Really appreciate the clarity here—your framing of democracy as a conflict-resolution tool rather than a truth-finding mechanism is refreshingly grounded. But I wonder if the analysis stops short of the realist critique it sets out to offer.
You describe disagreement as something to be managed, but not as something shaped—by capital, media control, and structural power. What looks like broad consent often hides the fact that key issues (public ownership, taxing wealth, opposition to war crimes) are consistently excluded from the platforms of those with any real chance at power.
If the arena is rigged before the contest begins, isn’t that a bigger democratic illusion than the pursuit of the common good?
You're absolutely right to highlight that disagreement is not only something to be managed, but is also shaped by structural forces—capital, media control, institutional design, and more. What appears as broad consensus can indeed reflect the limits of what is considered politically possible or visible, and key issues can be systematically kept off the agenda.
From a realist perspective, that’s precisely part of what the political game entails: social bargaining is not inherently fair, and power asymmetries are part of the terrain. But democratic institutions matter because they give more people more tools to contest what they don’t like. The arena is far from level, but it's also not fully closed. Political outsiders can gain traction when they articulate demands that resonate with excluded or discontented groups.
The risk, as I argue in the post, is in setting up democratic institutions to fail by judging them against ideals they are not designed to realise. Like markets, democracies are imperfect coordination mechanisms. Yet, just as markets are often better than the alternatives despite their flaws, democracies tend to offer broader avenues for contestation and change than more closed systems.
That doesn’t mean we shouldn’t criticise them—quite the opposite. But we should do so with an eye to reform and resilience, not disillusionment. The danger is that disappointment with actually existing democracy fuels support for alternatives that, in practice, leave most people worse off.
Something I’ve been wondering about as I reflect on your piece: you rightly point out that democratic institutions allow for more contestation than closed systems—but what about the possibility of more open systems?
You draw the line at defending what we have, but don’t really explore models that might go further: participatory budgeting, citizens’ assemblies, deliberative processes, even AI-supported consensus tools. These aren’t perfect, of course—but if we’re being realist, isn’t it also worth asking why these approaches are so often overlooked or excluded?
I suppose I’m curious whether we might expand the frame a little—not to romanticise ideal democracy, but to keep the imaginative space open for something better.
I fully agree that existing democratic institutions can and should be improved. My aim in this piece wasn't to draw a line at defending what we have, but to clarify what it is we do have, what existing democratic institutions are actually doing, how they function in practice, and why they’ve emerged as they have.
I believe that understanding current institutions from a realist perspective—seeing them as mechanisms for managing distributional conflict and not as devices for discovering a shared truth—is essential groundwork for any meaningful reform. If we don’t grasp their logic and limits, we risk supporting changes that sound good but are actually misguided (and potentially dangerous).
That’s why, before turning to institutional reform, I’ll be doing a series of posts exploring the foundations of political philosophy from this realist angle. A key influence here is Ken Binmore, whose book Natural Justice offers a powerful game-theoretic framework for thinking about fairness and political rules. I’ll be presenting his ideas and reflecting on what they might mean for reimagining political institutions—including precisely the kinds of mechanisms you point to.
I would be extremely interested in that. I was quite skeptical of well-founded normative political philosophy as being hopelessly idealist and weird. The only possible exception to this line of thought was Rawls, but, despite my initial interests, his conclusions have eventually struck me as grossly counter-intuitive, as I just can't bring myself to reject the idea of merit. Seeing Rawls' methodology being repackaged and refined with game-theoretic language would probably revive my interest in seriously thinking through political matters.
Binmore's work can be seen as a continuation of Rawls' work with naturalistic foundations. He disagrees with Rawls' conclusions, but Rawls himself became more sceptical of his Theory of Justice ideas and encouraged Binmore' in his enterprise.
Why are you pointing to Arrow? Ordinal utility yields insufficient information for a social welfare function, sure. But cardinal utility doesn't. So, on what basis do you claim there is no common good?
Thanks for your comment, Asher. Cardinal utility allows aggregation and thus sidesteps Arrow’s impossibility result. However, using a social welfare function assumes there is a correct scale of welfare to maximise—hence, a common good. That’s precisely the issue at stake: why would one particular welfare metric be the right one? Who decides on the weights, trade-offs, and distributional priorities? While aggregation with cardinal utility is mathematically possible, it embeds the assumption that the trade-offs between the interests of different social groups have already been settled. But by whom, and how?
Political institutions exist because people don’t agree on these matters. My point is that democratic institutions, for all their imperfections, are in expectation the best for most people.
You equivocate between different arguments here. As I understand it, the idea you're criticizing is that the ideal democracy _ought_ to be aimed at producing the public good.
You respond that this is misguided, because the public good is logically inconsistent, as shown by Arrow, and therefore there is no utility function which could reasonably be called the public good.
I respond that Arrow is insufficient proof of the claim that the public good can't exist, because there's no reason to assume individuals have only ordinal utility functions, which is required for the theorem to hold.
You now respond by saying that, rather than there being no utility functions which could reasonably be called the public good, there are actually several -- and then conflating the idea of pursuing the public good with the problem of _defining_ it. But, this is just an issue of vagueness of the statement you're responding to.
If I said, democracy ought to be aimed at maximizing the sum of individual utilities, you might agree or disagree, but you can't say it's not something a political system could do, that the goal is logically inconsistent and therefore the exhortation fundamentally misguided.
Cf. Suppose I said, "people ought to worship God". You respond that a being with traits X, Y, and Z is logically inconsistent, cannot exist, and therefore it's fundamentally misguided to worship it. I respond that actually, I don't consider God to have trait Z, and therefore you've not really made an argument against my position, but rather a straw man of it. You respond that since the Sikh, Jewish, Christian, Hindu, etc God(s) all exhibit traits X and Y, and there's no way for an individual to choose between them, then you're still right. But really, I just didn't make clear what I was talking about, and you've not made any argument against the point I'm actually making.
My critique is aimed at the procedural idealist vision of democracy—as a system that enables the discovery or expression of a uniquely valid general will through deliberation or aggregation. Arrow’s theorem undercuts this vision by showing that even under fairly weak assumptions, collective decision-making doesn't yield a coherent outcome.
You're right that with cardinal utility we can define social welfare functions and say democracy ought to maximise them. But that already presupposes a particular normative framework—what to measure, how to weigh different people's utilities, and so on. That’s not a neutral or consensus-based process; it’s inherently political. In short, you might say that we should maximise a given function, but what if I don't agree? How is that an obviously valid public good?
What I’m arguing is that the democratic discovery of a single, objectively correct public good is impossible in practice, because preferences are plural, contested, and grounded in different interests. Invoking a social welfare function doesn't solve the problem—it just relocates it.
"...the example set by classical Athens—particularly the image of citizens gathering in the Agora, the central public space, to discuss public affairs and make collective decisions."
That would be HALF the adults, women being kept well out of the public space and decision making process.
Indeed, the voting population was only around 10–20%, given that nearly half the population was slaves and there were also some non-voting foreigners. Athens is a model not for what it was in absolute terms but for how it compares to the typical authoritarian regimes of the past.
"Most political debates in democracies are not about finding a truth that everybody agrees on, they are about settling decisions in situations where interests conflict"
This gets part of it right, but it misses something big: politics isn’t just about interests—it’s about emotions, identity, and desire. Most voters aren’t weighing pros and cons, they’re reacting to stories and feelings. And the larger the selectorate, the more emotional and less rational it tends to get. History is full of examples where demagogues rise by tapping into that. Maybe the real challenge—unsolved since Plato—is figuring out what the optimal size of a selectorate should be to balance legitimacy with reason.
Thanks for your comment, Michael. From a realist perspective, the size and composition of the electorate, those who have a say in who governs, is not primarily a normative choice about who is “rational enough” to decide. It’s a consequence of bargaining dynamics. Inclusion in the selectorate is driven by who has enough power, individually or collectively, to be worth including, relative to the cost of excluding them.
Once a selectorate becomes large, including additional marginal groups doesn’t dilute power much, while exclusion can be costly—because excluded groups can resist, disrupt, or delegitimise the system. That’s why, historically, the expansion of suffrage often followed periods of social pressure or unrest. It wasn’t because elites suddenly saw the value of broader rational deliberation—it was because the cost of continued exclusion became too high.
So while arguments about knowledge, rationality, or emotional susceptibility are often raised to justify exclusion, these are better understood as post hoc justifications. The real underlying factor is bargaining power. This is also why democratic institutions tend to work better for more people—not because voters are always well-informed, but because inclusion expands the range of interests that need to be taken seriously.
I see you resisted the urge to put the classic Churchill quote which I think is quite on point and speaks to the mindset of someone defending democracy!
I also believe that there’s no reason why a majority of people will reach the right decision, assuming there is one. Even when a correct course for most people is available, it’s quite possible that the same majority will simply be mistaken about their own interest and the wrong vote will take place. So, I’m very sympathetic to other explanations of the function of democracy. And beating the prisoner’s dilemma is certainly one of them.
However, I do think that there’s another function which is kind alluded to here, but just to insist: a great feature of the system is that it has flexibility. It allows change in a relatively short period of time. So if wrong decisions are taken and recognized, they can be course-corrected.
A complex system, such as a modern society, has no easy solutions. What’s the best economic policy? How much investment in technology? How many immigrants? My contention is that no one knows the correct answer, but a democracy allows a trial-and-error process to at least take place. It’s like a giant random process of finding the best policies.
They mention it here in the article: "We can appreciate why democracy, with such political markets—full of imperfections but with limits to how much elite bargaining can veer away from the broad public interest—is the “worst system except for all the others”."
Good catch! I don't know how I missed it. That quote so perfectly summarizes the long process that led to democracy and how unintuitive a system it is.
The expression includes a frustration with the imperfection of democracy. In the perspective of this post, this frustration is based on our (misplaced) intuitions about what democracy should be.
Really appreciate the clarity here—your framing of democracy as a conflict-resolution tool rather than a truth-finding mechanism is refreshingly grounded. But I wonder if the analysis stops short of the realist critique it sets out to offer.
You describe disagreement as something to be managed, but not as something shaped—by capital, media control, and structural power. What looks like broad consent often hides the fact that key issues (public ownership, taxing wealth, opposition to war crimes) are consistently excluded from the platforms of those with any real chance at power.
If the arena is rigged before the contest begins, isn’t that a bigger democratic illusion than the pursuit of the common good?
You're absolutely right to highlight that disagreement is not only something to be managed, but is also shaped by structural forces—capital, media control, institutional design, and more. What appears as broad consensus can indeed reflect the limits of what is considered politically possible or visible, and key issues can be systematically kept off the agenda.
From a realist perspective, that’s precisely part of what the political game entails: social bargaining is not inherently fair, and power asymmetries are part of the terrain. But democratic institutions matter because they give more people more tools to contest what they don’t like. The arena is far from level, but it's also not fully closed. Political outsiders can gain traction when they articulate demands that resonate with excluded or discontented groups.
The risk, as I argue in the post, is in setting up democratic institutions to fail by judging them against ideals they are not designed to realise. Like markets, democracies are imperfect coordination mechanisms. Yet, just as markets are often better than the alternatives despite their flaws, democracies tend to offer broader avenues for contestation and change than more closed systems.
That doesn’t mean we shouldn’t criticise them—quite the opposite. But we should do so with an eye to reform and resilience, not disillusionment. The danger is that disappointment with actually existing democracy fuels support for alternatives that, in practice, leave most people worse off.
Something I’ve been wondering about as I reflect on your piece: you rightly point out that democratic institutions allow for more contestation than closed systems—but what about the possibility of more open systems?
You draw the line at defending what we have, but don’t really explore models that might go further: participatory budgeting, citizens’ assemblies, deliberative processes, even AI-supported consensus tools. These aren’t perfect, of course—but if we’re being realist, isn’t it also worth asking why these approaches are so often overlooked or excluded?
I suppose I’m curious whether we might expand the frame a little—not to romanticise ideal democracy, but to keep the imaginative space open for something better.
I fully agree that existing democratic institutions can and should be improved. My aim in this piece wasn't to draw a line at defending what we have, but to clarify what it is we do have, what existing democratic institutions are actually doing, how they function in practice, and why they’ve emerged as they have.
I believe that understanding current institutions from a realist perspective—seeing them as mechanisms for managing distributional conflict and not as devices for discovering a shared truth—is essential groundwork for any meaningful reform. If we don’t grasp their logic and limits, we risk supporting changes that sound good but are actually misguided (and potentially dangerous).
That’s why, before turning to institutional reform, I’ll be doing a series of posts exploring the foundations of political philosophy from this realist angle. A key influence here is Ken Binmore, whose book Natural Justice offers a powerful game-theoretic framework for thinking about fairness and political rules. I’ll be presenting his ideas and reflecting on what they might mean for reimagining political institutions—including precisely the kinds of mechanisms you point to.
I would be extremely interested in that. I was quite skeptical of well-founded normative political philosophy as being hopelessly idealist and weird. The only possible exception to this line of thought was Rawls, but, despite my initial interests, his conclusions have eventually struck me as grossly counter-intuitive, as I just can't bring myself to reject the idea of merit. Seeing Rawls' methodology being repackaged and refined with game-theoretic language would probably revive my interest in seriously thinking through political matters.
Binmore's work can be seen as a continuation of Rawls' work with naturalistic foundations. He disagrees with Rawls' conclusions, but Rawls himself became more sceptical of his Theory of Justice ideas and encouraged Binmore' in his enterprise.
Why are you pointing to Arrow? Ordinal utility yields insufficient information for a social welfare function, sure. But cardinal utility doesn't. So, on what basis do you claim there is no common good?
Thanks for your comment, Asher. Cardinal utility allows aggregation and thus sidesteps Arrow’s impossibility result. However, using a social welfare function assumes there is a correct scale of welfare to maximise—hence, a common good. That’s precisely the issue at stake: why would one particular welfare metric be the right one? Who decides on the weights, trade-offs, and distributional priorities? While aggregation with cardinal utility is mathematically possible, it embeds the assumption that the trade-offs between the interests of different social groups have already been settled. But by whom, and how?
Political institutions exist because people don’t agree on these matters. My point is that democratic institutions, for all their imperfections, are in expectation the best for most people.
You equivocate between different arguments here. As I understand it, the idea you're criticizing is that the ideal democracy _ought_ to be aimed at producing the public good.
You respond that this is misguided, because the public good is logically inconsistent, as shown by Arrow, and therefore there is no utility function which could reasonably be called the public good.
I respond that Arrow is insufficient proof of the claim that the public good can't exist, because there's no reason to assume individuals have only ordinal utility functions, which is required for the theorem to hold.
You now respond by saying that, rather than there being no utility functions which could reasonably be called the public good, there are actually several -- and then conflating the idea of pursuing the public good with the problem of _defining_ it. But, this is just an issue of vagueness of the statement you're responding to.
If I said, democracy ought to be aimed at maximizing the sum of individual utilities, you might agree or disagree, but you can't say it's not something a political system could do, that the goal is logically inconsistent and therefore the exhortation fundamentally misguided.
Cf. Suppose I said, "people ought to worship God". You respond that a being with traits X, Y, and Z is logically inconsistent, cannot exist, and therefore it's fundamentally misguided to worship it. I respond that actually, I don't consider God to have trait Z, and therefore you've not really made an argument against my position, but rather a straw man of it. You respond that since the Sikh, Jewish, Christian, Hindu, etc God(s) all exhibit traits X and Y, and there's no way for an individual to choose between them, then you're still right. But really, I just didn't make clear what I was talking about, and you've not made any argument against the point I'm actually making.
My critique is aimed at the procedural idealist vision of democracy—as a system that enables the discovery or expression of a uniquely valid general will through deliberation or aggregation. Arrow’s theorem undercuts this vision by showing that even under fairly weak assumptions, collective decision-making doesn't yield a coherent outcome.
You're right that with cardinal utility we can define social welfare functions and say democracy ought to maximise them. But that already presupposes a particular normative framework—what to measure, how to weigh different people's utilities, and so on. That’s not a neutral or consensus-based process; it’s inherently political. In short, you might say that we should maximise a given function, but what if I don't agree? How is that an obviously valid public good?
What I’m arguing is that the democratic discovery of a single, objectively correct public good is impossible in practice, because preferences are plural, contested, and grounded in different interests. Invoking a social welfare function doesn't solve the problem—it just relocates it.
"...the example set by classical Athens—particularly the image of citizens gathering in the Agora, the central public space, to discuss public affairs and make collective decisions."
That would be HALF the adults, women being kept well out of the public space and decision making process.
Indeed, the voting population was only around 10–20%, given that nearly half the population was slaves and there were also some non-voting foreigners. Athens is a model not for what it was in absolute terms but for how it compares to the typical authoritarian regimes of the past.
"Most political debates in democracies are not about finding a truth that everybody agrees on, they are about settling decisions in situations where interests conflict"
This gets part of it right, but it misses something big: politics isn’t just about interests—it’s about emotions, identity, and desire. Most voters aren’t weighing pros and cons, they’re reacting to stories and feelings. And the larger the selectorate, the more emotional and less rational it tends to get. History is full of examples where demagogues rise by tapping into that. Maybe the real challenge—unsolved since Plato—is figuring out what the optimal size of a selectorate should be to balance legitimacy with reason.
Thanks for your comment, Michael. From a realist perspective, the size and composition of the electorate, those who have a say in who governs, is not primarily a normative choice about who is “rational enough” to decide. It’s a consequence of bargaining dynamics. Inclusion in the selectorate is driven by who has enough power, individually or collectively, to be worth including, relative to the cost of excluding them.
Once a selectorate becomes large, including additional marginal groups doesn’t dilute power much, while exclusion can be costly—because excluded groups can resist, disrupt, or delegitimise the system. That’s why, historically, the expansion of suffrage often followed periods of social pressure or unrest. It wasn’t because elites suddenly saw the value of broader rational deliberation—it was because the cost of continued exclusion became too high.
So while arguments about knowledge, rationality, or emotional susceptibility are often raised to justify exclusion, these are better understood as post hoc justifications. The real underlying factor is bargaining power. This is also why democratic institutions tend to work better for more people—not because voters are always well-informed, but because inclusion expands the range of interests that need to be taken seriously.
I see you resisted the urge to put the classic Churchill quote which I think is quite on point and speaks to the mindset of someone defending democracy!
I also believe that there’s no reason why a majority of people will reach the right decision, assuming there is one. Even when a correct course for most people is available, it’s quite possible that the same majority will simply be mistaken about their own interest and the wrong vote will take place. So, I’m very sympathetic to other explanations of the function of democracy. And beating the prisoner’s dilemma is certainly one of them.
However, I do think that there’s another function which is kind alluded to here, but just to insist: a great feature of the system is that it has flexibility. It allows change in a relatively short period of time. So if wrong decisions are taken and recognized, they can be course-corrected.
A complex system, such as a modern society, has no easy solutions. What’s the best economic policy? How much investment in technology? How many immigrants? My contention is that no one knows the correct answer, but a democracy allows a trial-and-error process to at least take place. It’s like a giant random process of finding the best policies.
They mention it here in the article: "We can appreciate why democracy, with such political markets—full of imperfections but with limits to how much elite bargaining can veer away from the broad public interest—is the “worst system except for all the others”."
Good catch! I don't know how I missed it. That quote so perfectly summarizes the long process that led to democracy and how unintuitive a system it is.
The expression includes a frustration with the imperfection of democracy. In the perspective of this post, this frustration is based on our (misplaced) intuitions about what democracy should be.