THE OLDEST OF THESE ESSAYS was published forty-seven years ago, the most recent a year or two ago; they are a small but representative sample of my work over the intervening forty-five years. They possess a consistency beyond that of their authorship, but I would not wish to be tried for my life on behalf of every last sentence in every last one of them. Indeed, I would not have done when they first appeared; the object of intellectual exchange is to have good ideas reinforced and less good ideas corrected. I have also resisted the urge to rewrite them or to write a running commentary on them; everyone knows the mixed sensations provoked by encounters with our former selves and their thoughts. I have silently corrected typographical errors and adjusted some verbal infelicities—I have come to think that the use of “they” and “them” as gender-neutral singular pronouns reads better than “he or she” or “her or him”—otherwise, apart from eliminating some repetitions, I have left the texts untouched. I disavow in passing below some of what now seems incautious or wrong; but this preface is intended for the most part to explain what makes these essays part of a single intellectual project. These essays were written in response to requests from colleagues and have been scattered among different publications and different kinds of publications; the provocation for collecting them here was that former students and colleagues suggested that both the essays and their readers would benefit if some of these essays were collected in one place; I am happy to believe them, especially since it affords an opportunity to bring out their connections to one another.
Like many political theorists, I mix conceptual analysis with criticism of particular writers, and vice versa. If it were not improper to appeal to authority, I would take comfort from the fact that so much of Marx’s work was a “critique” of whatever it might be, and that he clearly felt that the best, and perhaps the only, way to articulate what he wished to say about economics and politics, and the methodological difficulties of their study, was to set his own ideas against those of his contemporaries and predecessors. John Stuart Mill, equally obviously, was another writer who thought with and against the writers whose work he discusses in the essays that make up Dissertations and Discussions or form the target of book-length works such as Auguste Comte and Positivism or the longer and less sprightly Examination of Sir William Hamilton’s Philosophy. If the concept of triangulation had not been discredited by its association with political bad faith, one could say that many of us navigate intellectually by triangulating on our discipline’s landmarks. As with navigation in general, that analogy might be the starting point for awkward questions about how and by whom the landmarks were chosen, and whether they lead us into rather than out of danger. I do not ask them here, but plainly, the tradition of political thinking with which I engage is an artifact, not a brute fact.
Intellectual traditions are never made from whole cloth: little of what our more distant forebears wrote was preserved at the time, and little of what was preserved survived thereafter. Generations of political theorists have turned some of what has survived into a canon, but the merits of their selection are far from uncontested, and my own, mildly skeptical view is that what entitles a work to canonical status is only that it turns out to be “good to think with” for a substantial time and a wide audience. But goodness is in the mind of the reader; some readers seek wisdom in a text, others seek the occasion for a rousing argument; some hope to be surprised, and others reassured. There is little more to be said, even though political theorists periodically work themselves into a lather on the subject. Political theory relies very heavily on rethinking the legacy of our predecessors, from Herodotus to Karl Marx, but it is not straightforwardly a historical discipline; it is primarily concerned with the coherence and credibility of the arguments on which it focuses and only secondarily with their causal antecedents and consequences. Historians, even historians of ideas, mostly reverse that emphasis. Political theory relies on conceptual analysis, but is not straightforwardly part of philosophy as taught in the academy; its interest in conceptual analysis is more instrumental than that of philosophy, because political philosophy is, in multiple senses, a “practical” discipline rather than a “pure” one.
There are many questions about the coherence of the project. It is often said that we engage in a “conversation” about “things political” with our departed predecessors; it is as often retorted that there is something odd about talking to the dead, who cannot answer back; and many of us have some difficulty in explaining the difference between “conversations” with silent interlocutors and talking to ourselves. One answer is that unless it is to be an exercise in ventriloquism, we must take seriously the historical identity of the writers we engage with, ensuring that their “otherness” is preserved. We must also exercise a form of self-control whose necessity goes without saying, but whose practice does not come easily. We must not credit past authors with our own favorite—or least favorite—ideas; we care about our forebears and value their insights, but they did not have us in mind when they came to them. Whatever else they were doing, they were not (then) talking to us. Necessary as it is to exercise this self-control, and hard as we should try to remember that our predecessors lived in their world and time and not in ours, it is not always easy; the more engaging a thinker, the harder to preserve the necessary distance. We say “what he must really have meant was . . .” rather than “I wish he had said . . .” usually because we do not notice that sympathetic interpretation has become rewriting.
The historical figures about whom I have written were often fascinating people; but my approach, both here and in the books I have written on Mill, Bertrand Russell, and John Dewey is not biographical as I understand the term. My focus is less on the writers’ lives than on their arguments, and the interest of those arguments lies in the way they illuminate the subjects we and their authors are concerned with. The point of getting the biographical context right is that it provides our only assurance that the arguments we explain, criticize, repair, or reject really are those that the author put forward and not figments of our imagination. Four decades ago, it was a common complaint that the “Locke” who featured in discussions of “Locke’s theory of private property” was rather distantly related to the historical John Locke; matters were worse with “Locke’s doctrine of government by consent.” The obvious way to fend off such complaints was to distinguish very sharply between the purposes that the historical John Locke may have had in mind when writing the Two Treatises and the literal meaning of what it was he wrote, no matter what purposes animated its writing. In retrospect, I think I was too quick to distinguish as sharply as I then did between “what he meant to say”—which I still take to be essentially a historical question about the author’s intentions—and “what what he said meant”—a conceptual question about the implications of the “plain meaning” of the text.
Setting aside the question whether a text ever has a plain meaning, to which the answer is a boring “yes, it is what an uncontroversial paraphrase reveals,” any argument over which there has been a good deal of dispute is unlikely to have a truly plain meaning, and a paraphrase is unlikely to be uncontroversial. I would no longer wish to draw as sharp a line as I once did between what an author meant by what he said and what what he said meant. Some sense of how the author intended his words to be taken and some sense of how the audience likely took them are needed before we can settle on the meaning of what he said, at any rate in the extended sense in which the speech act being performed is an aspect of the meaning. This does not mean that an author’s intentions are decisive in determining the content of an argument. To explain an argument is to settle, provisionally, on a view of its derivation from often barely visible assumptions, and to grasp its implications, not all of which the author will himself have perceived. Dialectical reasoning would be impossible if we perceived all the implications of what we say; reductio is the commonest device employed by critics. It is plausible that an author saw many or even most of the implications of what he wrote, but not that he could have had the foresight to see what his arguments would imply in very different conditions from those in which he lived and wrote. To take an instance in which I certainly misspoke, I would not now say that when Locke writes about property, this reveals a “bourgeois” sensibility; I doubt that the label “bourgeois” suits a British tradition more at home with such expressions as “the middle ranks” of society, and I remain puzzled about the connection in Locke’s own mind between “property” in the sense of “lives, liberties and estates” and property in the usual sense of the objects of ownership.
To think we can identify the arguments that writers in the past have put forward, and that it is worth our while to think about them as arguments, implies a limited degree of “platonizing.” That is, it implies the existence of the intentional object that, in much the same way as an author’s wishes, hopes, and fears, we identify as an author’s argument. If it we did not do so, we could hardly talk about Pythagoras’s theorem and similar entities. Indeed, we could hardly identify what we ourselves have said, are saying, and intend to say in future; oratio obliqua (indirect speech) would be impossible, and we would be confined to oratio recta (direct speech). There is much to be said about the utility of conversing with the dead, and I shall say only a little of it, and sketchily. The most obvious point is that it stretches the imagination; some people may be able to create out of whole cloth the ideas and insights that Dewey once described as our “funded intellectual capital,” but most of us must borrow heavily from the bank of the ages. To take a topic that recurs frequently here, the expressive and communicative aspects of work, it is impossible to use Marx’s ideas about species-being as a guide to the organization of a complex industrial economy. Working out why it is so is valuable, less because it is a destructive exercise than because the picture Marx paints of the way we can, in some contexts, achieve self-realization and happiness in our work for other people is so attractive. The thought that we see ourselves in our work, in what we make, and in what others make of it chimes with our reaction to gifts that others have created for us, in which we see an embodiment of their affection for us, and in which even the imperfections may be said to speak to us of them.
John Ruskin hated machine-carved statues in the restored churches of Victorian England because they frustrated that possibility: their perfection obliterated the personality of the sculptor, and the object no longer connected the spectator and the creator in the way a work of art, indeed in the way that all good work, should do. That Marx’s passion for vastly expanded productivity is at odds with this concern for individual creativity is not hard to see. But if we enter imaginatively into his youthful wish that all productive work should be expressive and communicative, we begin to see not only that it is a utopian aspiration in a modern industrial economy, as he later realized, but also why writers such as Mill, G.D.H. Cole, Dewey, and other enthusiasts for forms of industrial democracy looked for other ways of reconciling the benefits of the improved productivity brought about by modern industrial techniques with ways of giving workers a stake in the productive process beyond the instrumental stake of their wages. Unleashing their creativity elsewhere than in the productive process itself was not a utopian ambition. Nor was it a mistake to emphasize the communicative aspects of production and consumption. Even today, when aspirations to industrial democracy are at a low ebb, consumers are sensitive to the fear that their running shoes or T-shirts embody the forced labor of small children in distant countries, even if they do not much mind whether they rent their cars from a worker-owned or a more conventional enterprise. Selectively at least, consumers feel that the clothes they are wearing speak of their origins and the conditions of their creation. All these phenomena have a history, as does their reflection in the thoughts of the most interesting and imaginative political theorists of the past two centuries.
Beyond this, thinking about politics is essentially historical in a way in which thinking about physical nature, the subject of the physical sciences, is not. The physical sciences have a history, but it is irrelevant to their content. One aspect of this is obvious enough. The laws of physical nature are invariant; either we know them or we do not, but physical nature is not itself engaged in a process of refashioning the laws it obeys. We talk of molecules possessing “memory,” but that is strictly by analogy. Similarly, computers do not literally remember as we do: we remember with their assistance. This is not the proposition that all natural phenomena can be deterministically predicted; open systems defy prediction, and any natural phenomenon whose causation depends on human action is no more predictable than that human action itself. Plant biology cannot predict the fate of a tree in a rain forest encroached on by local tribes or industrial loggers. It is of the essence of the laws governing human societies that they are subject to change; the laws of the Medes and Persians may have been a byword for stability, but the empires of the Medes and Persians were overthrown and their laws became of no effect. Crucially, societies run on memory. Not only is their orderliness dependent on their members obeying laws whose validity is a question of pedigree—a matter of their having been accepted forever or having been passed by whatever local process it takes to create valid law; they rely on innumerable commitments above and beyond those encapsulated in law, and anyone intending to honor, or even to evade, those commitments must rely on individual and collective memory to know what they are.
Indeed, the identity of societies and their political institutions is very largely a matter of collective memory. As an organized political entity, a society has internal and external obligations that are binding on it only because it has a persisting identity as that society; even when a society experiences revolution and the postrevolutionary regime engages in wholesale repudiation of the obligations incurred by its prerevolutionary predecessors, the larger social entity persists, and its identity is a matter of language, culture, historical memory, and anticipation. Although a society is not a person writ large, the state that provides its order is, like other institutions, a persona ficta (an artificial person) with a life history and a capacity for making agreements and obeying norms. The French legal system regards corporate entities as personnes morales, and although they are matters of artifice, they have histories, purposes, and go through birth and death; the U.S. Supreme Court decided more than a hundred and twenty years ago that corporations were legal persons.
Nonetheless, it is only by the continuous, intellectually aware activities of human individuals that states and lesser corporate entities exist at all; without the thoughts and actions of personae non fictae there would be no such entities. This imports a degree of indeterminacy into our characterization of these entities, their goals, and their capacities; think of such questions as whether a religious group is a church or a sect. When writing about thinkers remote from ourselves in space, time, or social setting, the indeterminacy is redoubled, since we have to understand both what they wrote about and how they understood it. Was the Athenian polis the Athenian state? Did Aristotle write about “the Greek state” when he wrote about the polis? It is not infrequently said that before the seventeenth century, theorists “had no concept” of the state. What the claim means is hard to tell. One thing it has meant is that when Machiavelli speaks of the prince intending to mantenere lo stato, we should not think that he is intending to preserve the state but to hold onto his recently acquired power. It is more of a stretch to go on from that localized and simple point to suggest that when Roman writers wrote about the political system with which they were best acquainted, they were not writing about the Roman state. If that entity was not a state, it is hard to know what it was. Anthropologists who write about stateless societies would not include Rome among their number. By the same token, when we talk of city-states such as Greek poleis or the republics of medieval Italy, we appear to regard them as states. Yet we do not wish to say that when Sir John Fortescue writes about the corpus mysticum that is the “realm of England” and Hegel writes about the state that is “the march of God upon earth,” they are writing about the same thing. The fifteenth-century English state neither did nor could rest its authority on its capacity for the rational management of civil society; nor could the state that Hegel wrote about rest its authority on the thought that Frederick William III was the Lord’s anointed. Examples could be multiplied to the point of exhaustion.
In the context of the essays collected here, the three issues that raise difficult questions about the way other thinkers at other times understood their own societies are whether we can speak of liberalism before the term came into use in the nineteenth century, to what extent writers about freedom have written about the same thing, and what Greece and Rome thought about rights if they did not conceptualize subjective rights as we do—that is, rights amounting to claims that an individual may make or not make at will. These are familiar issues, much discussed elsewhere in the literature, but not much discussed by me except here. The simplest way of answering these questions here is to explain what provides the unity in what follows. I begin with my understanding of the nature of liberalism, continue to an account of the nature of freedom, offer some second thoughts on rights, property, and social justice, and end by explaining very briefly why I have valued the interlocutors on whom I have concentrated. Since these topics are inextricably intertwined with one another, some repetition is inescapable. Since they are also difficult and contentious, what follows is lamentably dogmatic for lack of space to be anything else.
Liberalism as I understand it is essentially a modern creed, but not simply a nineteenth- and twentieth-century one. Its modernity lies in the fact that it is, not in logic but in fact, an offshoot of Protestant Christianity. There is no economic or political reason why liberalism could not have arisen in Athens in the fourth century BCE, but Greek religion and ethics would have had to be very different. In other words, liberalism required a particular intellectual and moral outlook and ways of conceptualizing moral and political issues that existed in no ancient society, but it did not require any particular social, economic, or political structure. The difference is perhaps best seen in the contrast between ancient and modern notions of toleration. Ancient religion was generally less concerned with questions of belief than with ritual practice, and ritual practice was largely concerned with keeping the gods favorably inclined. In the Iliad, when Chryses implores Apollo to visit the Greeks with misfortune because the obstinate Agamemnon will not return his daughter for an enormous ransom, he begins by reminding Apollo of all the animal sacrifices he has made. He has done favors for Apollo, and now the god should return them. Greek and Roman cities set a high value on the performance of the appropriate rituals, especially sacrifice, but little on theological orthodoxy. Blasphemy was dangerous because it would lose the good will of the gods; the concept of heresy, as Thomas Hobbes pointed out, was unknown. They were tolerant in the sense of not much minding what anyone thought so long as they behaved like good citizens; Jews and Christians notoriously would not do so, refusing to sacrifice to the pagan gods or to the deified emperor, and so suffered persecution. Many ancient societies found it easy to incorporate other nations’ gods into their own pantheons less because they resisted the urge to proselytize on behalf of their own cults than because they seem to have felt no urge in the first place.
This kind of tolerance is collective and unprincipled. There is no suggestion that religious belief deserves a respect that is both grounded in and the ground of respect for the personality of the believer. One should respect priests because ill-treating them makes the gods angry; others must take their chances. “Mere” religious conviction has nothing to do with it. Religious practices should be taken seriously, and we should encourage the young to take them seriously, so that we may keep the gods favorably inclined toward us and preserve social cohesion by instilling respect for our forebears. Mockery should—mostly—be stamped on, and certainly when it amounts to blasphemy rather than Aristophanic byplay. The difference between this attitude and the core of liberalism is that modern toleration is focused on the individual; it rests on a commitment to the sanctity of the individual personality and the inviolability of the individual conscience. This thought lies at the heart of John Rawls’s liberalism and explains a great deal both of the apparatus of A Theory of Justice and of the concern of Political Liberalism to set out a constitutional scheme that can appeal to persons of very different religious or metaphysical convictions without relying on the truth of any of them.
This is obvious enough to allow me to emphasize something different. This is that we may take two very different views of what follows from a concentration on the individual; one might say that they yield the liberalism of fear and the liberalism of individual autonomy. Hobbes was not a liberal. Nonetheless, the idea of a Hobbesian liberalism is not pleonastic. Hobbes believed that authority must be absolute. He thought the belief that subjects have rights against their rulers invited anarchy. The right to private judgment in matters of religion, which was advocated by dissenters and which William Godwin later made the centerpiece of Political Justice, was anathema. Nonetheless, Hobbes was keenly conscious that individuals who were deeply committed to a particular view of their religious obligations and felt that their salvation hung on their fidelity to those obligations were vulnerable to a particular and deep kind of anxiety. A state that scrutinized their beliefs too narrowly set them to play for their salvation at cross and pile. Where the peace was not threatened, the state would enhance the happiness and diminish the anxiety of its subjects by leaving them as much freedom as possible to practice their faith as they saw fit. It is security rather than sanctity that is doing the argumentative work.
On my reading of both Hobbes and Mill, Mill is a Hobbesian liberal to the extent that he founded the defense of intellectual and social freedom in Liberty not on a doctrine of individual rights but on the promotion of utility. But he really was a liberal, as Hobbes was not, because the utility in question was that of individuals eager to exercise their own autonomous judgment in formulating a plan of life; one might say that Mill built a very un-Hobbesian superstructure on Hobbesian foundations. Since I am myself skeptical about rights-based political theories, I have always thought that a Hobbesian foundation for liberalism is as secure as any. Locke, of course, provides a much more obvious starting point, but because Locke’s conception of our inviolability as conscientious individuals with a life to live rests so squarely on the idea that we are the workmanship of God, sent here to fulfill his plans for us, it is vulnerable to anyone who thinks that we are not God’s handiwork but an interesting biological accident. It is a further consequence of this view of the genealogy of liberalism that just as “Hobbesian liberalism” is not a pleonasm even though Hobbes was no liberal, so “Augustinian liberalism” is not a pleonasm even though St. Augustine was even less a liberal than Hobbes. Hobbes, after all, thought that when men could be relied on to keep the peace, they should be left largely unconstrained. Augustine did not; they had very different views on original sin. The reason for thinking that Augustinian liberalism is not a contradiction in terms is less that Augustine is the patron saint of Protestantism, though there is much to be said for that view, than that Augustine’s emphasis on the strictly limited goods we can expect from the political order provides good reasons for states to focus on what they can do, which is to provide peace, security, and predictability in managing our earthly affairs, and to refrain from attempting to do what they cannot do, which is to guarantee our salvation. Augustine thought it right for the state to protect the church and suppress threats to its survival; but his political reasoning was—if the anachronism is permissible—Hobbesian.
Nonetheless, only when writers think of themselves as liberals does intense argument begin over the nature of liberalism, its compatibility with democratic politics, its vulnerability to mass society, and so on. The acquisition of a new vocabulary makes a great difference to what we think and how we think about it. Nonetheless, it is unduly dogmatic to insist that thinkers cannot hold a political position that they would not have articulated in the same vocabulary as we. This bears on some disputes in political theory to which I am not party: for instance, the claim of Leo Strauss and his followers that there was such a thing as “ancient liberalism.” On the view offered here, it is not a conceptual truth that there could not have been any such thing, but a factual truth that not enough of the right elements were in place for there in fact to have been any such thing. It is also true, of course, that when it became possible for politicians, commentators, and theorists to apply the term “liberal” to social and political positions, a good deal of linguistic chaos ensued. It is familiar territory that in (vulgar) American political usage, “liberal” means something not far removed from what a European would characterize as socialist or social democratic; it is less familiar territory that in everyday British political terms, it is an open question how far liberalism is committed to free-market economics and to the sanctity of private property, whereas on the European mainland, these commitments are all but definitive of liberalism.
Not only is it intellectually disreputable to try to win political arguments by stipulative definition, but it is also very unlikely to be effective. The most one can aim at is to provide a mixture of histoire raisonnée (reasoned history) and helpful definitional scaffolding. As the essay on Liberalism reprinted here suggests, I think the core of political liberalism has been defensive, and that as the term suggests, the history of liberalism is the history of a concern to protect individual liberty against a succession of threats. Religious liberty, the security of person and property, and perhaps the ability of those with no resources but their labor to organize to defend themselves against undue exploitation might all be counted among the liberties that the defense of freedom requires. By the same token, one might want to extend the threats to liberty to cases where others substitute their will for mine by means of manipulation rather than simple coercion; indeed, one might be unsure whether the psychological domination of individuals by public opinion that Mill and Alexis de Tocqueville so feared was coercive or manipulative. Mill plainly thought it was the first, but his successors have often thought it was all the more dangerous because it was the second. Indeed, Mill himself seemed to think that just because it did not feel coercive, it was all the more dangerous.
It is time to revisit the oldest of these essays. Although I do not resile from what I there say about freedom, I should add to it. What I say about the way in which what it is to be free gets its sense from the obstacles to whose absence the appellation “free” is pointing seems to me to be correct, but less bold than it should be, and can be illuminating in a political context only with a lot of fleshing out. It has seemed to me for the past thirty years that there is indeed one concept of freedom, and that it is the positive concept that Isaiah Berlin identified as equating “Am I free?” with “Am I my own master?” It has in itself no tendency to generate a taste for any or all of the horrors that Berlin was concerned to fend off, and half a century after Two Concepts of Liberty, it is no unkindness to say that Berlin generated more heat than light about the nature of liberty, and that we may wish he had not talked of two concepts, nor been so slapdash about just who thought just what. The negative concept of liberty that Berlin was defending was not a concept but a very simple theory. It amounted to the thought that for most people most of the time, being our own masters is a matter of there being nobody else who is our master; if what I do flows from what I choose to do rather than from what someone else dictates, I am to that extent free. That fact has the not irrelevant consequence that the dichotomy that the ancient world most cared about—free or slave—is indeed basic. When the Spartans fought the Persians at Thermopylae to defend their freedom, they fought to preserve the distinction between being a citizen and being a slave. The king of Persia’s subjects were slaves, not free men. Nobody has ever suggested that the Spartans were enthusiasts for the psychological, moral, and intellectual autonomy that Mill praised in Liberty; nonetheless, they fought for their freedom.
Hobbes clouded the issue, not only by objecting to the citizens of Lucca thinking that they were free in ways the subjects of the Ottoman Empire could not be, but by offering several, not obviously consistent, accounts of what freedom was. To say that a man is free when he is not hindered in doing what he has a mind to do was essential to his implausible claim that a man who swears allegiance to a conqueror when the conqueror’s sword is at his throat does so freely. It is, of course, true that he is not hindered in swearing allegiance; but it would be odd to suppose that he is not hindered from refusing by the prospect of immediate death. It may be true that if he were to wish to refuse to swear allegiance and so get himself killed, he would refuse; he would not physically be hindered in doing what he had a mind to do. More often, Hobbes seemed to think that freedom is what I am here arguing that it is. A man is free when he is master of his actions; he is master of his actions if nobody else is master of his actions; and another man can become the master of our actions—that is, have the power to dictate them—by two routes: by possessing the ability to make effective threats and, more importantly, by our having yielded ourselves subject or, in more limited spheres, taken on obligations to whomever it is. Hobbes drew some interesting conclusions from this, among them that it followed from the fact that we did not make a pact of submission with the Almighty that “the kingdom of God is got by violence.”
What mattered to Hobbes, and is worth emphasizing, is that where we have given someone else the right to determine some portion of our time and activity, we become “not free” with respect to the determination of our conduct in that sphere, but remain as free (or not) in respect of everything else as we were before. Conversely, the prisoner who is freed from jail is free, even if the society into which he is released is a dictatorship with no free speech and no room to organize politically. Again, the ex-prisoner may be married, and until he gets divorced, he is not free to contract another marriage, since he is, in Hobbesian terms, chained by his lips to his existing obligations. The political relevance of this analysis is indirect but worth insisting on. For the past two decades or so, there has been what one might unkindly describe as an outbreak of the politics of nostalgia, with commentators eager to revive an ideal of republican liberty that they contrast with what is taken to be liberal, or “let-alone,” liberty. Hobbes is the villain of the piece because of his insistence that the inhabitants of Lucca might or might not have more liberty—which is to say, immunity to the service of the republic—than their contemporaries in Constantinople. If those contemporaries by chance lived under a beneficent sultan, they would have much liberty, and if not, not. The obvious retort is that a slave with an idle master remains a slave; the subjects of the sultan had a master, and the citizens of Lucca did not.
One can admire Hobbes and still wince, since the reply is obviously correct. The crucial point is the answer of the Spartan Demaratus to Xerxes. Xerxes asked whether the Greeks would fight his enormous and invincible army, and was displeased to be told that they would fight to the death to preserve their liberty and to live under their own laws. Greeks were citizens, not subjects, and in their eyes, Xerxes’s subjects were not free men but slaves. As to why, Hobbes supplied the explanation: they were chained by the lips to the word of their ruler. Hobbes’s dismissive view of the views of the citizens of Lucca rides roughshod over distinctions that we need to preserve, and that he preserves elsewhere. There is, in the universe of Leviathan, no room for the distinction between our lawful rulers and the masters or owners of slaves, because there is no room for the idea that our rulers are subject to law; if law is the word of him that by right hath command, and if in earthly matters our rulers have a right to command all things, we are under their absolute command, and not our own masters. We may be slaves who are left to get on with our lives much as we choose almost all the time, and therefore enjoy a great deal of let-alone freedom, but in Spartan terms we are slaves. This, after all, is an implication of Hobbes’s insistence that tyranny is monarchy misliked.
Nonetheless, even if Hobbes was wrong about the incoherence of the republican view of liberty, the politics of nostalgia should be eschewed for the reasons that Benjamin Constant spelled out in La Liberté des anciens two centuries ago. The Spartans were collectively their own masters, but individually at the mercy of their fellows. There was no room for private judgment in matters of religion, no room to cultivate private and idiosyncratic tastes. What each individual possessed was his own small share of that collective liberty. Not all Greek city-states were as repressive as Sparta; because it was a commercial state, Athens allowed a great deal more room for individuality than Sparta. Nonetheless, the dichotomy on which Constant insisted was worth insisting on. And it is important to remember that Constant himself was clear that the choice was not between ancient and modern liberty, but between ancient liberty without modern liberty—which he thought that the purist republicans had hankered after during the Revolution—and ancient liberty with modern liberty. Indeed, in Constant’s view, modern liberty could not be preserved unless we took ancient liberty seriously. Unless enough politically active people thought of themselves as citizens with not only a right but also a duty to keep their rulers in check, their liberties of speech, religion, occupation, domicile, and the like would not be secure. Understanding that, modern citizens should understand what representative government can and cannot achieve; hankering after the chance to exercise in person the fraction of sovereign authority that purist republicans would allot us is mistaken, but a readiness to participate in the institutions of representative government—what we now call liberal democracy—is essential. As Mill emphasized forty years later in Representative Government, representative institutions cannot achieve self-government in any literal sense; they enable us to take securities for good government, but only if we use them as they should be used.
It is often said that the ancient world did not have our concern for individual liberty because it did not have our conception of individual rights. To dig deeply into this issue is beyond the scope of this introduction, but two things need to be said. The first is that trying to answer the question of what conception of rights the ancient world possessed runs into the difficulty already noted of describing in a non-question-begging fashion the ways in which other peoples at other times and in other places conceptualized their activities and institutions. Today, we plainly have a vocabulary in which we discuss all manner of rights; we distinguish between human rights and property rights, we argue at length about the extent of the obligations imposed by rights, distinguishing at one end of the spectrum a right that may impose no more than a duty of noninterference—as with the right to walk down the sidewalk—and at the other end of the spectrum rights that may impose elaborate duties of specific performance, as when our employer is obliged to pay us an agreed-upon salary. There is almost no simple claim about how we should conceptualize rights that will not be contradicted by someone. It seems tempting to say that nobody now holds the belief in natural rights that Jeremy Bentham savaged in the late eighteenth century; but there are many Catholic thinkers who believe that unborn children have rights under natural law, and it is hard to know what these are if not natural rights, at any rate when natural law is not identified with divine positive law. What is less obvious is what kinds of rights are grounded in that law.
This can be treated as a definitional issue of the kind that Wesley Hohfeld’s analysis of rights, powers, and immunities is well designed to handle. Indeed, I think it is best treated in this way. One quick way through the issue is to distinguish between an “objective” conception of rights, in which rights are essentially the immunities conferred by the obligations falling on the person or persons constrained by the duties imposed by law, whether positive, natural, divine, or customary, and a “subjective” conception, in which rights are powers to do or forbear at will, inhering in the possessor of such a right, and the corresponding obligations fall on others in consequence of the rights of the right holder. In the one view, rights are shadows cast by obligations, and in the other, obligations are shadows cast by rights. The rights of the unborn child are immunities to ill treatment or termination, but the unborn child has neither the physical nor the legal capacity to enforce, renounce, or modify those rights. A healthy adult’s right to life is more plausibly seen as their normative power to dispose of themselves as they choose. Modern property rights are paradigms of subjective rights. I may do as I please with what I own as a chattel or as real property. One should speak slightly more cautiously than that, of course; the fact that I have all the rights that anyone can have in a piece of property does not mean that I may do just whatever I choose with it. I may own a cow outright, but I have no right to kill it and leave the corpse to decay on the highway; it is infelicitous to think of that restriction on what I may do with the cow as a restriction of my ownership, however, because the things I may not do with my cow are things I may not do with anyone’s cow. There are certainly liabilities, distinct from restrictions on use, that can and do attach to the ownership of a cow, such as the duty to have it vaccinated against tuberculosis. An attractive way of making this point is to say that an owner has in principle all those rights of use and disposal that anyone can have, but that the incidents of property go well beyond rights alone.
Then the claim about the ancient world’s lack of our concept of rights becomes the claim that even if many of the incidents of ownership that the modern world recognizes were familiar in the ancient world, as they certainly were, the conception of ownership and the allied conception of rights were closer to an objective conception than a subjective conception. That is, what was absent was not so much property rights narrowly construed as the kind of individual rights enshrined in the American Declaration of Independence; and from that it follows that there was no room for the thought that property in external things was an extension of our rights in and over ourselves. John Locke originated something. The discussion of slavery in ancient authors marks this difference quite neatly. Even Aristotle, who seemed not to object to “man hunting”—that is, forcibly capturing slaves—as an economic activity, thought that slavery was a bad thing in itself; if plows could be set to work on their own and without human intervention, there would be no need of slaves. Since that was a fantasy, slaves were needed, and the only constraint he took seriously was that Greeks should not enslave Greeks.
In the seventeenth century, when Hugo Grotius defended the thought that a man may sell himself into slavery to secure his survival—no more alarming than Hobbes’s picture of the way we submit to a sovereign by acquisition—the starting point was self-ownership. It is by an act of my own will that I render myself subject or sell myself into slavery. That idea might seem to be at least latent in the Emperor Justinian’s explanation that he possessed absolute authority over the inhabitants of the Roman Empire because the people, who had formerly possessed it, had transferred it to the emperor. The crucial difference, perhaps, is that the authority that had been passed lock, stock, and barrel to the emperor was the property of the people—a collective, and not an individual, possession. What was lacking in the ancient world was the notion of individual sovereignty, or self-ownership, and especially its subsequent liberal, non-Grotian extension: our ownership of ourselves excludes the possibility not only of anyone else owning us, but even of our selling ourselves into servitude.
That perspective explains much of what appears here. Let me now say something about property, work, and social justice, then a very little about the particular authors with whom I have wrestled. Rights of all sorts, but property rights perhaps most intriguingly, demonstrate the human capacity to expand our control over the world and its potential usefulness by organizing ways in which we can guarantee to one another that something we wish to happen in the future will in fact happen. A paradox lies at the heart of the importance of rights of all sorts, particularly property rights: because we are able to generate obligations, we can extend the possibilities of action by giving others the capacity to constrain what we do under determinate conditions. Property rights, of course, go beyond the rights over our future conduct that we can create by promising and the like, since they are rights good against the world and inherent in whoever happens to own whatever it is. As several of these essays suggest, I am not persuaded that the concept of self-ownership illuminates our right to dispose of our own selves and capacities, nor that it provides an explanation of the importance of respecting property rights. A broadly utilitarian perspective is adequate to justify property rights of most sorts, and a proper attention to such matters as the irreplaceability of some sorts of property—a much loved house, for instance—and the ready replaceability of others—insurance contracts, say—illuminates much of what we intuitively feel about the importance of some sorts of ownership and indeed explains why we sometimes do and sometimes do not think that monetary compensation is sufficient for the loss of property.
But a broadly utilitarian justification does not appease two sorts of anxiety. The first is the sense that arrangements that may make very good sense by increasing overall welfare may make much less sense in the way the gains are distributed. There are evidently two ways in which this may be true. One is that the benefit accruing from property rights may simply reflect luck, as they plainly do in the case of holding a winning lottery ticket, and slightly less obviously in the case of being the first person to secure a patent, happening to own a piece of land with oil beneath it, or perhaps “owning” whatever talents and traits of character we happen to have. Whatever doubts I express in passing about the work of John Rawls, I share his doubts about what one might call the “depth” of our ownership of these benefits. The second is the familiar terrain of political philosophy from ancient Greece onward. The benefits of ownership tend to be cumulative; short of abolishing the family, it is extremely difficult to think of ways of preventing the owners of the advantages accrued in one generation from loading the dice in favor of their children in the next.
The problem this presents is in part that loading the dice undermines one of the most important modern conceptions of desert. Broadly speaking, the modern world holds that work, effort, and productivity form the most acceptable basis of a claim to reward, and that harder work merits greater reward—always ceteris paribus, of course. The strenuous production of harmful goods, such as attractive narcotics to be sold to schoolchildren, would merit a stretch in jail. Inherited inequality subverts this justification. But I have always been interested in a somewhat different view of the connection between work, reward, and ownership, one less directly tied to the usual concern with social justice and rather more directly tied to a more metaphysical or, more guardedly, more psychological notion of ownership. This is the idea that work provides an outlet for our need for self-expression and that it enables us to “own” the world in the sense of seeing the humanly created and modified world as our product, either individual, for which the obvious paradigm is the work of art, or collective, for which there is no obvious paradigm, though it is not implausible that we may, under favorable conditions, see the world around us as reflecting our talent, ingenuity, and capacity for cooperation. It is this that I call the “romantic” conception of ownership and tie to my view that an important and underexplored—though hardly unacknowledged—aspect of social explanation is this expressive dimension.
Last, then, the thinkers whom it has been good to think with. To a degree that surprises me, I have little sense of having had much control over this. Mill, as I said in the preface to my first book-length account of his importance, was one of the two thinkers who stiffened my teenage resolve to think for myself, and neither to be merely enraged by the prejudices of my teachers nor to give in to them; the other was Bertrand Russell. Because my interests always lay in political theory, Mill provided more sustenance—indeed, enough for a lifetime. Behind Mill as I read him stands Hobbes, the begetter of analytical political theory and an inexhaustible source of puzzlement. They offer opposite attractions. Mill’s allegiances are the more congenial, even if he was less liberal in temperament than one might wish. Some of Hobbes’s professed allegiances are hard to warm to, but he was liberal by temperament, or easygoing where his amour propre and physical safety were not at stake. Both, of course, repay almost any amount of engrossed poring over the details of what they said; it is almost impossible either to abandon the search for the utilitarian theory of rights that Mill was seeking when he wrote Liberty or to believe that we can square the circle as he hoped. By the same token, Hobbes’s attempt to give a coherent account of the inner lives of creatures who are fundamentally self-maintaining automata remains rivetingly interesting.
Mill relied heavily, perhaps too heavily, on Tocqueville for his view of where an increasingly democratic society and polity were heading. But he did not follow Tocqueville all the way either in his political enthusiasms or in his economic anxieties; nor was Tocqueville particularly attached to Mill’s enthusiasm for a liberal ethic of self-fashioning. Nonetheless, Tocqueville is not easily going to be evicted from the liberal pantheon, and certainly not by me. He provides a counterpoise to the “philosophical” mode of argument to which Mill was attached. Tocqueville was sometimes accused of having gone to America knowing already what he was going to see and what he was going to say about it. This is unfair, but there was a French view, articulated perhaps most clearly by François Guizot but widely accepted, that European society had been becoming steadily more egalitarian for many years and that it was a process that might be steered but not stopped. In that view, America was the society that embodied “modernity.” In Europe, the increasing “equality of condition” by which Tocqueville was impressed was emerging from societies that had previously been governed aristocratically and whose economies had been feudal; in America, these ancien régime drags on progress were absent. Although Dewey had nothing of the literary charm that Tocqueville possessed in such abundance, I have found him a fascinating interlocutor because he essentially accepted Tocqueville’s identification of America with the condition of modernity, though I do not think he ever mentions him.
There are many scarcely visible connections between the thinkers to whom I have been most attracted; some are curious. For instance, Sidney Hook saw Dewey’s pragmatism as the philosophy, or post-philosophy, implicit in Marx’s more humanistic writings, and even though Dewey himself resisted the thought, Hook was right. In the same vein, Dewey’s identification of modernity, democracy, and rationality underlies Rawls’s social theory in spite of the world of difference between Rawls’s architectural style of argument and Dewey’s more meandering prose. Sometimes it is hard to resist the temptation to wish that history had been otherwise, and that connections made that were not. Mill’s stepdaughter Helen became friends with Karl Marx’s daughter Eleanor; one wishes that stepfather and father could have worked through Kapital together. Mill and Tocqueville should have spent time in each other’s company; and as long as the better rather than the worse characteristics of each had rubbed off on the other, it would have been a good thing if Russell and Dewey had got on better. Dewey’s patience with the intractability of people and politics would have done Russell good, just as Russell’s crispness would have done Dewey’s literary reputation good. This is to stray self-indulgently, but it allows me to end by saying that I cannot imagine political theory as a discipline that does not encourage the idea of a conversation across the centuries between imaginary friends who might form a revolutionary vanguard or an anxious rearguard, might often find each other exasperating or evasive, but would rarely find each other less than humanly interesting.