Technocracy as a Form of Life

On Seeing Everything as a Problem to Be Solved

When you think about the reasons that may justify the adoption of a democratic regime rather than any other alternative system, most of us will think of the value of political equality, the respect and dignity of persons viewed as moral equals, making sure that everyone’s interests are taken into account in collective-decision making, leaving the possibility of getting rid of incompetent/corrupt political leaders open, favoring the competition between a plurality of views, and probably a few others. While all of them have been largely explored by political philosophers, there is a significant and interesting trend in what is sometimes called the “deliberative democracy” research program that tends to emphasize a different and less intuitive reason. More and more, deliberative democrats are insisting that the justification of democracy lies in the fact that it is the best political problem-solving procedure at our disposal. In this post, I’ll explain a bit more what that means and then reflect on the meaning of this view as part of what could be called the technocratic form of life.

Even though it is surely not new, the idea that the justification of democracy must be somehow related to its epistemic, truth-tracking properties has been particularly well exposed by David Estlund in his book Democratic Authority.[1] This book is part of what Hélène Landemore has characterized as the “epistemic turn” of deliberative democracy.[2] As Landemore notes, the kind of “epistemic abstinence” that is characteristic of Rawlsian political liberalism where it is argued that the truth-value of moral views cannot be a (public) reason to justify coercive institutions implementing them is less and less accepted, even by “Rawlsians democrats” (Estlund being one of them).[3] I shall not rehearse here the motivations for excluding truth from the domain of public reason that Rawls and others have developed, the point that matters is that this is no longer the received view among political philosophers and especially deliberative democrats.

The retreat from epistemic abstinence has however significant implications. Once the justificatory relevance of truth is acknowledged, deliberative democrats face a potential danger: what if it happened that a democratic regime is not the best procedure to track the truth? In particular, what if, as Jason Brennan and others argue,[4] something like an epistocratic regime happens to be epistemically superior? There are obviously two possibilities. Either you go back to traditional justifications of democracy based on political equality or other non-epistemic considerations. Or you provide an argument for the claim that democracy is epistemically superior to any other alternative regimes. This is the road taken by many deliberative democrats such as Landemore and Estlund. A popular approach in this perspective has been to appeal to formal (mathematical) models or theorems whose results suggest that a democratic regime is more likely to lead to the correct (collective) decision than any other political regime. Three such models are cited again and again in the literature:[5]    

·       Condorcet’s jury theorem. Suppose a population of n agents who have to form a binary judgment (e.g., true or false) on some issue I. We assume that each of them is more likely to form the correct judgment than not (competence hypothesis) and that their judgments are not statistically correlated (independence hypothesis). Then, the higher n is, the more likely the majority judgment in the population will be correct.

·       The miracle of aggregation. Consider a variable x defined over some interval and whose true value is x*. Each individual i in a population of n agents forms a judgment x_i over x’s true value and a collective judgment is formed by looking at which individual judgment is majoritarian. Suppose that a subset of individuals knows the true value of x while others form their judgment randomly according to a uniform distribution across the interval. Then, as n increases, the probability that errors ‘cancel each other out’ increases, allowing the subset of knowledgeable individuals (even if very small) to decisively shift the balance in favor of the correct answer.

·       The “diversity trumps ability” theorem.[6] Consider a population n of “problem-solving agents” addressing a problem P. To tackle P, each agent uses some heuristics h. Some heuristics are better than others, i.e., they are more likely to solve P. A problem-solving team composed of randomly selected agents is more likely to solve P than a uniform team composed of agents who use the same heuristics, even if it is the best. The probability that this happens increases with n.

These results suggest that an increase in the size and/or diversity of the population of “problem-solvers” makes it more likely that the majority judgment will be correct. Or, more simply put: number and diversity are more relevant than competence and expertise to track the truth. If we consider that a democratic regime is, compared to any other alternative regime, the one with the largest and most diverse electorate, several deliberative democrats go a step further and argue that democracy is epistemically superior to any other plausible political regime.

I must admit my long-standing puzzlement at the face of this “jump” from the results of formal theorems to a strong (epistemic and, by implication, political) conclusion. At the most general level, we all know (or should) that the inference from a formal result to a normative conclusion is conditional on many additional assumptions and premises. Obviously, these theorems warrant the conclusion that democracy is epistemically the best regime only if we have good reasons to think that the comparison between political regimes falls into the domain of validity of those theorems. In other words, do we have good reasons to think that the conditions under which these theorems are valid apply when we compare democracy with, say, epistocracy. There is also the more complicated issue of whether the way the interpretation of the theorems is true to the mathematical result. As it has been pointed out multiple times, including by some deliberative democrats,[7] we may rightfully doubt that the three aforementioned theorems say anything about the epistemic value of democracy. Regarding the first two theorems, it is just plainly obvious that the conditions for their validity hardly apply to real-world democracies. With respect to the third theorem, its interpretation in terms of the epistemic value of diversity has been challenged and pointed out as an example of “the misuse of mathematics in the social sciences.”[8]  

Now, and to finally get to my main point, whatever you think about these arcane debates about the philosophical and political relevance of mathematical theorems, I think the most problematic aspect is that this approach frames democracy and political regimes, in general, uniquely in terms of problem-solving procedures. Basically, the big assumption lurking behind all this sophisticated discussion is that there is a correct answer to all political issues and that the justification of political institutions entirely rests on their ability to find it. While this is a natural assumption, it is far from being obvious – witness the fact that deliberative democracy has not relied on this premise until recently.

We could discuss the relevance of this assumption but my interest is elsewhere. The more relevant question is why so many people are prone to making this assumption. I think indeed this more generally betrays the pervasiveness of what could be called the technocratic form of life in contemporary democracies. The concept of technocracy is generally used to refer to a political regime where collective decision-making is essentially left in the hands of technicians or experts charged with finding the best means to reach given ends. However, following the late Jeffrey Friedman in his book Power Without Knowledge,[9] I’ve come to realize that technocracy is more than a set of formal rules allocating political power. Technocracy is more generally a mindset and a practice consisting of seeing every issue in terms of the means/ends framework and where the only real problem is to find the best means to achieve given ends, i.e., to solve a problem for which it is assumed everyone agrees on what the solution should be.

Indeed, and rather counterintuitively, Friedman argues that in contemporary democracies the technocrats are not only the experts or specialists with the relevant credentials (whom he labels the “epistocrats”) but also the citizens themselves. Indeed, according to him

“[citizens’s] political decisions are heavily influenced by perceptions of whether or not public policies can be expected to ‘work,’ or are already ‘working,’ in a vague but real sense; that is, a vague sense of whether they solve, mitigate, or prevent social and economic problems – and in the process, whether they do more good than harm.”[10]       

This makes experts and citizens alike sophisticated or casual cost-benefit reasoners who frame every issue in terms of a problem for which the best or at least a satisfactory solution must be found. This mindset is constitutive of what Friedman calls a democratic technocracy, defined not as a political regime but rather as the conjunction of a set of formal rules to allocate political power and a set of frames, forms of reasoning, and practices that lead to political decisions. Friedman convincingly shows that the epistocratic critique of democracy makes sense only in the context of the democratic form of life. I would add, based on what I’ve written above, that the same applies to the epistemic turn of deliberative democracy.

In his book, Friedman goes on to attack democratic technocracy on the ground that its actors (experts and citizens) display “naïve technocratic realism,” i.e., a tendency to assume that there exist self-evident technocratic truths. This tendency is also shared by the “political epistemology” tacitly or explicitly endorsed by political philosophers and political theorists who are assessing political regimes in terms of their truth-tracking properties. The problem is that we do not and cannot have the knowledge to discover these truths (which are therefore not self-evident) nor any reason to think that they must exist. The problem with the democratic technocracy as a form of life is that all its actors (the epistocrats, the citizens-technocrats, and the technocratic political philosophers and theorists) behave as if they had the relevant kind of knowledge while they are actually “radically ignorant,” i.e., they are unaware of the fact that they don’t have the relevant knowledge or information to behave as problem-solvers.

I’m far from agreeing with everything in Friedman’s critique of democratic epistocracy that at times borders with radical skepticism if not nihilism regarding the possibility of technocratic knowledge. If we follow him till the end, then the distinction for instance between specialized scientific knowledge (that is indeed essentially technocratic) and cultural practical knowledge (that may be technocratic in Friedman’s sense without reducing to it) that I used in my previous essay becomes meaningless. That would be going too far. I fully agree that our political knowledge is fundamentally limited but this is not because technocratic knowledge is an illusion. In so far as the kind of technocratic knowledge natural science is producing has resulted in a considerable increase in our control of nature, it is an indisputable reality that has permitted the spectacular improvement of living conditions over the last couple of centuries. Matters are surely more controversial with social sciences but here too, there are indications that something akin to Popperian “piecemeal engineering” is possible. As always with relativistic or skeptical views, if you take them too seriously you end up at best with a form of epistemological anarchism where forms of knowledge cannot be hierarchized, at worst with plain contradictions. If the former, then one should ask if we should continue to publicly fund scientific research, if the latter, then why should we even take the time to listen to someone claiming that there is no knowledge anyway?

Admittedly, Friedman does not say that we have no ground to believe anything, only that we don’t have the kind of technocratic knowledge we think we have and that we need to behave as technocrats. Independently of one’s view about this, another interesting question is whether we can escape the technocratic form of life and how. Friedman also has things to say about this, and that will surely be the topic for a future essay.            

---

[1] David Estlund, Democratic Authority: A Philosophical Framework (Princeton, N.J: Princeton University Press, 2007).

[2] Hélène Landemore, “Beyond the Fact of Disagreement? The Epistemic Turn in Deliberative Democracy,” Social Epistemology, 2017.

[3] I’ve discussed about political liberalism’s epistemic abstinence several times here, see this essay for instance.

[4] Jason Brennan, Against Democracy (Princeton (N.J.): Princeton University Press, 2016). See also, for instance, Garett Jones, 10% Less Democracy: Why You Should Trust Elites a Little More and the Masses a Little Less, (Stanford University Press, 2020). I’ve made a (very) modest contribution in this debate: Cyril Hédoin, “The ‘Epistemic Critique’ of Epistocracy and Its Inadequacy,” Social Epistemology 35, 2021: 502–14.

[5] E.g., Hélène Landemore, Open Democracy: Reinventing Popular Rule for the Twenty-First Century (Princeton: Princeton University Press, 2022).; Robert E. Goodin and Kai Spiekermann, An Epistemic Theory of Democracy, (OUP Oxford, 2018).

[6] Lu Hong and Scott E. Page, “Groups of Diverse Problem Solvers Can Outperform Groups of High-Ability Problem Solvers,” Proceedings of the National Academy of Sciences 101, no. 46 (2004): 16385–89.

[7] Estlund is for instance dismissive of the relevance of Condorcet’s jury theorem to make the case for the epistemic superiority of democracy. See Estlund, Democratic Authority.

[8] Abigail Thompson, “Does diversity trump ability?”. Notices of the AMS, 61(9), 2014, 1024-1030. But see Daniel Kuehn, “Diversity, Ability, and Democracy: A Note on Thompson’s Challenge to Hong and Page,” Critical Review 29, no. 1 (2017): 72–87.; Daniel J. Singer, “Diversity, Not Randomness, Trumps Ability,” Philosophy of Science 86, no. 1 (2019): 178–91.

[9] Jeffrey Friedman, Power without Knowledge: A Critique of Technocracy (Oxford University Press, 2019).

[10] Ibid., p. 13-4.