The argument of Anarchy and Legal Order is relatively simple and straightforward. There are some things we may never reasonably do to each other, and other things we have very good reason not do. Territorial monopolists—states—do these things persistently. Their putative value as maintainers of social order might be thought to render their actions justifiable (when the prohibitions they’re violating aren’t exceptionless, so that no violation would be defensible); but, in fact, there are multiple reasons to think they’re not needed to maintain social order, and further reasons to believe they actively undermine it and cause serious harm to people’s interests. Consensual legal institutions can justly enforce law; we don’t need territorial monopolists for that. The law they enforce should leave people as free as possible and should focus on remedying injuries rather than on punishing or deterring. This kind of law will create space for the emergence of a free culture and will facilitate effective responses to such problems as poverty and workplace hierarchy. Links with English and American radicalism justify seeing this kind of anarchist approach as rooted in the socialist tradition.
Jason Brennan did not like Anarchy and Legal Order.
As his NDPR review makes abundantly clear, Brennan finds the book unappealing as a defense of anarchism. He believes key points are insufficiently defended, with assertion taking the place of argument; that the natural-law position in moral theory I adopt is both obscure and mistaken; that the book’s arguments are insufficiently independent of this position to be persuasive to anyone who does not already share it; and that the book is therefore likely to be unhelpful as a contribution to the task of making anarchism persuasive to non-anarchists. He also clearly finds the book tedious and over-long.
I am sure that Brennan is right that I am guilty at points of expressing my reactions rather than justifying them. It is important to emphasize, though, that I incorporate by reference the more elaborate defenses of others at various points. I think, for instance, that I have offered significant reasons to believe that social order does not depend on Leviathan, or that poverty relief is quite possible without the state; but I have also referred readers to much more extensive discussions of these topics.
I am inclined to agree with Brennan that a book like Michael Huemer’s The Problem of Political Authority, which features fewer theoretical assumptions than mine, might well be more accessible and persuasive to non-anarchists. And of course I understand why someone sympathetic, like Brennan, to public reason liberalism might find this advantage particularly important. But the broadly strategic concerns that seem to underlie some of Brennan’s criticism were not my primary foci. Instead, I wrote the book I did because, for over a quarter-century, I have found the new classical natural law (NCNL) theory appealing and fruitful—this is my second book on the topic—and because I wanted to spell out what a version of anarchism articulated in terms of my preferred moral theory might look like.
I am less confident that the other criticisms Brennan levels are worth accepting. I touch on some key instances in the remainder of this response.
He is doubtless correct that NCNL theory is both controversial and not the majority view among Thomists. Whether this is, on its own, a reason to criticize it is presumably another matter. Compare: libertarianism is a minority position in political theory; and bleeding-heart libertarianism of the sort Brennan endorses is a minority position among libertarians. That hardly means it’s not interesting or important. In addition, of course, various features of NCNL theory could be embraced by people who weren’t Thomists.
Obscurity isn’t Brennan’s principal charge, of course; he simply finds NCNL theory mistaken. After noting its minority status in the academy, he focuses, in particular, on its commitment to a robust form of incommensurability.
The NCNL theorists, whose position on this issue I share, deny that there is some one thing that welfare or value or well-being is. Talk about value (or well being, etc.) is a way of talking about the various particular reasons for action we actually have. To maintain that friendship, say, is an aspect of welfare is not to say that friendship realizes some independently specifiable quantity called “welfare” (or something else—happiness, pleasure, preference-satisfaction, etc.) but simply that one’s participation in friendship is one of the ways in which one’s life can go well, and that initiating or enriching a friendship or connecting with a friend is thus something it makes sense to do for its own sake.
This sort of analysis helps to explain why the NCNL theorists maintain that the various sorts of welfare or play or well being or fulfillment or flourishing are incommensurable.
Suppose friendship, or æsthetic experience, or knowledge, or bodily well being, mattered just, and to the extent that, it produced or embodied some common element—say, some sort of happiness or pleasure. In this case, the common element would provide a yardstick that would make comparative measurement possible. By contrast, absent a common element, there’s no basis for comparative measurement. We can say that friendship is valuable, or that play is valuable. But in doing so we’re just saying that play and friendship give us reasons for action, not that they do so because they produce or realize something else, something that’s really valuable. Sensory pleasure seems valuable in its own right; but knowledge isn’t valuable because of its contribution to producing or realizing sensory pleasure.
If the various aspects of welfare aren’t valuable instrumentally, then the kind of aggregation required for consequentialist calculation isn’t going to be possible. The consequentialist invites us to consider the states of affairs that might be produced by our actions (or rules, policies, character-traits, practices, institutions, etc.) and to ask, of the available options, which yields the most value (adjust the formulation as needed if you prefer a satisficing variant of consequentialism). (If you don’t think this describes consequentialism as you understand it, substitute your preferred term for this sort of approach.) This injunction obviously makes sense only if there’s a quantity to maximize. And it does not in fact seem that there is any such quantity underlying our choices: there are only multiple reasons to do various things, not some maximizable master value that underlies those reasons.
If the various ways in which lives can go well are irreducibly different from each other, approaches to practical reasoning that involve choosing among options in light of the quantum of value each produces won’t be viable. Given that operationalizing this sort of approach requires aggregation, that aggregation requires commensurability, that commensurability makes sense only if there’s a common substrate of value, and that there is no such common substrate, maximizing and similar strategies seem non-viable.
Obviously, no consequentialist supposes that, as a matter of course, we can in fact quantify the relevant values in many situations. But the consequentialist does need to assume that the notion of maximization makes sense, even if we’re not typically capable of maximizing precisely; otherwise, the consequentialist injunction seems to collapse into triviality. Thus, I want to argue, attempted justifications of intellectual property or of coercive public-goods provision that are framed in consequentialist terms are unsuccessful.
Brennan offers no arguments in favor of consequentialism, per se, but he finds the notion that comparing overall states of affairs in terms of the value they embody is impossible simply odd. He reports his intuition that one can declare some states of affairs better than others.
If there is no way of ranking states of affairs quantitatively, it still might be possible to rank them qualitatively. And of course NCNL theorists will be perfectly willing to do this in at least two ways. (i) A state of affairs involving no value at all, if such a state of affairs is conceivable, would be inferior to one involving some value, any value. (ii) As an object of choice for me, a state of affairs that includes my acting reasonably is superior to a state of affairs in which I act unreasonably.
NCNL theorists will typically be uncomfortable with focusing on states of affairs external to one’s actions as the foci of the deliberation in which one engages en route to making a decision. The only context in which questions about value actually arise are in the context of our choices. The goal ought not to be to find the most efficient means to bringing about desirable states of affairs but rather to choose reasonably. And there will certainly be reasons to choose some paths, paths directly or indirectly effecting some states of affairs, rather than others. But on the NCNL view one will reasonably exclude some paths because of what would be involved in choosing them (unreasonableness with respect to others, to oneself, or both), and one will opt for other paths because of particular goods to which they lead. In neither case will one need to rank or otherwise compare the aggregated goods embodied in the relevant states of affairs. But one will be able to rank states of affairs as objects of one’s own choice in light of the reasonableness or unreasonableness of one’s choices as embodied in those states of affairs.
In addition, of course, nothing about NCNL theory prevents someone from preferring one state of affairs to another and identifying it as clearly better in terms of her own attitudes and commitments. I need not regard a given option as superior to all of the alternatives in order to select it, but I certainly can do so.
For all I have argued, other sorts of qualitative comparisons among states of affairs might be possible. For instance: something like Jonathan Dancy’s particularism might be correct. On Dancy’s view, different sorts of factors in a given situation can be seen to be salient to action in ways quite particular to that situation. But Dancy’s particularism provides little comfort for the consequentialist: it is anti-consequentialist by intention and doesn’t lend itself to the kind of generalization required for the justification of large-scale social practices. In any case, Brennan has not invoked Dancy’s view in support of his position.
In any event, though the incommensurability thesis, and the Principle of Respect derived from it, can be seen to be at work at various points throughout the book, much of the book’s position could be worked out without any reference to it. And the other key element of NCNL theory on which I draw, the Principle of Fairness, is a version of a requirement familiar from many other moral positions, so that its frequent use hardly renders the book dependent on an obscure and especially controversial view. In fact, the Principle of Respect plays little or no role, as far as I can see, in the rationale for possessory rights articulated in Chapter 2, the elaboration in Chapter 3 of obligations to the state, of factors making for social order without the state, and of the state’s contributions to social dis-order, the discussion of legal obligation in Chapter 4, the discussion in Chapter 5 of remedies for injury (apart from the treatment of deterrence), and all of Chapters 6 (on the development of a culture of freedom) and 7 (on leftism, anti-capitalism, and socialism). I think Brennan has overstressed the book’s reliance on what he regards as an indefensible premise.
It may be that Brennan thinks that, by referring to my position as “anarcho-capitalist,” he’s simply helping to situate it for readers unfamiliar with the conversation to which it seeks to contribute. But, as his footnote reference to Markets Not Capitalism makes clear, he is perfectly aware of the work colleagues and I have done to resurrect the radically market-oriented socialism of the nineteenth-century individualist anarchists. And, of course, both Anarchy and Legal Order and Markets Not Capitalism identify multiple considerations that dispose me to disavow the “capitalist” label and, indeed, to identify my position as anti-capitalist. I am unsure why Brennan declined to give these considerations more weight or to respond directly to my arguments on these points.
Brennan doubts that I have successfully explained how a stateless society could meet the requirements of the broadly Rawlsian variety of social justice he embraces, and that I can show that the worst-off people in a stateless society would be at least as well off as those in, say, contemporary Denmark. I don’t accept the Rawlsian account of social justice, and Brennan offers no extended justification for doing so. But I do seek to explain in some detail why I believe a stateless society could deal satisfactorily with the problem of poverty.
I discuss wealth redistribution and poverty relief at considerable length. I emphasize the role of the state in both creating structural poverty and in making the vulnerabilities associated with structural poverty more significant. I treat the role of background injustice in misshaping the contemporary distribution of wealth. And I discuss a variety of mechanisms for wealth redistribution that would be appropriate in a market-oriented stateless society. I devote approximately 14,000 words to these topics.
Brennan has elsewhere rightly emphasized that offering a legal guarantee isn’t the same thing as ensuring that a desirable outcome will actually occur. He is not committed to the view that an elaborate welfare state is needed to secure the well being of the least well-off. And of course Huemer’s Problem of Political Authority, which he cites approvingly, devotes very little time to addressing the problem of poverty in a stateless society. So it’s not clear just what he would regard as a satisfactory treatment of the issue of economic vulnerability. I have offered several considerations in favor of avoiding coercive social welfare measures, along with multiple reasons to believe that institutions in a stateless society could respond effectively to the problem of poverty and that, absent various structural factors evident in contemporary society, poverty would be a significantly less severe problem without the state. Brennan is obviously to free to argue that I should regard myself as accountable to a Rawlsian standard, but he hasn’t offered any reasons why I should. And I think I’ve done the same sorts of things that Brennan himself, John Tomasi, and David Schmidtz have done in pointing out the effectiveness with which markets and civil society institutions can respond to the problem of economic vulnerability and with which markets can maximize well being and opportunities for self-authorship.
I’ve sought to defend anarchism in light of contemporary natural-law theory because I find the NCNL approach appealing and insightful, whether or not it is currently fashionable. The incommensurability thesis captures an important feature of human choice, and I believe it is well worth defending. I also believe I have offered good reasons to treat a stateless society as an attractive alternative to the state where the problem of economic vulnerability and insecurity are concerned. Brennan is obviously more than free to conclude that I’m mistaken on these points. But the arguments on offer in his review don’t suffice to convince me.
The primary exponents of the general approach I adopt have been Germain Grisez and John Finnis. Among the other prominent members of their school are Robert P. George, Joseph M. Boyle, Jr., and Chris Tollefsen. Other philosophers who have embraced similar positions (and whom, for present purposes, I treat as NCNL theorists) include Mark C. Murphy and Timothy Chappell.
For one spirited defense of the approach, see Natural Law and Practical Rationality (Cambridge: CUP 2001), by Brennan’s Georgetown colleague Mark C. Murphy.
This is obviously true of what I’m calling the Principle of Fairness, which is similar to the sort of consistency test R. M. Hare considers in Freedom and Reason, as also to various other sorts of universalizability or generalization requirements defended in diverse approaches to ethics. The NCNL theorists’ incommensurability thesis may be unique to them; but, for instance, While not a supporter of the incommensurability thesis, Bernard Williams recognizes that “goods may not be homogeneous,” acknowledges “that goods are not necessarily inter-substitutable” and maintains: “That there must be something which compensates for a finite loss is just a dogma, one which is more familiar in the traditional version to the effect that every man has his price.” Bernard Williams, “A Critique of Utilitarianism,” Utilitarianism: For and Against, by J. J. C. Smart and Williams (Cambridge: CUP 1973) 144-5. He earlier maintains that “[i]t would be idle to pretend that in many more restricted connexions we had no idea what course would lead to greater happiness” (80); but of course I should want to question the notion that happiness, per se, is what matters (as I take it Williams would in some important sense as well).