The Knowledge/Power Separation Problem
On One of the Predicaments of Modernity
Disclaimer: Despite the appearances, there is no Michel Foucault in the following essay (quite the contrary actually)!
Let’s consider a fictional place called “Wisdom Town.” Wisdom Town is a small town, or maybe a village of a few dozen individuals at most. There is a very low level of division of labor, no real scientific knowledge is possessed and used, and inhabitants share the overarching goal of maintaining peace and fulfilling basic needs. The governance of this town is based on a very simple principle. The older you are the more you’re assumed to know about how people should live in Wisdom Town and the more political power you’re granted. In practice, there is a “council of the Wise” that people above a certain age join and that makes most of the collective choices that concern all the inhabitants of the village. Within the council, the same principle formally applies – the older you are, the more power you have – though the actual hierarchy may be less rigid than that. Because the division of labor is limited and much of the knowledge that is possessed is related to ancestral traditions that individuals progressively learn about as they get older, Wisdom Town essentially conflates knowledge and power. Power is given to those who know more and knowledge is the sole justification for allocating political power.
Wisdom Town is a simple (political) model that most likely doesn’t correspond to any human society that has existed ever. Still, it captures an important property that is constitutive of any such society, i.e., how knowledge and political power are related. In Wisdom Town, the two are merged. Knowledge gives both de facto and de jure political power. This is made possible by the assumptions I’ve sketched, especially the fact that there is almost no division of labor. Now, it is obvious that modern societies do not exhibit such a simple and one-way relationship between knowledge and political power. Knowledge may be a source of power, both de jure and especially de facto, but it is not the only one. More importantly, the fact of being knowledgeable tends to not be viewed as a strong enough reason to grant political power to someone. In modern societies, knowledge and power are kept separate. Let’s call that the Knowledge/Power Separation Problem.
To understand why I call the separation of knowledge and power a “problem,” we should start by considering the kind of knowledge we’re talking about. In societies with a high degree of division of labor and specialization, there are at least two forms of knowledge that are relevant. The first is specialized scientific knowledge, the second is cultural practical knowledge. The former is mostly formal and results from the systematic use of scientific methods to increase our control over nature and society. The latter is essentially informal and tacit and is encapsulated in practices that have evolved through a complex discovery and evolutionary process. Specialized scientific knowledge is thought to be truth-tracking and is amenable to accumulation and growth. It is mostly general. Cultural practical knowledge is more “local” and of the “know-how” kind. It is constitutive of social practices in which members of the population are participating.
Note that the distinction is irrelevant in Wisdom Town because I’ve stipulated that knowledge there is essentially cultural-practical. The emergence of specialized scientific knowledge creates however a difficulty because there is no reason to think that both kinds of knowledge must be convergent. Indeed, there are good reasons to think that very often they will starkly conflict. This is not surprising as specialized scientific knowledge is cumulative and truth-tracking while cultural practical knowledge is local and about social practices. By definition, it cannot be acquired without being a member of the community participating in the practice. While sociology or anthropology aims to produce scientific knowledge about social practices, it is not really a substitute for the practical knowledge constitutive of the corresponding practices. Also, and crucially, both kinds of knowledge have a different kind of “value-relevance.” Truth, or at least the ability to control the world, is what gives scientific knowledge its value. The value of practical knowledge – what makes it valuable – is attached to some conception of the good life and more generally to a social form that may be unrelated to the pursuit of truth or control.
The existence of two different and potentially conflicting forms of knowledge implies that knowledge and power cannot be conflated anymore. At best, only one of those two forms of knowledge can ground political power. Reasoning in terms of ideal types, what can be called a “technocracy” is a regime where power is entirely grounded on specialized scientific knowledge. Wisdom Towns corresponds to a hypothetical society where power is entirely grounded on practical knowledge. In terms of modern political regimes, a fully direct democracy (or maybe even an anarchy) is what comes the closest to the case where power is mostly grounded in practical knowledge. None of the alternatives is attractive. To ground political power on specialized scientific knowledge is a delusion that has been pointed out multiple times. While it is a fact that humans have achieved incredible control over their environment, this control remains nonetheless limited. The problem is at least twofold. First, it lies in the fact that scientific knowledge is unavoidably fallible and bounded. There are things that we thought we knew and realized later that we were just wrong about. Also, to be complete, scientific knowledge should incorporate cultural practical knowledge, but as I have pointed out, this is impossible. The result is the spectacular failure (for instance) of the kind of ambitious architectural or agricultural projects well-documented by James Scott in his classic Seeing Like a State.[1] Second, specialized scientific knowledge tends to be oblivious to the plurality of values (and the potential conflict between them thereof) that is constitutive of social practices. On the one hand, scientific considerations will very often underdetermine political decision-making just because even considering a set of true propositions about the world, what ought to be done depends on value judgments that cannot be assessed by scientific methods. On the other hand, and as a logical implication of the previous point, if one tries to impose political decisions only based on scientific knowledge, that implies that one is disregarding the value judgments of a fraction of the population. In this case, we oscillate between authoritarianism (if the attempt succeeds) and plain failure (because people react in such a way that the attempt at imposing some values is defeated).
Now, disregarding specialized scientific knowledge is not an option either. There are multiple indications that well-established social practices based on cultural practical knowledge can turn out to have adverse effects on the populations participating in them.[2] Downplaying specialized scientific knowledge is also a well-entrenched habit of populist politicians who argue that political legitimacy and authority are entirely encapsulated in the people’s general will. A characteristic feature of populist politics is indeed to rule out all forms of “rational-legal” authority that, by definition, is not tied to the leader (who is impersonating the general will) or the tradition. Science, along with the constitutional and legal framework, figure among the source of authority and thus of political power that populists reject. The idea that scientists, experts, and judges can have even a modicum of political power is anathema to populist politics.
Since we are left with the conclusion that neither specialized scientific knowledge nor cultural practical knowledge can fully ground political power, the relationship between power and knowledge cannot be simple and straightforward. Contemporary liberal democracies reflect the kind of intricate solutions that have been progressively worked out to tackle the knowledge/power separation problem. They have an indisputable technocratic component. This is reflected in the fact that many political decisions are outside of the scope of proximate democratic control and are largely grounded in the specialized knowledge produced by scientists and diverse kinds of experts (including private consulting groups for instance). On the other hand, there are signs of a growing awareness that the use of specialized knowledge for political decision-making is radically value-laden and that this is relevant for political practice. The emergence of new democratic tools such as the “conventions citoyennes” that have been used twice recently in France can be interpreted in this perspective as innovative attempts to deal with the knowledge/power separation problem.
All this is also relevant to philosophy and the use of (social) sciences in the public space. Eric Schliesser recently commented on the short book of Eric Winsberg and Stephanie Harvard, Scientific Models and Decision-Making.[3] As Eric points out, the book’s chapter discussing Ferguson’s ICL model considers how policy-relevant scientific models should account for the values that prevail within the population that will be affected by the political decisions eventually based on these models. The view that Winsberg and Harvard defend (i.e., that the models should essentially appeal to publicly held values) is part of a more global approach to the knowledge/power separation problem. In essence, the idea is that scientific modeling (and thus scientific knowledge) should incorporate as much as possible the values that are constitutive of the social practices in which the members of a population participate. However, as Eric notes, this is not so straightforward once it is realized that people disagree about values – something that presumably is unlikely to happen in Wisdom Town.
Winsberg and Harvard’s book also has a chapter focused on the public role of scientists and experts and on how they can (mis-)use their models and the knowledge they produce.[4] The discussion illustrates another aspect of the knowledge/power separation problem related to the way specialized scientific knowledge can enter the public sphere as a justificatory resource for political decision-making. This is an issue that Eric discusses in another recent essay where he notes that “how to select and how to know which specific intellectual or influential to listen to, and then, how to turn their ‘warnings’ into collective agency remain to be resolved.” Both Winsberg and Harvard’s and Eric’s discussions somehow echo my point about the political ethic that scientists-turned-public-intellectuals should display.
Ultimately, in liberal democracies, while they are separated, power and knowledge must still be articulated in the public sphere. The fundamental problem is to publicly justify and legitimize political decisions by making them sensitive to the plurality of values and practices while at the same time grounding them in specialized scientific knowledge. There is (or there should not be) no political power without public justification, and public justification requires forming and using the appropriate kind of “synthetic knowledge” (to borrow Eric’s expression). The nature of this synthetic knowledge and how to achieve it remains a largely open issue.
[1] James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven, CT London: Yale University Press, 1999).
[2] My favorite reference is Cristina Bicchieri, Norms in the Wild: How to Diagnose, Measure, and Change Social Norms (Oxford University Press, 2016). Bicchieri also shows how perilous it is to try to change people’s social practices without a proper understanding of the underlying values and motivations, as well as of the social dynamics that sustain them.
[3] Eric Winsberg and Stephanie Harvard, Scientific Models and Decision Making (Cambridge University Press, 2024).
[4] They target a NYT’s op-ed and a tweet from the evolutionary biologist Carl Bergstrom.