, lacks reliable evidence sufficient for it to know that the defendant is guilty.

The second example exploits the ubiquitous phenomenon of distributed cognition, which involves a division of cognitive labour into sub-tasks assigned to individuals in the group with the relevant expertise. Scientific research groups typically involve distributed cognition on a huge scale; for example, CERN involves collaboration between physicists, engineers and other researchers from dozens of institutions and countries. The following example, inspired by Alexander Bird, helps fix ideas:

Imagine a physicist and a mathematician, collaborating on a project to demonstrate the existence of parallel universes. The physicist builds models, conducts experiments, and analyses data to confirm an interpretation of string theory, while the mathematician provides a mathematical proof that such universes exist if the physicist’s interpretation is correct. The physicist and the mathematician do not communicate about their respective tasks. Instead, they independently pass on their results to an assistant who simply completes a pre-written draft, and then publishes their joint paper showing the existence of parallel universes.

The research team thus comes to know that parallel universes exist, yet neither of its members, the physicist nor the mathematician, has that knowledge, since neither is aware of what the other has achieved. Nor does the assistant, whom we assume lacks sufficient expertise to even grasp string theory or the relevant maths. This example shows that a group can have knowledge that none of its members have, where such knowledge fulfils the aim or purpose of the group.

In sum, the examples of the criminal court and the research team together show that the possession of knowledge by all (or a majority, or some) members of a group is neither necessary nor sufficient for the group to have that knowledge. Consequently, knowledge is not the preserve only of the individual, but can be held by groups as they have been understood here. This raises many other intriguing questions. Do groups and individuals acquire and store knowledge in different ways? Can groups genuinely act on their knowledge in ways that cannot be reduced to their members’ behaviour? Much work is still to be done here.

Jesper Kallestrup is professor of philosophy at the University of Edinburgh. This essay is based on his paper ‘Group Virtue Epistemology, forthcoming in Synthese. His research interests are primarily at the intersection of philosophy of mind and epistemology.

Image credit: T. A. Vyacheslavovich, ‘Grand Orchestra’