The Grounds of Political Legitimacy

Fabienne Peter is Professor of Philosophy at the University of Warwick. She has written extensively on political legitimacy, and is interested in the question of what, if anything, justifies democracy. She has published a book on Democratic Legitimacy (Routledge, 2009), and is the author of the Stanford Encyclopedia of Philosophy entry on “Political Legitimacy.” One of her research projects at the time of the interview focuses on the normative foundations of conceptions of political legitimacy. She started this project when she held a Leverhulme Trust Research Fellowship in 2010/2011. A key question this project explores is how expertise bears on the justification of political decisions.

Before coming to Warwick in 2004, she taught at the University of Basel. Prior to that, she was a postdoc at the Harvard School of Public Health, where she worked with Sudhir Anand and Amartya Sen on a project on justice in health. In 1997, she was a Hoover Fellow at the Université Catholique de Louvain.

Fabienne Peter is currently writing a book on The Grounds of Political Legitimacy.

She was invited at the EHESS on the 9th May 2017 to present her paper “The Grounds of Political Legitimacy” (now published in the Journal of the American Philosophical Association, 2020, Vol. 6, n° 3, pp. 372-390) at the CESPRA Séminaire de Philosophie Politique Normative.


This interview was conducted prior to the seminar by Luc Foisneau (CNRS director of research).

Edited by Serge Blérard

Luc Foisneau – Why did you choose to work on political legitimacy rather than on justice?

Fabienne Peter – The embarrassing thing is that I have been working on legitimacy for a very long time. I started working on legitimacy in the context of my doctoral work. This is now more than twenty years ago. Even though I was interested in justice then, and I still am interested in justice, I had the impression—very much influenced by Rawls in his Political Liberalism1—that, as a political concept, legitimacy is at least as important, if not more important, than justice.

And, to give you just a little bit of personal background, as I started working on legitimacy, I was actually doing a PhD thesis in economics, and I was looking at social choice theory. After having read a lot of social choice theory, I basically got bored with it. I had the impression that the really interesting debates were happening in political theory and in political philosophy. It was a time when there was a lot of work on deliberative democracy. Bernard Manin has done a very important work on this2, and I was very much influenced by his work, which emphasized deliberation as a source of legitimacy. For me, something clicked when I read that work. With the social choice theory background on the one hand, and then this work on deliberative democracy, it seemed to me that what I should do—and what I did in the thesis—was to analyze the conception of legitimacy inherent in social choice theory and compare it to the conception of political legitimacy inherent in debates in political theory, specifically the debate around deliberative democracy. So, I contrasted the aggregative approaches that one finds in social choice theory, with Arrow and others, with the deliberative approaches one gets in the literature in political theory and political philosophy.

So, that is my biographical background, why I started working on legitimacy and how I started thinking about legitimacy in this contrast between economic approaches and philosophical approaches. But more broadly speaking, it seems to me—coming back to the point I mentioned earlier, about Rawls’s influence on me—that justice is, to use a word that is now used a lot in debates in political philosophy, an ideal concept. Legitimacy, however, is a normative concept that responds to what Rawls calls non-ideal circumstances. Rawls drew the distinction between ideal theory and non-ideal theory on the basis of compliance with normative principles3. I’m particularly interested in a question that is related to but distinct from Rawls’s own question: what should we do, or how should we decide, in a political context where we don’t know what justice requires?

My sense is that this is a very important question. And this is an important question because the circumstances of politics are such that we usually don’t know what justice requires. In general, in the political context, we are facing a lot of uncertainty, not just about normative facts, not just about the facts of justice, but also about empirical facts. Political debates are very complex. And it seems to me that the right response to that complexity is a normative concept that can do the work for us in this context.

And justice isn’t the right one, it seems to me. It is too demanding. What justice requires from within political debates is often more or less out of reach. But legitimacy cannot be. So, we need to think through the requirements of legitimacy in such a way that we can establish whether or not particular political decisions, and particular political institutions, are or are not legitimate. And I find that a fascinating question, a fascinating type of normative analysis.

Luc Foisneau – What is your conception of legitimacy?

Fabienne Peter – Let me explain how I think about legitimacy, first, before I answer your question. I think that one feature one finds, not just in debates on justice, but also in debates on legitimacy these days, is that they tend to be dominated by a moral perspective. So, the question is: “What does justice require morally?” And, analogously: “What does legitimacy require morally?” And that way of approaching things has always struck me as incomplete.

That relates back to what I said earlier. I think that one feature of political life is the uncertainty that we face when we are trying to make political decisions, when we are debating in a political forum, in newspapers, about what we should do. These debates are characterized by massive uncertainty. So, there is a certain epistemic predicament that comes with the political domain, which I find is on the whole insufficiently theorized. The moral perspective tends to take over too much.

Now there are, of course, important moral issues in politics. And justice is an important concern in politics. I don’t want to deny that for a second. I just don’t think a moral approach tells the full story, contrary to the dominant moral perspective on legitimacy, which is found in what I call in the paper “will-based conceptions of political legitimacy4”. They are trying to give a specifically moral answer to the question of what legitimacy requires, building on values such as freedom, equality, democracy, and so on. On Rawls’ own conception of political legitimacy, for example, political legitimacy requires that political decisions are justified on the basis of reasons that all persons, understood as free and equal citizens, can share5. On Philip Pettit’s republican conception, political decisions must be made in a way that secures equal control of all citizens over those decisions6. Will-based conceptions are particularly well-suited to capture the importance of democracy for political legitimacy.

A second family of theories of legitimacy, by contrast, has less of a say when it comes to debates on democratic legitimacy. But they highlight a very important feature of political life: part of what is at stake in good decision making, in politics as elsewhere, is to make the right decisions and political legitimacy is tied to how well political decision-making responds to a superior epistemic access to what the right decision is. In the paper, I call such conceptions of political legitimacy “belief-based”. Joseph Raz’ service conception of legitimate political authority is a prime example7. Raz argues that the legitimacy of political decisions depends on whether the authority is able to discern what the right decisions are, that is to say, make decisions that allow the citizens to better comply with what they should do anyway, independently of the decisions of the political authority.

So, there is a contrast here between this way of thinking about legitimacy, which focuses on getting to the right decisions, and the other one, which relates the problem of legitimacy to respecting people’s equal freedom in one way or another.

Now, my objection to will-based conceptions is that they neglect the question of how to get it right. So, I think the belief-based conceptions have something to offer there. On the other hand, there is a reason why the belief-based conceptions are not all that influential, especially when it comes to debates on democratic legitimacy. The main problem I see is that the epistemic circumstances of politics are often such that we don’t know what the right decision is, as I have already mentioned. Each of these two families of conceptions of legitimacy thus has shortcomings, especially when we are thinking of political legitimacy in a context of democratic politics as we have it today in countries like France or the UK or the US, and so on.

In the paper, what I try to do, beyond offering an interpretation of these two ways of thinking about legitimacy, is to argue that we need to preserve what is correct in each of those two ways of thinking about political legitimacy and combine these features in a hybrid conception of political legitimacy.

Luc Foisneau – As his work is central on legitimacy and he is not well-known in France, could you sum up the main ideas of Joseph Raz in two minutes?

Fabienne Peter – Ok. [Laughs] Well, Joseph Raz has made very important contributions to this topic8. Anyone who works on legitimacy has to grapple with Raz’ contributions, I think. Raz is a very influential legal philosopher who has thought a lot about legal authority and the duty to obey the law. But his work on authority, in general, is highly relevant for spelling out what legitimate political authority would entail.

So, on Raz’s conception, a practical authority, whether a legal or a political one, is legitimate just in case the directives of that authority allow those subjects to the authority (e.g., the citizens of a particular state, in the political domain) to better conform with reasons that apply to them anyway, independently of the directives, than if they relied on their own judgment. And if an authority such as a government satisfies this condition, then the citizens ought to do as they are told and not follow their own judgments.

A simple case to illustrate the idea is a coordination or cooperation problem. Without an authority telling you “Drive on the left!” or “Cooperate (say, by paying your taxes), otherwise we are going to punish you!”, the good equilibrium might not be obtainable. So, this is one way to think about how a political authority might help you to better conform with reasons that apply to you anyway. You have reasons to solve coordination and cooperation problems, but solving those problems might be difficult, impossible even, if we relied on the judgment of each individual. To clarify, in the driving case, for example, the point is not that driving on one particular side of the road is the right decision as such – driving on either side of the road is equally optimal, as long as everyone sticks to one side. The point is, rather, that there are optimal and suboptimal solutions to coordination problems, which can be very complex in political life, and citizens relying on their own judgment might a) not be able to discern the optimal solutions and b) not be able to coordinate on an optimal solution without being directed.

The important feature of Raz’s work, as I see it, is this notion that political legitimacy need not derive from the judgment of the citizens, contrary to what will-based conceptions of political legitimacy claim. Some political decisions are legitimate because they are made by someone in a good epistemic position to grasp what we all have reason to do.

That is where my challenge comes in. This is assuming too much, I think. Given the circumstances of politics as I described them earlier, very often even well-intentioned governments might not have good grasp of what the right decision is. And then the question is: what happens to the Razian conception of political legitimacy if the epistemic circumstances are as I described them?

One problem that Raz has in this case, I think, is that his conception seems to imply that, when the epistemic circumstances are such that no political authority would be in a better position than the citizens, all decisions are necessarily equally illegitimate. But that can’t be right. So, there is a problem which comes from the epistemic circumstances of politics. And what I argue in the paper is that the hybrid approach has the advantage that it deals with these epistemic circumstances, that it has something to say about what legitimacy requires even in those adverse epistemic circumstances.

Luc Foisneau – What I understand of your position is that it is not possible to base legitimacy only on normative authority, that is, an authority that clearly says what has to be done and gives good reasons to justify it, since there are political circumstances that don’t allow for such a decision-making procedure. Hence there is need for a second approach to legitimacy when what matters is the will of the subject and not the nature of the believes involved in the decision. But why do you consider this second kind of legitimacy secondary? Why do you call it “derivative”?

Fabienne Peter – The hybrid conception of political legitimacy recognizes two grounds of legitimacy: normative authority and the will of the citizens. This raises the question: are they equally important or is one ground more fundamental than the other? On the particular hybrid conception that I favour, normative authority has priority and the will of the citizens is a secondary ground of legitimacy. The value of this second way of generating legitimacy derives from its ability to respond to an epistemic problem that arises with normative authority.

Let me explain the basic thought by contrasting it with Rawls’s work on public justification. Rawls captured something very important with this notion of public justification. Rawls argued that some form of agreement among the citizens, loosely speaking, is the source of normativity in the political domain. Rawls thought that public justification is the right approach to justification in matters of justice as well as legitimacy. Many working in the Rawlsian tradition have followed Rawls in this regard. This is most obvious in the public reasons conceptions of political legitimacy or justice9. Such conceptions hold that justice and legitimacy derive from some form of agreement among free and equal citizens. In this way of thinking about justice or legitimacy, some form of agreement among the citizens is thus normatively fundamental.

As I said, I believe that Rawls’ work on public justification is extremely important and has been very fruitful. But what Rawls hasn’t done is ask when it is appropriate to adopt some form of public justification. He just postulated that this is the right approach in the political domain and contrasted it with more traditional ways of thinking about justification that come from epistemology, such as reliabilism, evidentialism, and so on. By focusing on public justification, he embraced a divorce between moral and political theory, on the one hand, and epistemology, on the other.

As it happens, I think that Rawls is right that in many political contexts, some form of agreement-based justification is the right approach. But there is still an interesting question of when exactly public justification matters, and when it matters more that we make the right decisions, based on what we have reason to believe is the right decision. That question is at the heart of the work that I currently engaged in and Rawls’ work doesn’t cover this question.

The comparison with Rawls gives us another perspective on the hybrid conception that I’m advocating. Given the importance of making the right decisions, my starting-point is the question of whether we’re in a sufficiently good epistemic position to grasp what it is. I thus reject the idea that agreement-based justification is normatively fundamental, even in the political domain. But I argue that agreement-based justification matters as a secondary ground of political legitimacy when we don’t have a sufficiently good grasp of what the right decision is.

Luc Foisneau – What are the normative reasons why democracy can be, as it is nowadays, under pressure? Has it to do with the fact that Enlightenment values are also coming under pressure?

Fabienne Peter – In the paper, I focus on the contemporary literature exclusively. But the distinction I draw resonates with positions people have taken in the history of political thought. Divine command theories of legitimacy provide an example of belief-based conceptions. According to divine command theories, the right decision is the one that God truly wills, and legitimate political authority is afforded to those who are in a position to discern God’s will. But with the Enlightenment, this conception of legitimacy became discredited. What emerged instead was a conception of legitimacy which, in one way or another, built on equal freedom and on everyone’s equal ability to reason. This way of thinking about legitimacy corresponds to what I call will-based conceptions of political legitimacy.

To jump into the present, there is now a strange resurfacing of the idea that we need strong leaders to tell us what’s what. It is strange because the development doesn’t seem to resonate with belief-based conceptions. Instead, it is a kind of unmitigated authoritarianism. So, Enlightenment values are definitively coming under pressure. And I think two things are important to note in this regard.

First of all, we need to take this seriously as a development. I don’t think it would be wise to just dig our heels in and defend democracy no matter what. I think that we should be open to the question that there may be alternatives to democracy, even if it’s currently unclear what they might be. For a long time, democracy seemed to be the inevitable political regime. But there are now criticisms from all sides, and we need to take them seriously.

In particular, I think that we need to take the possibility seriously that democracies can get it badly wrong. And the hybrid approach that I propose allows us to do that. It allows us to take seriously the idea that sometimes there are right decisions, and that political legitimacy depends on political collectives’ making the right decisions and avoiding wrong decisions. At the same time—and this is my second comment—we should be wary of the rise of unmitigated authoritarianism. And the reason is that strong leaders are only as good as the decisions that they are making. Given what I said about the epistemic circumstances of politics, which tend to make it very difficult to discern how we should live together, we should be very careful with relying on strong leaders. In all likelihood, they don’t know what they are doing.

Unfold notes and references
Retour vers la note de texte 10451


John Rawls, Political Liberalism, New York, Columbia University Press, 1993.

Retour vers la note de texte 10452


Bernard Manin, The Principles of Representative Government, Cambridge, Cambridge University Press, 1997.

Retour vers la note de texte 10453


For Rawls’ distinction between ideal theory and non-ideal theory, see John Rawls, A Theory of Justice, Cambridge (Mass.), Harvard University Press, 1971, p. 8f.

Retour vers la note de texte 10454


Fabienne Peter, “The Grounds of Political Legitimacy,” Journal of the American Philosophical Association, vol. 6, n° 3, 2020, p. 372-390.

Retour vers la note de texte 10455


John Rawls, Political Liberalism, New York, Columbia University Press, 1993.

Retour vers la note de texte 10456


Philip Pettit, On the People’s Terms. A Republican Theory and Model of Democracy, Cambridge, Cambridge University Press, 2012.

Retour vers la note de texte 10457


Joseph Raz, The Morality of Freedom, Oxford, Oxford University Press, 1986.

Retour vers la note de texte 10458


Joseph Raz, The Morality of Freedom, Oxford, Oxford University Press, 1986. See especially Part I, “The Bounds of Authority.”

Retour vers la note de texte 10459


See Jonathan Quong, Liberalism Without Perfection, Oxford, Oxford University Press, 2011; and Gerald Gauss, The Order of Public Reason: A Theory of Freedom and Morality in a Diverse and Bounded World, Cambridge, Cambridge University Press, 2011.