Chris Freiman’s Unequivocal Justice

Christopher Freiman  has just published a fabulous book, Unequivocal Justice, the first book in Routledge Press’s new “Political Philosophy for the Real World” series.

It is a tour de force of philosophical excellence. It may well be the best book of political philosophy published in 2017. I certainly haven’t read anything this year that comes close to competing with it.

Imagine a person said, “I have a solution to the problem of drunk driving. However, my solution works only in a world where alcohol hadn’t been invented.” There’s something deeply silly and incoherent about that.

Well, it turns out that the mainstream of political philosophy over the past 50 years has precisely this problem. The mainline of political philosophy, when it tries to defend or critique various institutions, has largely been a joke, Freiman shows us, though he’s too polite to put it that bluntly.

What Freiman shows is that Rawls, Freeman, Ackerman, Dworkin, and a number of other left-liberal philosophers are making this kind of mistake over and over. His critique is so devastating that you might as well take Rawls’s writings about institutions and throw them in the trash; they are now, thanks to Chris, nothing more than artifacts of historical interest.

Chris starts by saying,

A perfect state is a pointless state.

The point of a state is to mitigate injustice. If Rich would donate his 40% to the poor, the state wouldn’t need to tax his income. If Mimi would buy a hybrid instead of a Hummer, the state wouldn’t need to cap her emissions. But since virtue alone won’t do the job, the state needs to redistribute equitably and regulate efficiently.

…But here’s the problem: the very reasons why the state is needed are reasons why the state won’t work.


Rawls writes mostly at the level of ideal theory. But, Freiman shows, an ideal theory of the state is incoherent. (Yes, he responds to Kavka’s argument otherwise.) Under ideal conditions–in which people are stipulated to comply fully with the requirements of morality and justice–there simply is no need for a state, period. There is no need to create an institution which claims a monopoly on violence and which enforces rules through threats of violence. Ideal theory must be anarchist.

Coercion is needed to defend justice only when society is less than fully just. But when society is less than fully just, we cannot stipulate the ideal justness of the state itself. So we arrive at the dilemma for ideal theories of the state: either (i) society is fully just, in which case there is no need for a state, or (ii) society is not fully just, in which we case we may not stipulate the state itself is just.

In order to create a need for a state, Rawls (and his followers) equivocate. They posit bad behavior in the private sector. But then, in order to defend their favored regime and in order to avoid the criticism that the regime itself might be corrupt and make things worse, they imagine away all bad behavior in the public sphere.

For example, Rawls claims that we need to equalize incomes in order to prevent the rich from buying power for themselves. (Freiman thinks that’s a weird argument to begin with; in order to stop people from polluting, we don’t equalize income; rather we regulate pollution.) But here’s the dilemma.

…The only way to ground both (i) the need for regulation and (ii) the stipulation of the regulation’s success is to equivocate in precisely the way Rawls does.

So, to restore consistency, Rawls needs to resolve a dilemma: Either (i) the rich aren’t buying up state power, in which case equalization isn’t necessary, or (ii) the rich are buying up state power, in which case they can subvert equalization by buying up the state power unleashed to do the equalizing. Neither option justifies an a priori demand for equalization.

A few other philosophers, including G. A. Cohen and me, have pointed out that Rawls makes cartoonishly bad arguments like this here and there. But Freiman methodically goes through Rawls and a few others, and finds they make such arguments constantly. Rawls’s version of the public goods argument, his argument for redistribution taxation, his argument for the existence of the state, and so on, all have the same form: He’s giving us a theory about how to solve drunk driving, but his solution can only be stipulated to work in a world where alcohol had never been invented.

In the end, the mistake is that Rawls is trying to make a priori arguments for institutions, regime-types, and rules. These arguments all fail. They are no substitute for doing careful PPE-style empirical institutional analysis. Freiman closes by warning left-liberals not just to presume that empirical analysis will vindicate the exact institutions they were defending on entirely a priori grounds.

Again, the book is a tour-de-force. You should read it. It will make you a better thinker.

Here’s my blurb for the book:

Unequivocal Justice, with its delightful and engaging prose, is a devastating critique of the dominant arguments and methods in political philosophy. It shows that almost everything Rawls and other left-liberals have said about institutions over the past 50 years is not merely wrong, but incoherent. It should–if philosophers have an intellectual integrity–change the field forever.

Strong words, but entirely deserved.

  • Salem

    So the world of philosophy has finally discovered Federalist No. 51?

    If men were angels, no government would be necessary. If angels were to govern men, neither external nor internal controls on government would be necessary. In framing a government which is to be administered by men over men, the great difficulty lies in this: you must first enable the government to control the governed; and in the next place oblige it to control itself.

    It’s a shame that Rawls, Dworkin et al were writing so many years before Madison, and so did not have the opportunity to respond to his critique.

    • Jason Brennan

      In a sense, yes, rediscovery.

      Freiman has a section where he quotes that and a number of other papers. He says that the left-liberal insistence on behavioral asymmetry is bizarre and puts them at odds with the rest of the liberal tradition.

      • Nkaplan

        That is because they are not liberals, they are egalitarians. At best they are ‘liberal-egalitarians’, which is to say, egalitarians who are more squeamish than most socialists about the means actually required to achieve their (worthless) end of much greater equality of outcome. They are not – as they often describe themselves – ‘egalitarian liberals’, which is nothing more than a contradiction in terms.

        • Sean II

          I endorse your comment because your main point is right: there’s nothing bizarre about the left being at odds with the liberal tradition, because they aren’t part of it and really haven’t been for 100 years. This callout won’t give them a moment’s pause.

          But I disagree with your diagnosis of egalitarianism as the primary tumor. It can’t be that, because well…look how lefties behave. These are the least egalitarian people around, and they could hardly be more obvious about their ambition to become a confucian style overclass-by-examination. Hence why they love school so much.

          Or to put it more bluntly: when they complain about “inequality caused by the market”, it’s the “market” they object to, not the “inequality”. They don’t even blink at forms of inequality created by credentialing, bureaucratic rank, etc. On the contrary they relish inequality when it’s created by a selection process they understand.

          A better, simpler reason why the left persists in pushing behavioral asymmetry against all evidence is frank authoritarianism. They like stories with a leader, and they don’t much get stories where the protagonist is some sort of private action or spontaneous order. The The West Wing offers a nice glimpse into their way of thinking: nothing good ever happens except some boss decides it should, and the only question that really matters is which boss you get.

    • brandonberg

      As I observed several years ago, if men were angels, government would work.

      • A. Alexander Minsky

        According to Mises, even if men were angels, socialism would still fail. The reason for this is that an efficient economy is impossible sans the information one gets from the pricing system.

        • brandonberg

          Depends on whether the angels are omniscient, or just really nice.

          • A. Alexander Minsky

            Is there any religious faith in which angles are omniscient? In the Abrahamic tradition, only God is all knowing.

        • Albionic American

          The Mises Institute ironically disregards price signals when it gives away digitized versions of Mises’s work “for free” over the internet. Why doesn’t this practice misallocate scarce resources and cause economic chaos?

          • Abeed Khader

            Digitized version? How is it a scarce resource?

    • David Alexander

      And Federalist 51 is based on Calvin’s commentary on Galatians:

      “If we were like angels, blamelessly and freely able to exercise self control, we would not need rules and regulations. Why then do we have so many laws and statutes? Because of mans wickedness; for he is constantly overflowing with evil. This is why a remedy is required.”

  • So THIS is the long awaited work you have been talking about… I am very excited!

  • Blain

    Based on this post at least, this book’s argument seems to rest on a misunderstanding of Rawlsian ideal theory.

    Strict compliance is assumed only within the 1st stage of the original position, specifically, the stage at which the parties initially select which conception of justice should govern the basic structure of the society in which the citizens whom they represent will live out their lives. The 2nd stage of the original position involves determining whether a fully just well-ordered society—a society with a basic structure that is organized in accordance with the conception of justice selected at the 1st stage—would be stable over time for the right reasons, namely, through the free compliance of its reasonable citizens (see Justice as Fairness, §§54–55). Here compliance is *not* assumed. Rather, it must be shown that citizens raised within such a society will acquire an effective ‘sense of justice’, and, moreover, will regard their sense of justice as appropriately regulative of their behavior over time. (Hence Rawls discusses, inter alia, the ‘strains of commitment’, i.e., psychological factors that may prevent real citizens from being able to comply adequately with the requirements of justice.) State coercion is necessary at the 2nd stage in order to overcome the assurance problem with respect to citizens’ compliance with the requirements of justice in a large-scale society (i.e., to assure citizens that others will act reliably as justice requires). However, such coercive power would be exercised quite rarely (if at all) with respect to most citizens (See: A Theory of Justice, pp. 237–238, 277, 296, 305–306).

    So there are 2 stages to Rawls’s overall argument for selecting the principles of justice as fairness. In the 1st stage of the original position, strict compliance is assumed by the parties; within the second stage, compliance is *not* assumed, but must be shown to be justified given (inter alia) realistic assumptions about human psychology and the functioning of institutions.

    I recall nothing in Rawls’s theory that suggests that different kinds of behaviours (‘bad’ versus ‘good’) are assumed vis-à-vis private (economic) versus public (political) agents or institutions in the 2nd stage.

    Also, regarding: ‘Under ideal conditions–in which people are stipulated to comply fully with the requirements of morality and justice–there simply is no need for a state, period.’ Presumably, even under ideal conditions so described, the state would still be necessary to solve various collective action problems and ensure public knowledge of the relevant rules governing public life (even if it would not need to exercise coercion to ensure compliance).

    • Chris Freiman

      Thanks for the comment. I address the assurance argument in the first chapter of the book. My reply, in brief, is that positing the state as a solution to the assurance problem merely generates another assurance problem. If I won’t make a contribution unless I have assurance that others will contribute as well, then I won’t contribute to the state itself (e.g., by voting) unless I have assurance that others will make their own contributions.

      As I put it in the book, “But how do we get assurance that others are contributing their good votes? When I’m in the voting booth deciding whether or not to vote for the best candidate or to goof off and vote for Santa Claus, how do I know that the person next to me isn’t goofing off? Rawls’s move—appealing to the state as an enforcer—doesn’t work because the contributions in question are to the state itself. You cannot appeal to the state as a device for assuring me that others are also voting well because voting well is a precondition of the proper functioning of the state. Rawls’s proposed solution to the assurance problem simply creates another assurance problem. We need a well-functioning state to solve the assurance problem, but we can’t get a well-functioning state because of the assurance problem.”

      • Blain

        Sure, the state cannot assure its own justice. The role of the threat of state coercion (rarely if ever exercised) within a well-ordered society is to ensure *ongoing* compliance on the part of citizens, to assure them that just decisions already made will be carried through and complied with by all citizens. State coercion serves as a collective self-binding mechanism by citizens.

        The state cannot ensure that its own laws are just through *its* exercise of coercion. That requires, instead, that public officials shape just laws, and that voters hold public officials accountable for doing so.

        If I understand you correctly, your worry is that *this* process is (also) vulnerable to an assurance problem. (‘How can be assured that other citizens are voting for the laws that they believe are most just? And if I don’t have that assurance, perhaps I should simply vote for laws that I think best serve my interests, and not comply with my sense of justice?’)

        But here Rawls relies on the role of public reason within the ‘public political forum’. By justifying political decisions in terms of public reasons (which are themselves drawn from a reasonable political conception of justice, such as justice as fairness), public officials assure others (including citizens) that they are enacting politically just laws. (Of course, they may be mistaken, and we can vote against them, but doing so would not be because we doubt their commitment to a reasonable political conception of justice—at least if we think that they are in fact using public reasons.)

        So Rawls has two solutions to the two separate assurance problems. (A) With respect to ‘everyday’ compliance with just laws, the threat of state coercion (almost never actually exercised within a well-ordered society) overcomes this assurance problem. And (B) with respect to ensuring that public officials (and citizens when voting) shape laws that are just (through legislation, interpretation of laws, etc.), we hold public officials to the idea of public reason.

        Perhaps you also address this aspect (B) of Rawls’s account. I guess I’ll have to read the relevant chapter…

        • Chris Freiman

          I appreciate the comment. I posted a reply earlier, but it hasn’t shown up yet, so I’ll try again. My worry centers on the claim that “voters hold
          public officials accountable for” shaping just laws. Holding public
          officials accountable for shaping just laws via the vote is costly–I
          have to do some research on the public officials, justice, etc., drive
          to the polls, and so on. By Rawls’s own lights, I won’t pay those costs
          unless I have assurance that other citizens are doing the same. But I
          don’t see how stipulating that public officials are committed
          to”justifying political decisions in terms of public reasons (which are
          themselves drawn from a reasonable political conception of justice, such
          as justice as fairness)” overcomes this problem. To see why, suppose
          that candidate A justifies their political decisions in terms of public
          reasons. As a voter, I still need to put in work to learn about
          candidate A and their platform, to get to the polls, etc. However, I’m
          not, on Rawls’s assumptions, going to do those things because I have no
          assurance that other citizens will do those things as well (and the
          other citizens won’t do it either for the same reason). So how do
          candidates that are committed to public reason get elected in the first

      • Andrew Lister

        That’s an interesting point, that the solution to the assurance problem has its own assurance problem. It’s true that we don’t have assurance that other individuals are voting for justice if political liberty includes the freedom to vote as one pleases. However, I think the duty to vote for justice isn’t conditional on reciprocity. The fact that others are voting selfishly on tax policy, say, doesn’t give me the right to do so – but if no one else is paying their taxes, I won’t either. The duty to help create institutions and to help reform unjust institutions is unilaterally binding, in Rawls (I take it), which is why it’s subject to a demandingness constraint. In contrast, the duty to obey laws is subject to reciprocity condition. Why should I obey rules I disagree with (within some limits of course) if you aren’t willing to obey rules you disagree with? Of course these normative points don’t settle the emprical question of the extent to which formal specification and centralized monitoring / enforcement can provide assurance of compliance with rules.

        I suppose what the argument you summarize above shows is that if people think that *all* duties are conditional on assurance of compliance on the part of others, then the state provides no solution, since it depends on compliance with contributive duties that themselves require assurance. That’s back to Hobbes, in a way, isn’t it? One could read Hobbes as basing his political theory on a moral view with a very wide-ranging reciprocity condition. All of the Laws of Nature are binding ‘in foro externo’ only with assurance of compliance on the part of others.

        • Blain

          Good point, Andrew.

          By way of clarification, my comment regarding public reasoning has to do with assuring citizens *within* a WOS of their ongoing commitment to maintaining a just basic structure when legislating. Justifying political decisions in terms of public reasons is one way to do this. I agree with you that the duty to *create* just institutions is not reciprocity-based (as you may recall from our discussions re. your J of Pol Phil article).

        • Chris Freiman

          Thanks for this, Andrew. My claim is slightly different from the claim that my duty not to vote on the basis of pure self-interest is conditional on reciprocity. Rather, my thought is that, by Rawls’s own assumptions about conditional cooperation, the duty to incur the costs of voting well is conditional on reciprocity. Take a simple case. I ask myself whether I should (1) do some research on the candidates, learn what experts think about their various policy proposals, drive to the polls, and vote for the best candidate or (2) sit on my couch and watch Netflix (I may or may not make a well-intentioned but uninformed vote at the last minute). It seems that a conditional cooperator will only do (1) if they have assurance that other citizens are also opting for (1) over (2). Rawls has a dilemma. If citizens will voluntarily opt for (1), then Rawls must back off of his claim that coercive state enforcement is required to ensure cooperation (since people are cooperating here without coercive state enforcement). On the other hand, if citizens will not voluntarily opt for (1), then they aren’t making the needed contributions to the state itself–in which case, Rawls is deprived of the right to stipulate that the state is going to work well (for instance, in its role as an enforcer of assurance).

          • Andrew Lister

            So it’s informed vs. non-informed, informed being costly… I will have to buy the book!

  • David Levey

    Certainly will be reading it. Jacob Levy made many of the same points, especially in relation to Rawls and Cohen, in his paper, “There’s No Such Thing as Ideal Theory.”

  • Scott

    ouch, I’m going to have to see if the library gets this one. You guys need to come up with some sort of peer publishing system for books.
    Jason I was watching the camping episode of Mickey Mouse Clubhouse with my daughter “Camping is more fun when you share!” Was that the inspiration for Mickey Mouse Clubhouse imagined as ideal theory capitalism?

  • Andrew Lister

    I look forward to reading the book, but like Blain, want to quibble with the account of Rawls implicit in Jason’s precis.

    Re: “Under ideal conditions–in which people are stipulated to comply fully with the requirements of morality and justice–there simply is no need for a state”

    The statement above makes Rawls’s ideal theory sound very ideal. Yet his theory is arguably too realistic, in certain respects. Rawls didn’t assume that people would comply with the requirements of morality and justice whatever they are, no matter how unilateral their requirements. He assumed that willingness to comply with duties of justice was conditional (for some duties, to some extent) on the willingness of others to comply. If one thinks of justice as being conditional on reciprocity, in this way, one will be led to attribute a kind of logical priority to ideal theory. In order to know what I owe others when they are *not* complying with requirements of justice, I have to first know what I would owe them if they were complying. Not all theories have this structure, e.g. utilitarianism. The debate about ideal theory is thus in part a substantive normative dispute.

    More recent research in psychology and economics does seem to have borne out Rawls’s assumption that people are motivated by reciprocity – for better and for worse (tit for tat can mean revenge as well as reciprocating kindness, though I know that the TFT strategy in Axelrod’s competition kept no grudges). Whether our ideals of justice *should* be limited by reciprocity is another question. Rawls has been accused of watering down justice for the sake of feasibility. Of course we want our institutions to be workable. But it might be important to know when these institutions fall short of true justice, realizing only the amount of justice that is currently possible given what we know about the limits of human nature.

    • Jason Brennan

      This is good quibble, but Freiman is very precise in his book and takes this into account. My standing on one foot rendition doesn’t do his argument justice.

      • Peter from Oz

        Quibble is a lovely word, innit?

  • Peter from Oz

    The left-liberal political philosophers remind me of astronomers who believed in the Ptolemaic
    universe. When new data became available that was inconsistent with the
    Ptlomeic model these astonomers didn’t question the model, they adjusted
    it by putting in side-orbits and other such tricks to show that the
    earth was in fact the centre of the universe. The fixes these
    astronomers proposed were in fact quite brilliant. They went about their work in amanner that no-one could deny was empirical. But they were wrong.
    The left-liberal idea that government derives virtue from the fact that it is designed to impose virtue on a corrupt and immoral populace is the philosophical equivalent of the science of the geocentric universe. Thus have many left-liberals spent much scholarly energy on keeping this model going whenever any new data comes to light showing that government is just as corrupt and silly as private enterprise. The thought that the lack of a profit motive must lead to puirty, without reasoning that bureaucratic empire building and the need for bigger budgets were thecivil service version of seeking profit. They don’t se rent seeking as a logical result of bigger government, but as an evil perpetrated by private enterprise.
    These left-liebrals could have saved themselves some time by reading Mandeville’s ”Parable of the Bees” which argues persuasively that it is man’s vices that have led to his greatness, not government’s virtues.

  • urstoff

    What a bargain at $140!

    • Sean II

      “For the Real World”, indeed.

  • Pingback: Rawlsian Anarchism – An Exercise in Charity()

  • Pingback: Rawls, Ideal Theory, and the Public Goods Argument for the State - Bleeding Heart Libertarians()

  • David

    Good post. Now, $140 for a physical book and $40 for an ebook? He must be kidding. I’m definitely not buying it, no matter how interesting it is.