In a new series, PEA Soup is hosting book discussions in which philosophers introduce a book that influenced them or that they found important. The first post is by Daniel Viehoff, who introduces The Morality of Freedom by Joseph Raz. The comment section will be open Monday 3rd – Friday 7th of April.
Joseph Raz’s The Morality of Freedom is a densely argued book about foundational topics in political philosophy. To many readers (including the students in my classes to whom I sometimes assign small parts of it) the book seems rather forbidding: its style can be austere, and the ideas it develops are often difficult to grasp in their complexity. And yet I find myself returning to it on a regular basis.
Part of the book’s attraction is simply its philosophical range. It advances influential accounts of practical authority and personal autonomy. It develops incisive objections to a prominent neutralist interpretation of liberalism. It offers probing critiques of familiar liberal pieties about rights and equality. And it articulates distinctive views of fundamental ethical issues like wellbeing and value incommensurability. Each of these topics could be the subject of a blog post in its own right (though I doubt that I, at least, could do justice to any of them within the confines of such a post). I won’t, however, discuss any of them here in detail (though I will briefly say more about some of them in the final part of this post). Instead I want to make some more general observations about Raz’s book – about how it is written, about why its philosophical approach seems especially apposite when doing political philosophy, and finally about the distinctive version of liberalism that The Morality of Freedom advances.
1.
What has always struck me, from when I first read The Morality of Freedom as an undergraduate, is the great density of arguments and observations it contains.
Sometimes we meet someone who just seems so much more alive to what the world has to offer than most other people, and when they share with us how they view things, we too come to experience them more intensely and vividly. They may see connections and significance in paintings that others miss, and when they point them out to us, we suddenly perceive depths of which we were previously unaware. Or they may alert us to the variety of flowers and plants where we had just seen undifferentiated nature, or draw our attention to the musicality of bird songs to which we would otherwise have been deaf. Reading The Morality of Freedom offers, at least to me (and, I should say, only on some occasions), the philosophical counterpart to these experiences. It makes me see explanatory connections to which I had been blind; and I recognize structure in our normative world where I had seen none.
What makes Raz’s discussion so illuminating are two features. First, Raz has an exceedingly sharp eye for the role that normative phenomena like authority, autonomy, or rights play in our social lives. After reading his discussion, I recognize how authority, as he characterizes it, is central to my relations to my children, or my students. I feel the draw that the ideal of personal self-authorship exerts in our culture. And so on. Second, Raz offers a systematic account of the complex and pluralistic domain of reasons and values, and the proper place that authority, autonomy, or rights have within it. Correlatively, where most philosophers just take for granted the normative or theoretical background against which they develop their core argument, Raz frequently sheds further light on the structure of the domain within which the main target of his analysis is located. The result is, at least for me, a vivid sense of philosophical depth. I mean this quite literally: It is as if Raz sees where the phenomenon stands in relation to many different parts of the domain of values and reasons, and by reading his writings, I too acquire this depth-perception.
I think the forbidding character of Raz’s writings that I mentioned earlier is to some extent the flipside of Raz’s recognition that the domain of values and reasons is governed by pluralism and incommensurability; his aptitude for seeing normative phenomena’s role in our social practices; and his desire to shed light on these phenomena by locating them within the complex domain of values and reasons. There is often so much happening in Raz’s discussion that it can be disorienting. It can be difficult to know what to pay attention to. There is a frequent sense that one may have missed something important (and chances are that one has). What redeems it, for me, is precisely that, whenever I return to the book, I see further connections between the normative phenomena it discusses, and get a new grasp on how they fit together.
2.
Some of the features of Raz’s philosophical approach that I already mentioned are, I think, of special importance when doing political philosophy. The Morality of Freedom shows how we can make genuine philosophical progress when thinking about political issues. When doing political philosophy, I often feel at risk of simply restating, in so many words, political commitments that I anyway have. I worry that I do not get enough distance to these commitments, so that what I offer in their support lacks the independence necessary to provide a genuine argumentative point d’appui. One of the most distinctive features of Raz’s political philosophy is how thoroughly it is integrated with his views in practical philosophy more generally. Rather than just checking the compatibility of one political judgment with another, Raz invites us to think about the place of our political judgments within the broader structure of practical reason, the nature of wellbeing, the value of mutual respect, and so on. As a result, we make progress in political philosophy by thinking about these other issues, which are yet a step removed from, and do not simply track, political commitments and the disagreements we have about them.
A closely related feature is Raz’s willingness to doubt familiar philosophical articulations of liberal ideas (as when he rejects the commitment to government neutrality, or questions the foundational character of rights and the independent moral significance of equality). By recasting various political commitments within the broader structures of reasons and value, we come to see how what is often deemed foundational is better seen as resting on further values, and is perhaps more contingent than is often assumed. This in turn opens up space for a much more probing reflection on the details of our political morality than one encounters in many other works of contemporary political philosophy.
3.
There is finally another, more distinctly political, reason why I find myself regularly returning to The Morality of Freedom. The book (which Raz describes in the introduction as an “essay on the political morality of liberalism”) offers a distinctive way of thinking about a society committed to individual freedom. Central to it is the observation that individual freedom is a social achievement; that it requires the provision of common goods; and that the role of the government is to make people’s lives go better, by providing them with genuine choices among valuable options.
This vision of liberal political morality stands in stark contrast to the kinds of policies that have been adopted, in the decades since Raz’s book was published, by many purported proponents of individual freedom. In the United Kingdom, recent governments have (often in the name of personal responsibility and liberty) effectively starved public services, including schools, universities, the arts, etc. This is bad for many reasons (including some broadly egalitarian ones to which Raz himself is relatively unattuned). But at least one reason is precisely that it deprives citizens (at least those who lack independent wealth) of the kinds of choices that contribute to individual self-authorship. (Just consider how a student’s decision what university subject to study is very often constrained by fear for her financial future; how certain otherwise attractive career options are foreclosed because they are effectively incompatible with ever owning a house; and how having a family is off the table, not because would-be parents don’t wish to, but because they doubt they can afford it.) In the United States, similar political attitudes have similarly put pressure on an already significantly underdeveloped set of institutions for fostering the common good.
The individualistic and competitive conception of personal freedom that is very often presupposed when the value is invoked in contemporary Anglo-American political culture is, Raz’s book shows, only one possible interpretation of it. And once we think in more detail about alternative conceptions – indeed perhaps think about them with the care and insight that The Morality of Freedom exemplifies – we recognize that there are other, better alternatives worth embracing, and that our society should adopt quite different policies if it indeed values personal freedom.
That’s lovely, Daniel. Thank you!
Thanks Daniel. A really helpful and sympathetic overview. Let me ask you a question, in the spirit of your section 2 point about “Raz’s willingness to doubt familiar philosophical articulations of liberal ideas.” Is Dignity foundational for you (or if you prefer to speak for Raz, then for Raz)? I tend to take it as foundational; though I have an analysis of what it implies, I tend to take it for granted that we should treat all who have certain basic capacities, over a certain basic threshold, as beings with dignity. But I’m starting to wonder about how and whether to argue For that, especially when confronted with the idea of “honor” (as in an “honor culture).
Thanks so much for the fascinating post Daniel. Like you, I re-read parts of the Morality of Freedom pretty often, and I think that Raz’s exploration of the connections between fundamental features of our personal and social lives and political questions is inspiring. The range of the book is extraordinary, and I think that this was by far his best book. I sometimes find Raz obscure and unclear, and the arguments aren’t always made explicit. That is true of this book to some extent too, but less so than some of his other work. And the book is littered with insights and controversial and sometimes compelling ideas. Here is something that I have always found missing in the book though – how we understand the permissibility of the exercise of state power, and limits on that power, depends on the truth or falsity of basic deontic principles, such as doing and allowing and the means principle (or DDE). The Morality of Freedom has surprisingly little to say about these principles even though discussion of them was very common in moral and political philosophy in the period where Raz was developing the views that led to The Morality of Freedom. Perhaps Raz was just sceptical about them, but if he was, he doesn’t defend that view or even say much about it, and if he wasn’t they seem important in understanding the relationship between reasons for action and the exercise of state power. It seems to me a real limit of the book that it creates the impression that we can move seamlessly from thinking about reasons in everyday life to the state without thinking so much about restrictions. I wonder whether you had any thoughts about this feature of the work.
In response to Victor, I agree. In fact, the biggest weakness I found in Morality of Freedom was his treatment of rights, which was a quickly asserted and unargued reductionism (as I recall it). I came to think that was a fundamental mistake, because he assumes away the possibility that rights (like deontology) could have deep structural impact on HOW we reason about morality and politics. Thoughts on that?
Thanks, Alec and Victor, for these comments! I think they push in a similar direction, and they highlight a feature of Raz’s discussion that I too have misgivings about. But before I explain what these misgivings consists in, let me say something about Raz’s discussion of the kinds of deontic categories that both of you are drawing attention to.
In several places, Raz discusses deontic constraints and personal prerogatives. These includes his discussion of Nozick on coercion, in Chapter 6; and his discussion of Nozick’s, Nagel’s, and Williams’ objections to consequentialism, in Chapter 11. But in each of these passages, Raz’s primary concern is critical. He seeks to cast doubt on attempts to use purported deontic constraints as the basis for various familiar liberal/libertarian constraints on the use of state power. And, as part of this effort, he also casts doubt on the explanatory significance of these deontic constraints: he suggests that particular judgments ultimately depend on further accounts of people’s independent interests, and that at least some of these judgments can be explained by appeal to the value of autonomy, the importance of action- rather than outcome-reasons, etc.
I do think that this results in certain gaps in Raz’s discussion. He would have done better, for instance, if he had accepted something in the vicinity of the means principle to make sense of some judgments about coercion that he seeks to explain by appeal to personal autonomy. Specifically, Raz introduces a dimension of autonomy he calls ‘independence’ to account for the strength of our objections to coercive threats that only have minimal consequences for the range of options open to us, or our capacity to choose among them. But this invocation of independence has always seemed rather ad hoc to me. It doesn’t fit well with the official understanding of personal autonomy as self-authorship. (Why should the fact that you coercively foreclose one option but leave me with many others threaten my authorship over my own life?) Raz tries to address this by suggesting that coercion “invades autonomy not only in its consequences but also in its intentions. As such, it is normally an insult to the person’s autonomy. He is being treated as a non-autonomous agent, an animal, a baby, or an imbecile.” (MF 156) But again, if your coercive threat deprives me of an option that has no further significance for my autonomy, why should your intentionally depriving me of it amount to treating me as a non-autonomous agent (etc.)? A better explanation of what makes coercion distinctly objectionable (if it is) is, I think, that it deprives the victim of an option to which she is entitled, and does so for the sake of advancing an end to which she is not required to contribute her resources (including the option of which the coercive threat deprives her).
The most interesting question this raises, for me, is what effects such a revised account of our objection to coercion has on our understanding of legitimate political power (which I take to be Raz’s ultimate concern in the book). I don’t think it undermines Raz’s critical discussion of liberal arguments for anti-perfectionism, or his argument for the limited role that rights play in political morality. (It seems relevant here that the revised account presupposes prior entitlement claims, which are plausibly grounded in a concern for autonomy of the sort Raz outlines.)
But I do agree that it might push us to rethink some positive claims Raz is often thought to make. It seems plausible that the deontic concern with depriving us of options to which we are entitled applies not just to coercion, but also to authority. That might in turn affect what kinds of considerations matter for the justification of one person’s authority over another. In particular, that I have reason to do something, and that someone else’s having authority over me would help me better conform to that reason, might not be sufficient to justify her authority. (Why? Because, though I have reason to do it, I am not required to do it, or have to make myself or my resources available for it being done.) Some might think this requires revisions to Raz’s account of authority, and in particular his normal justification thesis. I am less certain that such revisions are needed – not because I think the justification of authority sketched a few lines up really is sufficient, but because I think Raz’s account already leaves room for recognizing that it is not.
Thanks Daniel; really interesting.
I do think that Raz’s formulation of authority needs adjusting for the reason you give. But I also wonder whether more needs to be said than Raz does about the fact that the state backs up its authority with threats, which are subject to deontic constraints.
The question about the relationship between these principles and anti-perfectionism is interesting and I think under-explored. Matthew Clayton has been exploring the idea that there might be limits on bringing up one’s children to endorse or advance certain values that one holds because doing so is tantamount to using children to serve values that, when they are autonomous agents, they would not be required to serve. And that concern might favour a kind of anti-perfectionism in education. So your objections to coercion that you think Raz doesn’t account for might also provide the basis for a challenge to the more perfectionist view found in MF. But, of course, that would need a bit of work.
Hey Daniel,
Just a quick thought about something you wrote in response to Victor and Alec. I agree that accounting for the wrongness of coercion in terms of reducing someone’s autonomy (independently conceived) doesn’t seem promising. (The late Stephen White makes this point forcefully in his Philosophy and Public Affairs paper on coercion). But you could still think that depriving someone of an option to which they are entitled somehow threatens their authorship of their life in a distinctive sort of way. In fact, that seems very plausible — people do feel, I think, that the extent to which they are ‘in control’ of their lives is highly sensitive to the impermissible interference of others. And it seems natural to mark this phenomenon with something like an ‘independence’ requirement on autonomy. It’s just that the content of the ‘independence’ requirement on autonomy is explained not by exploring the idea of life-authorship in insolation, but rather by appeal to (partly?) independent deontic constraints/entitlements, many of which are filled out conventionally, including conventions for assigning control over resources.
It’s an interesting question whether this is compatible with the move you briefly suggest, whereby these entitlements are themselves justified (in part?) by appeal to their role in securing autonomy. I suspect it is, but I’m not sure.
Hi Daniel, Victor, and Jake.
Daniel, your reply is really helpful. Before I say more, I’m going to reveal that I never made it through all of MF, so perhaps I’m just going to show my ignorance. But let me share a few thoughts about what you wrote. First, when you write about the “value of autonomy” I’m not sure about two dimensions of what that might mean. There’s the “to whom” dimension. I assume it’s not consequentialist value but value to the person. Even so, there’s the question of what is meant by “autonomy.” There’s a capacity notion: I am *able* to make choices for myself, and in particular to bring reason to bear in a competent way in doing so; there’s a value notion: I *value* being left to make such choices for myself; and there’s a status dimension: I have a kind of fundamental *right* to being left free to make my own choices. Did Raz distinguish these dimensions? My sense is that he was focused on the first two, but for me, the deontic work is done in the last: to treat me as an autonomous person has a right to be treated is not fundamentally about not insulting me with your attitude towards me (not avoiding *expressing* the belief that I’m an imbecile or being motivated by an *intention* that presupposes that I’m an imbecile); it’s about respecting my right to make choices for my own life unless I really am incompetent as an autonomous agent, and in particular about not coercing or forcing me to serve as a means for pursuing projects that I don’t value (a restriction that manifests in both limits on what can be demanded of me as an agent and what can be taken from me as a patient). Note, this means that there’s something wrong in framing the issue in terms of removing an option. That’s done all the time. Every time you buy some property from me, you remove the option I had to use it; and every time you buy from someone else, you remove my option of going to that person for permission to use it. What matters is when you take *my* property or my body, time, or energy. But that’s because control of my own property is fundamental to autonomous agency. You need a property regime, in other words, before you can make sense of coercion, respect for autonomy, etc. You can’t start with the idea of options. (I think, by the way, that I’m fundamentally agreeing with you when you write “coercion [is] distinctly objectionable (if it is) [because] it deprives the victim of an option to which she is *entitled,* and does so for the sake of advancing an end to which she is not *required* to contribute her resources.” I am just trying to link that point to the notion of autonomy as a status and the importance of a property regime for spelling out how that status works.)
I agree with Victor that the issue of legitimate threats is also key, and relevant to authority. You say that Raz leaves room for the thought that there’s more to authority than the normal justification thesis. But it’s somewhat odd for someone to write so much on authority and not make it clear just how that more deontic notion, backed up by legitimate (rightful) threats is supposed to go… or am I missing something?
Finally, I agree with Jake when he says the “‘independence’ requirement on autonomy is explained not by exploring the idea of life-authorship in insolation, but rather by appeal to (partly?) independent deontic constraints/entitlements, many of which are filled out conventionally, including conventions for assigning control over resources.” I’d only suggest that not all notions of control over resources are conventional. There’s a “natural” notion of control over “me” (my body, my time, my energy) that seems to me to be prior to convention; it’s second property that is largely conventional in the way it is acquired, held, transferred, etc. But as Rawls pointed out, legitimate expectations over income and wealth are just as binding as any other rights.
Thanks, everyone, for these really interesting responses.
Victor, I agree that the coercive threats that back up (much of) the law are important for assessing the morality of our political and legal institutions. And I also agree that such threats are subject to deontic constraints. But I am also inclined to think that there are very similar deontic constraints that apply to practical authority. If so, then switching our focus from authority to coercion might make less difference to the substantive issues we are discussing that one might have thought.
I may be less sympathetic than you are to the suggestion that a concern for deontic constraints supports anti-perfectionist conclusions. (I have, in light of our discussion here, wondered whether one reason why Raz doesn’t take deontic constraints as seriously as he perhaps should have is precisely because he too thought it might push us in this direction.) I think the deontic constraints involved in our objections to coercion presuppose that certain options or resources are mine, rather than yours, and are thus asymmetrically available to me, rather than you. There are different ways to spell out what this asymmetrical availability amounts to. My own view is that it is, at bottom, availability for the pursuit of my personal projects and goals, rather than someone else’s. Correlatively, what violates these constraints is that my resources are being used for the pursuit of someone else’s personal projects and goals; and when the resource is sufficiently closely associated with me (if it is my body, or my agency), then this amounts to me being made to serve another. But in neither case is the underlying problem that I am being made to serve ‘a value’ I don’t share, unless that just is a form of serving another person. And I don’t think that is generally the case.
Put more positively: I think the complaint that you are using what is mine can often be answered by showing that the person using our resource or option is serving us, by advancing and being guided by our personal interests. And someone can (at least in certain circumstances) serve us even if they use our resources or options in the pursuit of valuable ends we don’t share (in the sense that we don’t endorse them), and aren’t required to share, as long as it is the case that advancing these ends is in fact in our interest. (There is evidently much more to say here. I take a stab at articulating some of this in a recent article [https://doi.org/10.1111/papa.12218], and in a book that I am currently completing.)
Jake, I think I largely agree with your comment. I will add that there might be three different ways in which autonomy as self-authorship could relate to the idea that we have certain entitlements to options (or to resources that help us pursue options). First, it could be that self-authorship explains quite directly why we have an entitlement over this particular option or resource. Second, it could be that self-authorship explains why we need a system of entitlements that make certain options or resources asymmetrically to us, even though the particular assignment of options or resources depends on various considerations, including conventions. Third, one could take the view that entitlements are prior to any views of self-authorship, but that, once something counts as mine, your interfering with it threatens my autonomy simply as such. I am very sympathetic to the first two views. I am somewhat less sympathetic to the last, not because I think that there isn’t a justification for such entitlements, but because I don’t see why (assuming that justification is not grounded in autonomy, as it was on the previous two views) we should then think that the complaint we have against their violations sound in autonomy, rather than in another value. (Alec, I think that this might be a point on which you and I disagree.)
Alec, thanks for raising these questions. About autonomy: for Raz, the relevant unit here is that of an autonomous life, which is, as such, good for the person whose life it is. I think this is in fact different from each of the proposals you mention: While leading an autonomous life requires certain capacities, it amounts to more than that. While I ordinarily value making choices for myself, that is neither what autonomy consists in, nor what makes it valuable. And if we have certain fundamental rights to make our own choices, this is, for Raz, ultimately because it is good for us to lead (or have the option of leading) a life shaped through successive choices, and the latter is not exhausted by the former.
I think you are right that the status notion of autonomy is not central to, and indeed rather alien to, Raz’s account. This is not because he denies that there might be autonomy-related rights. Instead it is because his entire project is to emphasize that the concern for personal autonomy central to an attractive conception of liberalism is not, first and foremost, a source of constraints on state power. It is instead a value that state power, properly deployed, can advance rather than set back. On this picture, autonomy is not a status to be respected by the state, but an achievement to be enabled or promoted.
As I suggested above, in response to Victor, I believe one can hold on to much of this picture of autonomy and yet introduce the deontic constraints that help us make sense of certain judgments about coercion. We can do this because those judgments presuppose a certain formal structure of entitlements. But they don’t commit us to any particular view either of what grounds the formal structure, or of what grounds the particular assignment of entitlements that constitute the structure. And I think the understanding of autonomy and its value that Raz adopts can in principle be combined with other considerations (whether relatively abstract values like fairness, or relatively concrete accounts of the role that these entitlements play in structuring social interactions) to account for the formal structure. Let me concede right away, however, that I am here merely outlining what I take to be the conceptual space. Actually developing such an account, and showing it to be explanatorily adequate, would require (to borrow Victor’s phrase) a bit of work.
Daniel… rich and though-provoking reply. I especially like the emphasis on the thought that autonomy is a kind of life-long achievement. I have deep worries about that thought, but that’s for another time. For now I’ll simply thank you and say: I look forward to the other time!