The flow of evidence is pervasively impacted by social dynamics. One straightforward illustration of this point is that individuals routinely give evidence to one another. If I want you to believe that my dog’s name is Buffy, I might simply tell you her name or I might show you her passport. But if for some reason I don’t want you to know my dog’s name, I might withhold this evidence from you, and might even discourage or prevent others from giving you the relevant evidence.
The ability to control the flow of evidence is a substantial sort of power, one that enables its possessors to influence what others believe and what others can rationalize. Notably, this ability, and hence the sort of power it implies, is not evenly distributed—some have a far greater degree of control over the flow of evidence than others. I use the term epistemic domination to refer to imbalances in the ability to control the evidence available to the other party.
In general, epistemic domination occurs when one party can control the evidence available to another party, but not vice-versa. So understood, epistemic domination can come in degrees. Parents typically stand in a strong position of epistemic domination over their children, but their capacity to control their children’s evidence typically wanes over time. Prominent media figures exert a degree of epistemic domination over audiences, but this domination is substantially mitigated by the high-choice nature of the contemporary media environment. In authoritarian systems, the state stands in a relatively strong position of epistemic domination over subjects. Notably, authoritarian states’ capacities to control the evidence available to subjects consists not only in the ability to force evidence onto subjects, but also in the ability to restrict subjects’ access to alternative sources of information.
As I understand the concept, one party can epistemically dominate another even if the former does not in fact take measures to control the evidence available to the other. It is the capacity to control the evidence available to others, rather than the exercise of this capacity, that is necessary for epistemic domination.
Taken by itself, epistemic domination, even when exercised, does not imply any ill-intent. Indeed, epistemic domination is regularly exercised with both positive motivations and beneficial effects. For example, as noted above, parents may deliberately withhold low-quality information from children, while promoting their access to higher-quality information. Much the same is true of teachers.
But epistemic domination can be both attained and practiced with ill-intent, and with severe epistemic consequences. I’ll mention just some of these here. First, parties in positions of epistemic domination may influence what others believe, or fail to believe, for the worse. In an extreme case, an authoritarian government might flood the airwaves with propagandistic messaging, causing subjects to believe various falsehoods. That same authoritarian government might withhold access to valuable evidence, and might deter others from sharing it. By restricting the flow of evidence in this way, authoritarian governments might prevent subjects from forming true beliefs based on that evidence. Restrictions on speech are hence stark examples of the exercise of epistemic domination. Notably, such restrictions also make relations of epistemic domination visible and hence may promote suspicions on the part of dominated parties. I return to this point below.
Even if exercises of epistemic domination do not cause those subject to them to form or sustain false beliefs or to fail to form true beliefs, such exercises might still have ill-effects at the level of epistemic justification. Suppose a particular person believes some scientific proposition, and that a party in a position of epistemic domination floods her epistemic environment with misleading counter-evidence against that proposition. Even if the person is not swayed by that counter-evidence, her justification for believing the proposition might still be compromised.
In extreme cases, including some cults and authoritarian systems, there is one party in a position of epistemic domination with respect to others. This party may be expected to push evidence favorable to itself onto subjects, while suppressing evidence unfavorable to itself. The expected effect of such unipolar epistemic domination is the promotion of certain beliefs and the suppression of others. But, in many real-world cases, different parties with different interests compete for control over the flow of evidence. In such cases, rather than being given a consistent line of evidence, those subjected to epistemic domination may find themselves whiplashed by evidence pointing in different directions. Rather than finding themselves strongly pushed toward some beliefs and away from others, those in this situation may experience disorientation—a state in which they are at a loss as to who to trust and what to believe. Notably, although such a state of disorientation might arise as a side effect of competition for epistemic domination between parties, many have noted that some contemporary propaganda strategies aim at producing this effect.
One reason contemporary propaganda strategies sometimes aim at disorientation, rather than the promotion of specific beliefs, is that the modern social and technological environment makes it difficult to exert thoroughgoing epistemic domination. Even those subjected to state control of online content can, for example, often elude such control by way of virtual private networks. Relatedly, especially heavy-handed exercises of epistemic domination tend to be both obvious and suspicious, which in turn drives subjects to actively pursue alternative information. This dynamic leads some parties in positions of epistemic domination to eschew the most heavy-handed forms of evidence control in favor of techniques like porous censorship, which limits the influence of unwelcome evidence through inconvenience and distraction, rather than outright suppression.
The struggle in authoritarian systems to exert control over evidence without making this obvious, and hence inviting suspicion, points at struggles faced by other parties, including some more sympathetic ones. A degree of epistemic domination is all but inevitable. Researchers and others occupying roles in various specialized fields effectively control much of the evidence available to the broader public about issues within these fields. This imbalance often leads to resentment and suspicion. As an example that is very current as of the time of this writing, consider how members of both major American political parties have expressed concerns about manipulation of documents related to Jeffrey Epstein. Consider also how conspiracists and other science skeptics often premise their skepticism on the suggestion that scientists might manipulate data for their own gain. For example, skepticism about climate change is often motivated by reference to “Climategate” and related conspiracy theories.
Attention to the ways in which the flow of evidence is controlled can be beneficial, dampening the impact of propagandistic efforts by agents in positions of domination. But concerns about epistemic domination can tip easily into paranoia. Some forms of science skepticism, in particular, reflect an oversimplified picture of scientific institutions as monoliths that have unified interests and that manipulate evidence to promote these interests. Insofar as science skepticism is predicated on such concerns, efforts to make science more transparent and to emphasize the divergent interests of individual scientists can be expected to reduce such skepticism. In effect, such efforts can serve to reduce the perception that scientific institutions are tools for epistemic domination.
Keith Harris
Dr. Keith Raymond Harris is a postdoctoral fellow at the University of Vienna, where he is part of the Knowledge in Crisis project. He is the author of Misinformation, Content Moderation, and Epistemology: Protecting Knowledge. His recent articles include “Higher-order misinformation,” “Synthetic media, the wheel, and the burden of proof,” and “Smoke machines: Emotive artificial intelligence and deceptive signaling.”
