I. Berlin's Axiomatic Monism

Berlin's characterizations of monism are scattered throughout his work. Sometimes his remarks draw sweeping generalizations about entire periods of thinkers, and sometimes they focus on specific philosophers or ethical systems. The reading I present here is implied by Berlin's clearest attempts to define monism, it accounts for his contemptuous rhetoric in criticizing monism, and it illuminates the conception of value that tacitly underpins both his account of monism and his pluralism.

According to Berlin's clearest statement, monism has three parts.Footnote ¹⁹ First, monism holds that all genuine ethical questions must have one, and only one, true answer; all the rest are errors. Second, monism holds that there must be dependable paths to discovering these true answers, which are in principle knowable even if currently unknown. Third, monism holds that these true answers are compatible with one another and form a single, harmonious whole.

Let us start by considering the first and third parts of this description. Together, they plausibly express an assumption shared by all theories of value that, in a clear sense, are “monistic.” This is the assumption that value pluralism is false. Value pluralism, in general, is the thesis that genuine normative questions can have multiple, objectively true, conflicting answers so that in particular situations an agent can be subject to multiple, objectively true, conflicting normative demands with no possibility of reconciling them. In such situations, since normative demands are irreducibly plural, in permanent conflict, and incapable of resolution, a tragic choice must be made to follow one instead of others.Footnote ²⁰ Pluralism thus asserts a certain kind of skepticism concerning right answers to normative questions. But unlike some forms of normative skepticism, pluralism does not claim that there is no objective normative truth in a particular situation, nor merely that it is difficult to know what that truth is, but rather claims that sometimes there is too much normative truth and that this truth is inconsistent with itself. Monism, on the other hand, is pluralism's contrary: it holds that normative requirements speak univocally, and that the truth about what we ought to do in any particular situation is in principle consistent. This formulation of monism leaves room for disagreement among monists about many other questions concerning, for example, the nature of value, the grounds of normative truth, and the best ways to reason toward these truths. Although all monists are at some level committed to the idea that ethical theory seeks some form of unity, coherence, and mutual support among true ethical propositions, some might seek this unity by postulating an overarching metaprinciple from which all others derive and in light of which conflicts might be resolved, others by proposing various ordering criteria for prioritizing values in conflict situations, and yet others by articulating methods for refining or limiting the scope of values in light of one another in order to approximate something like Rawlsian reflective equilibrium.Footnote ²¹ But a characteristic claim of monism, which distinguishes it from pluralism, is that conflicts among values are in principle resolvable.

Now consider the second part of Berlin's statement of monism. What does Berlin mean by saying that the monist accepts that there are dependable paths to discovering normative truths? In his preferred and most loaded slogan, Berlin calls this the faith in “final solutions,” and he claims to reject it.Footnote ²² He rejects the assumption that conflicts of value are “soluble, and soluble with finality.”Footnote ²³ He rejects the prospect of a “final harmony” in which “all riddles are solved, all contradictions reconciled.”Footnote ²⁴ He associates the search for final solutions with modernity's hubristic assumption that “conclusive solutions must always in principle be discoverable,” and that “all questions could and would one day be settled.”Footnote ²⁵ According to Berlin, the great monists—he includes, among others, Socrates, Plato, the Stoics, the authors of the Christian Gospel, Christians, Muslims, Rationalists, Empiricists, Rousseau, Kant, and Tolstoy—all share a commitment to discoverable solutions, to a single all-embracing system that can reliably answer all questions of value.Footnote ²⁶ The monist's sense of certainty about discovering these truths flows from the method of ethical inquiry Berlin thinks the monist assumes. The single-minded monist, according to Berlin, accepts that there is a single method of inquiry appropriate for all fields and through which we might discover “eternal, timeless truths, identical in all the spheres of human activity—moral and political, social and economic, scientific and artistic.”Footnote ²⁷ Underpinning this method is “some single central principle… which once found, will govern our lives,” and which can serve as a “blueprint” for essentially mechanical and deductive reasoning, a notion Berlin describes as “complex computation.”Footnote ²⁸ This master principle serves as an “infallible measuring rod” before the bar of which all problems can be brought, and which is capable of settling all questions of value.Footnote ²⁹

Berlin uses mathematical metaphors like confetti to illustrate the role of this ethical master principle in monism's account of practical reasoning. For the monist, ethical reasoning amounts to “operations which a slide rule can perform,” where values are “graded on a scale” and can be reduced to a “common denominator.”Footnote ³⁰ He portrays Descartes's geometric method of demonstration—of inferring irrefutable conclusions from axiomatic premises—as an exemplar of monism's method. Descartes's ambition was to discover a single system of knowledge embracing all areas of inquiry and answering all questions, including ethical questions. According to Berlin, these Cartesian answers “could be established by unbreakable chains of logical argument from universally valid axioms, not subject to refutation or modification by any experience of an empirical kind.”Footnote ³¹

Berlin apparently understands an ethical axiom as a discoverable, brute, incorrigible, fixed point from which all propositions within an ethical system can be derived with certainty. An ethical axiom could come in different forms and be known by different means. It could be an object of intuition, introspection, or revelation, or given by God or by nature. It could be identified with the nation, church, party, class, progress, or the material end of history.Footnote ³² But whatever the axiom's source, and although it might be difficult to discover, it is in principle discoverable with finality and certainty.

This notion of a single, ultimate, ethical axiom readily accounts for Berlin's suspicion that monism supports despotism that parades under the banner of positive liberty. Once the axiom is known, the ethical whole is then given, and practical deliberation can be reduced to technical means-end reasoning.Footnote ³³ The only rational disagreements that are then possible concern the axiom's entailments. Disagreement, uncertainty, and conflict about ultimate values are signs of confusion, vice, or mental disease.Footnote ³⁴ The despotic monist's dogmatic faith justifies coercively submerging these disagreements and conflict in the name of right reason and positive self-liberation. If the despot's reasoning from incorrigible axioms is sound, then coercion cannot be regarded as domination but is rather a form of positive liberation, for coercion which forbids someone from doing what he or she cannot rationally wish to do is not a restraint on freedom at all because it fully agrees with what that person's free, rational self truly wills.Footnote ³⁵

It is well known that Berlin endorses value pluralism as a prophylactic against monism's supposed dangers.Footnote ³⁶ Less often appreciated, however, is that the axiomatic conception of ultimate values that Berlin associates with monism also underwrites Berlin's own account of value pluralism. This will strike some as a surprising and perhaps controversial claim, but it is nevertheless one to which Berlin's pluralism is committed. Let me explain why before addressing two potential objections to this reading.

Berlin's pluralism does not merely declare that we are often uncertain about what the right solutions to value conflicts are, but rather asserts that it is sometimes certain that principled resolutions to such conflicts are impossible.Footnote ³⁷ This certainty is implied by pluralism's central idea of permanent, irresolvable—indeed, final—conflict among irreducibly plural values.Footnote ³⁸ If, in a conflict situation, we were merely uncertain whether the conflict could be resolved, then we could not reasonably conclude, as Berlin does, that such conflicts are permanent or inescapable and that values are irreducibly plural. Rather, if we were merely uncertain whether the conflict could be resolved, then we could at most conclude that such conflicts are sometimes very difficult to resolve and a source of great anguish. The pluralist, therefore, cannot justifiably maintain pluralism's characteristic claim that there certainly are multiple, permanently conflicting values unless he also assumes that the particular demands of some values can be known with certainty. Moreover, pluralism's key assertion that value conflicts are insoluble can be sustained only by assuming that conflicts cannot be resolved by refining or adjusting the scope of values in light of normatively relevant facts about a situation, or by reflecting on higher-order values against which conflicting values might be commensurated and ordered. For if, in a conflict situation, adjustment of competing values or reflection on progressively more abstract values remains in the cards, then we would not be entitled to assume that deliberation could never resolve apparent conflicts. But the impossibility of resolving value conflicts is precisely what value pluralism asserts—that is what value pluralism is. To deny the possibility of refinement or the appeal to a higher tribunal of value is just to take the conflicting values themselves to be fundamental hypotheses that play a fixed, axiomatic role in practical reasoning, and this is just to affirm that there are some fixed, ultimate values that we can reliably know. This implies that although Berlin denies the first and third features of monism, his pluralism tacitly assumes the second feature he attributes to monism, which is that some values are axiomatic and discoverable with finality.

Berlin's rhetoric about the nature of values supports this reading of his position. He treats values as if they are stubborn fixtures of reality or of our own ethical personalities. Value conflicts, he contends, are “the essence of what [values] are and what we are,” and “choices must be made for no better reason than that each value is what it is.”Footnote ³⁹ Berlin assumes that the content of these values is brute, not something that can be limited or adjusted in a way that might resolve conflicts between them. This explains why he derides as “procrustean” any attempt to reconcile conflicting values by reinterpreting their scope or meaning.Footnote ⁴⁰ Nowhere is this clearer than in his discussion of the concept of negative liberty, where Berlin's determination to use words in a certain way is most apparent.Footnote ⁴¹ He insists that “liberty is liberty, not equality or fairness or justice or culture, or human happiness or a quiet conscience,” and “nothing is gained by a confusion of terms.”Footnote ⁴² He assumes that the content of this value is fixed and that any attempt to interpret it in a way that eliminates conflict with other values amounts to gerrymandering a settled definition. For Berlin, the meaning of liberty is clearly distinct from the conditions under which it is valuable.Footnote ⁴³ We may regret, on egalitarian grounds, that a person born into poverty and who receives poor education lacks the resources and opportunities to make worthwhile life choices, but according to Berlin this person is no less free as a result.Footnote ⁴⁴ This is simply a hard fact about liberty that we have to accept even though it sometimes makes liberty seem utterly valueless.Footnote ⁴⁵ Liberty for the wolves, after all, is death for lambs.Footnote ⁴⁶

I want to address two objections to this reading of the conception of value underpinning Berlin's monism and pluralism before proceeding to Kant's alternative.Footnote ⁴⁷ First, it may be objected that this reading inaccurately takes Berlin to hold that under value pluralism there can be no rationally superior choice in any conflict situation. Yet, the objection continues, this overlooks the fact that Berlin appears to  accept the more plausible view that rational choices can indeed be made in particular contexts where general rules or principles conflict. For instance, Berlin writes:

Here Berlin says that conflicts among priority rules or principles that can neither be adjusted nor fully analyzed can nevertheless be arbitrated in light of a more abstract ambition or “general pattern of life.” The principle prescribing conduct that least disturbs that pattern is, according to Berlin, the rational principle to follow. Elsewhere, in an essay cowritten with Bernard Williams, Berlin echoes this point in remarking that we have no reason to think that judgments based on a “simple priority rule” (e.g., that justice always trumps loyalty) are any more rational than what he calls “judgments of importance” whereby one of two conflicting values is judged to be weightier than another in a particular case.Footnote ⁴⁹ Berlin and Williams conspicuously omit an explanation of what such “judgments of importance” involve beyond suggesting that we make them by attending to “context,” but it seems reasonable to infer that such judgments involve selecting whichever value or conduct least disturbs what Berlin calls one's “general pattern of life.”Footnote ⁵⁰

The conception of practical rationality Berlin articulates in these discussions is at once revealing and mysterious. If, on one hand, the concrete conflict Berlin has in mind is the type of conflict that value pluralism stresses—a permanent, irreconcilable conflict among incommensurables that necessitates a tragic choice—then it is revealing that he describes it as a conflict between principles that are “unadjustable.” As I argued above, value pluralism depends on treating the content of competing principles and values as fixed and therefore as incapable of refinement or adjustment that could release us from one of the conflicting demands; that is part of what it means to treat these principles as axioms.

On the other hand, if following whichever principle is judged to be most “important” in light of one's “pattern of life” may be, as Berlins suggests, “the right policy” in such a case, then it is mysterious whether such a case is well described as a genuine pluralist conflict at all, rather than as merely an apparent conflict from which an agent can be released through the other strategy I described above, namely by reflecting on a higher-order standard. Again, the characteristic claim of value pluralism is that some conflicts are not resolvable through such strategies precisely because in some cases there is no single common standard—no single “pattern of life”—in light of which to compare and order values. That is what it means for values to be incommensurable. It is also the reason why Berlin elsewhere states that “rationality and calculation can be applied only to means or subordinate ends, but never to ultimate ends.”Footnote ⁵¹ It is, furthermore, the reason why Berlin repeatedly insists that we are forced by the nature of ultimate values to make tragic choices among them.Footnote ⁵² The Pickwickian pluralism Berlin seems to describe here buys rationality at the cost of pluralism's distinctive features.

There is a more general point worth stressing about the relation between pluralism and practical rationality. At a minimum, a rational choice among competing values or courses of conduct is one that renders the choice both rationally intelligible and nonarbitrary. A choice is minimally rationally intelligible when it is based on considerations that explain how the chosen option is at least as likely to serve some worthwhile purpose as competing alternatives are. A choice is minimally nonarbitrary when the purpose thus served enjoys authority in guiding the choice and is not based on arbitrary factors such as brute inclination, whim, or blind prejudice. A purpose generally enjoys such authority when it itself can be located within a broader scheme of aims that renders the purpose intelligible because necessary to one's more general projects. In this way, and as we will see Kant recognizes, practical rationality strives to achieve a kind of systematicity among actions, purposes, and the wider network of projects we pursue and try to integrate within our lives. The rational agent is someone whom we at least presume to be capable of actively, though perhaps not always reflectively or successfully, integrating values at different levels of abstraction in this manner. It is not at all clear whether value pluralism, whose distinctive claim is that some values are essentially fragmented and incapable of reconciliation, is compatible with even this minimal standard of rationality. On what basis could an agent resolve a genuine, permanent, irreconcilable conflict among incommensurables other than arbitrary considerations? If the basis were not arbitrary but rather principled, then we could not assume that the conflict is permanent and irreconcilable.Footnote ⁵³

In any case, under either the revealing or the mysterious interpretation of Berlin's conception of practical rationality, Berlin's remarks seem to confirm my contention that conflict resolution requires that we keep either adjustment or higher-order reflection in the cards. Pluralism—the view that there is sometimes no resolution among incommensurables so that we must simply choose—works only by denying that either or both of these possibilities is available. In other words, pluralism works by assuming the axiomatic account of value.

The second objection I mentioned suggests, contrary to my interpretation, that Berlin does not in fact regard values as “stubborn fixtures of reality” similar to brute facts. Normative truths, says Berlin, are “not statements of fact at all, but orders, commands, ‘imperatives,’ deriving neither from an artificial convention, like mathematics, nor from the observation of the world, like empirical statements.”Footnote ⁵⁴ Instead, this objection continues, Berlin derives essential aspects of his conception of value from Kant.Footnote ⁵⁵ As Joshua Cherniss writes, “Kant, Berlin explained, held that individuals are ends in themselves ‘because they were the sole authors of moral values.’ Values existed, not in nature, but only in the wills of individuals.”Footnote ⁵⁶ On this Kantian reading of Berlin, the value of our choices derives not from anything outside of us, but from the fact that we will them. We, as individuals, are agents whose capacity for choice is itself the basic bearer of value, “the sole source of all morality, the beings for whose sakes alone whatever is worth doing is worth doing, because the notion of ends in themselves is one of the ends which men invent for themselves.”Footnote ⁵⁷

First, notice that this Kantian reading of Berlin's conception of value is at least controversial. Alex Zakaras, for instance, recently attributes to Berlin the distinctly un-Kantian view that normative values are known empirically through history and anthropology.Footnote ⁵⁸ It is not obvious how to square this reading with the view that, for Berlin, all values exist “only in the wills of individuals.” If Zakaras is correct, then Berlin not only holds an un-Kantian theory of value, but he also treats values as a species of brute fact. As I argued above, the assumption that values impinge upon us like brute facts is one way to guarantee that values conflict irreconcilably.

But even if we assume that the objection describes the best interpretation of Berlin's conception of value, it nevertheless highlights another gap between him and Kant. As we will see in section III, the Kantian view is not that an action possesses value because it is willed by an agent. Rather, an action has value—indeed fully counts as an action at all—only to the extent that it issues from the agent's own conscious representation of the action as required by principles of pure practical reason, in particular by the categorical imperative.Footnote ⁵⁹ These principles are not ends that agents author or freely “invent for themselves,” but rather state normative demands to which we must conform in order to conceive of ourselves as free authors of any ends at all. The agent who acts according to these principles must actively construct, interpret, reinterpret, revise, and attempt to reconcile his or her particular ends so that these ends comprise a coherent practical system. The Kantian agent is under a standing obligation always to be prepared to revise ends and courses of action in order to try to resolve apparent tensions among them. An agent who fails to do this is, to the extent of the failure, ethically heteronomous, not autonomous.

Moreover, as I argue below, Berlin's conception of practical freedom is fundamentally different from Kant's. For Berlin, freedom of the will does not include the capacity to self-legislate actions in accordance with principles of practical reason, but is rather merely the executive capacity to steer the hinge of one's will toward one of many conflicting possible courses of conduct.Footnote ⁶⁰ Unlike the Kantian agent, the Berlinian agent remains passive in relation to the content of the courses of conduct between which she must choose. It is in that sense—that Berlin's agent is a chooser between a plurality of, as Berlin says, “unadjustable,” permanently irreconcilable courses of conduct—that Berlin regards values as stubborn fixtures of reality.

Berlin's pluralism is axiomatic monism multiplied. Like Berlin's axiomatic monism, his pluralism does not deny the existence of discoverable normative axioms, but merely proliferates them and presses them into the breast of the individual. If his view really is that we create our values, then the values we create function in our lives as warring sovereign deities with fixed, conflicting wills between which we must tragically choose.

II. Kant's Asymptotic Monism

Kant's monism is fundamentally different from Berlin's. According to Kant, we never discover normative axioms from which to derive normative conclusions. Instead, Kant conceives of the ineradicable impulse toward unity as a working postulate, never a theoretical given. Practical reasoning presupposes that there are unique answers to any particular normative question, that practical reason ultimately speaks in one voice, not many inconsistent voices, and that practical deliberation gropes, though tentatively and uncertainly, toward those answers using different strategies for refining values in order to resolve apparent conflicts among them. But Kant's monism does not assume that this process ever achieves closure or finally comes to rest upon unshakeable normative bedrock. On the contrary, the search for the whole of value is always a work in progress, somewhat like solving a cosmic jigsaw puzzle where pieces are continually added and changing.Footnote ⁶¹

This initial characterization might strike you as an odd account of Kant. The notion of a categorical imperative alone might seem to qualify Kant as an axiomatic monist par excellence.Footnote ⁶² So too might recent interpretations of Kant's ethics that emphasize the role of “procedures” of construction that supposedly ground moral truth in Kant's theory.Footnote ⁶³ There does not seem to be much room in this familiar image of Kant for the kind of holism I am now claiming is so central to his ethics and political philosophy. But we fundamentally misunderstand Kant if we think of his system of duties as resting on a master imperative from which we can derive final solutions with certainty. Berlin himself encourages this misunderstanding by suggesting that Kant's ethics, like utilitarianism, embraces “monist decision procedures.”Footnote ⁶⁴

Although Kant's system of duties does indeed presuppose a fundamental principle (the categorical imperative) and a fundamental end (humanity as an end in itself) to which all other practical principles are in a sense subordinate, the categorical imperative serves a very different function than, say Bentham's Greatest Happiness Principle serves within classical utilitarianism. Whereas Bentham's principle instructs us to act in accordance with principles conformity to which would produce an empirically knowable state of affairs containing the largest surplus of pleasure over pain, Kant's categorical imperative instructs us to seek systematicity among the principles according to which we act for the sake of systematicity itself. As we will see, only by acting in accordance with a scheme of systematically interrelated ethical and legal principles is individual freedom possible.

To understand Kant's notion of systematicity and its relation to freedom, we must grasp Kant's general account of a rational system in both science and ethics. According to Kant, what distinguishes a system from a disordered “heap” or “mere aggregate” of particulars is that the parts of a system cooperatively relate to an organizing idea that serves as a principle of their unity.Footnote ⁶⁵ But although we often judge and understand systematic or purposive objects or events, such as human artifacts or human actions, in light of some given, fixed, material principle that organizes and makes sense of them (e.g., for a watch, its time-keeping function; for a human action, the intended end suggested by a desire or inclination), we need not understand all systems in relation to such a principle. This is because, according to Kant, the concept of a system's organizing principle includes more than a given, fixed, material end. A system's organizing principle may also be understood merely as an idea of order that we impute to the objects or events that comprise the system, and which helps guide our efforts in interpreting and systematizing our conception of these objects and events. Sometimes we find systematicity among seemingly disparate parts using only the relation among the parts themselves to one another as our guide. In this way, although we may assume, as a working postulate, the idea of a systematic whole, we may never in fact conclusively possess the actual whole or any part of it from which to derive the rest.Footnote ⁶⁶ Famous philosophical metaphors capture this notion of ungrounded systematicity or coherence. Otto Neurath compared the task of inquiry to that of sailors “who must rebuild their ship on the open sea, never able to dismantle it in dry-dock and to reconstruct it there out of the best materials.”Footnote ⁶⁷ W. V. O. Quine similarly held that it is only in the light of all of our cognitions, adjusted and understood as a systematic whole, that any particular belief should be accepted.Footnote ⁶⁸ On this view, there is no fixed, Archimedean standpoint from which we might judge or construct a systematic body of knowledge.

Kant's conception of systematicity is holistic in this way. His holism follows from his conception of the function of reason in empirical and practical thought. Reason is our highest cognitive power and its fundamental aim is to unify the concepts of the understanding into a system, arbitrating and resolving conflicts between them in order to achieve unity and lawlikeness among them.Footnote ⁶⁹ Reason's function is to transform indeterminate concepts and unprincipled distinctions into a single, coherent whole, by weaving together as much as possible in terms of as little as possible, minimizing higher-order explanatory and classificatory concepts and maximizing lower-order concepts.Footnote ⁷⁰ Unlike the rationalist monists Berlin criticizes, Kant emphasizes that this idea of unity is merely sought after; it is not an end point for inquiry, but rather an ideal toward which reason strives. Kant explicitly rejects the notion, central to Berlin's account of monism, that we can even in principle realize this unity. Kant writes that the ideal of systematic unity is something “we can only approximate and never fully attain.” It is “only a projected unity, which one must regard not as given in itself, but only as a problem,” something we realize “merely by approximation, without ever reaching them.”Footnote ⁷¹ Reason's questions never cease; we approach unity “asymptotically.”Footnote ⁷²

Kant expounds this conception of reason while presenting his metaphysics of nature, but he is clear that reason performs an analogous function in morality.Footnote ⁷³ In the Critique of Practical Reason, Kant presents the moral law through an analogy to his system of nature. He argues that practical judgment requires one to ask oneself whether, if the action one proposes to take were to take place “by a law of the system of nature of which you were yourself a part,” one could still regard it “as possible by your own will.”Footnote ⁷⁴ Here Kant is proposing that a principle of action must fit into a coherent system of practical laws in the same way that a principle of nature must fit into a system of natural laws. This analogy follows from his general view that morality and natural science comprise two grand domains of inquiry distinguished by two different kinds of objects and systems of laws to which each domain corresponds. On one hand, there are natural objects and the “laws in accordance with which everything happens” and, on the other hand, there are free actions and the laws governing them, namely the laws “in accordance with which everything ought to happen.” The former are the subject matter of “physics” or natural science, and the latter are the subject matter of “ethics” or the “doctrine of morals.”Footnote ⁷⁵

Whereas scientific reasoning seeks a systematic union of physical concepts and efficient causes and effects that comprise physical events, moral reasoning seeks a union of moral concepts and final causes, or ends, that determine a free agent's will. Just as, in constructing a system of nature, scientific reasoning does not settle for inconsistencies in empirical laws governing how things happen, practical reasoning, in constructing a system of morals, does not settle for inconsistencies in how things ought to happen.Footnote ⁷⁶ In empirical reasoning, we withhold assent to propositions that contradict our experiences or other empirical propositions our best theories affirm. If, for example, we do not have a law of optics that accounts for highway mirages, then we had better adjust the laws of convection to account for our perceptions of vanishing pools of water on highways on hot summer days. In this way, when we confront recalcitrant experiences, we inquire in order to arrive at principles of nature that square with our experience. We do not settle for absurdities, but investigate alternative hypotheses that are consistent with both old and fresh observations.Footnote ⁷⁷ Similarly, inconsistencies in potential normative principles must be settled with fresh hypotheses, distinctions, and refinements of proposed acts in particular cases. There is no given end, such as pleasure, or well-being, or god's will, or the force of history that can guide practical reason's search for unity.

Kant famously describes different ways in which conflicts might enter this ethical system. His several examples offered to illustrate the categorical imperative describe different types of “contradictions” that might arise among agents' “maxims,” which are the subjective principles according to which an agent acts or proposes to act.Footnote ⁷⁸ The moral significance of these inconsistencies is that a morally permissible maxim must be one that can serve as an objective law by virtue of its membership in a system of principles that applies consistently to all free agents. The point of morality, for Kant, is to constrain our maxims by requiring them to be inter- and intrapersonally systematic: no principle may apply to any person on some occasion or in some circumstance that cannot also apply to all in identical or relevantly similar circumstances. Immoral behavior always involves acting on considerations whose validity for others or for ourselves on different occasions we cannot rationally accept. When we apply different reasons to others, or to ourselves, on different occasions, without making principled distinctions between situations to justify the different application, we flout the characteristic demand of practical reason, which is that it must apply systematically and universally to all rational agents.

We saw in section I that Berlin maintains that the content of a value has a fixed core that must not be confused with the conditions that make the value good or useful. According to Kant's conception of value, however, to say that something is good or valuable is to say it is something we have reason to pursue or to which we have reason to respond.Footnote ⁷⁹ Once again, the analogy to science is useful. Whereas physical concepts systematically classify natural substances (such as gold) and causal explanations (such as evaporation) of what in fact happens, values and normative concepts systematically classify instances of actions that ought to happen. So whereas scientific inquiry aims to unify empirical concepts, for instance by adjusting our conceptions of refraction, sublimation, and gravitation, ethical inquiry aims to unify normative concepts by adjusting our concrete conceptions of goodness, freedom, equality, beneficence, and so on. Just as we might refine our empirical concepts in light of recalcitrant observations or experiences, we can also learn about and refine our conceptions of normative concepts in light of recalcitrant convictions we might have about what ought to be done in particular situations. This implies that the meaning of our normative concepts is no more fixed by our definitions of them than the meanings of “water” or “gravitation” are fixed by our definitions of those substances and phenomena. For example, just as the scientific discovery that water is H₂0 refined our conception of water, the emerging conviction that slavery is abominable refined our conception of justice.

Onora O'Neill, in presenting Kant's view that practical judgment requires us to “limit one maxim of duty by another,” compares practical judgment to a process of “triangulating” acts in relation to one another in order to bring maxims into a more coherent relation.Footnote ⁸⁰ Like solving a system of simultaneous equations, practical judgment must aim to take into account the constraints of multiple principles of obligation in order to try to make sure that all are satisfied. Unlike Berlin, who assumes that in conflict situations we must always do some wrong no matter what choice we make, the kind of reasoning O'Neill attributes to Kant involves finding ways to perform permissible actions that do not also involve some wrong, such as identifying ways of being truthful and honest that are not also cruel.Footnote ⁸¹ We adjust and refine possible principles of action by reflecting on our convictions about virtue and vice, good and bad, and what really ought to be done, until we come to some provisionally satisfactory understanding of what particular situations coherently demand.

III. Kant's Monism and Ethical Freedom

As we have seen, Berlin's view that monism supports despotism leans heavily on the axiomatic character he attributes to ultimate values. The pluralism that results, he suggests, mitigates the threat of despotism by declaring that these axioms are many and irreconcilable. His normative prescription—that we must value negative liberty as the unhindered freedom to choose among alternative paths—follows from this supposed fact of pluralism. Kant, by contrast, inverts Berlin's order of explanation and reaches a different conclusion. Instead of arguing from his conception of monism to the need for a liberal political order to protect our practical agency, he rather argues that we must regard values as comprising a coherent system because the value of our rational agency—our freedom—demands that we do. For Kant, monism thus follows from his commitment to individual freedom. Acting in accordance with a system of principles itself constitutes proper valuation of our freedom. Acting on that system and respecting freedom are equivalent tasks.

In order to grasp this profound insight, we must consider carefully the meaning of some familiar ideas in Kant's ethics—especially his conception of respect for humanity—and their relation to the conception of reason I have just described.Footnote ⁸² Kant held that there is an incomparable dignity in elevating ourselves above the laws of nature by the free exercise of our rational nature, or what he called our “humanity.” Despite suggestions from Berlin and his followers that Kant and Berlin share a conception of freedom as the power of “pursuing ends for their own sake by deliberate acts of choice,” Kant's conception of practical freedom is fundamentally different from Berlin's.Footnote ⁸³ The difference is crucial because each conception of freedom sponsors a very different theory of value. Berlin understands practical agency as the power to steer the hinge of one's will, and thus one's choices and actions, toward one among several possible unobstructed paths.Footnote ⁸⁴ If, as value pluralism maintains, ultimate ends conflict irreconcilably, then our agency to choose between these paths is very important because it will always be called upon to select a path in conflict situations. Kant, by contrast, locates freedom in our capacity as rational agents to act for reasons.Footnote ⁸⁵ In the Critique of Practical Reason, Kant argues that our freedom can be shown only by acknowledging the fact of our subjection to laws of reason.Footnote ⁸⁶ The free will, he writes, is “a will that is a causality inasmuch as reason contains its determining ground.”Footnote ⁸⁷ By describing the will as a “causality” Kant is not suggesting that freedom is a capacity to step outside the natural order and initiate a causal series independently of antecedent natural causes. Such a conception of the will as an uncaused cause of action results from conceiving the world of rational actions in mechanistic terms. Instead, Kant is arguing from the practical standpoint of an agent who must deliberate about how best to act. From that standpoint, we see that we regard our will as free in all our rational actions because we regard our actions as actions we consciously perform under principles of practical reason, not as movements that simply happen to us or that we passively observe. We are “noumenally” free from the causal transactions of the physical and biological world, not when, as Berlin assumes, our movements are causally indifferent as between possible future courses of conduct, but rather when our actions issue from our own rational capacities in a way that makes it appropriate to hold us responsible for those actions. The difference between what we do and what happens to us depends on our having a rational capacity to form justified beliefs about our environment, to set ends, and to rationally pursue these ends—that is, the difference depends on what Kant calls our humanity.

Valuing our freedom properly, therefore, requires valuing our humanity properly. But what is our humanity and how must we value it? By “humanity” Kant does not mean a biological characteristic of Homo sapiens. Kant often uses “rational being,” “rational nature,” and “humanity” interchangeably. In its practical role, humanity is the capacity to set, revise, and pursue ends through reason, to systematize different ends into a rational order and so to form an idea of our happiness as a whole.Footnote ⁸⁸ These capacities together comprise the basic forms of the exercise of freedom of choice and freedom of action. Kant plainly writes in the second section of the Groundwork that all rational action must aim at an end, and that if there are any universally binding principles then there must be an “objective end” that is “valid and necessary for all rational beings.”Footnote ⁸⁹ Kant postulates that this objective end simply is humanity or rational nature itself, that humanity is an end in itself whose worth is unconditioned by desire or inclination or any particular act of willing, and that humanity is necessarily an end for all rational beings.Footnote ⁹⁰ Kant held that the supreme worth of humanity, the form of our freedom, is presupposed in all rational willing. We value and stick to our plans because we already, even if only tacitly and unreflectively, have a conception of our own self-worth as rational end-setters who can freely form and pursue plans.

Kant stresses that humanity, and so also freedom, is an “existent” end, an end to be esteemed and preserved wherever it arises. He contrasts an existent end with an end to be effected or produced such as our own material welfare or happiness.Footnote ⁹¹ In contrast to an end to be effected, the end of humanity is something for the sake of which we act with no aim of producing a new state of affairs, as when we doff our hats or observe a moment of silence. Acting for the sake of humanity fundamentally involves adopting a certain attitude that appropriately expresses recognition of humanity's special status. For Kant, we adopt this attitude when we treat humanity as an end in itself rather than merely as a means to some contingent end. To treat humanity as an end in itself is to treat it as having objective and unconditional value, rather than as something that can be subordinated to or used merely in service of a particular end of merely conditional value. The imperative always to treat humanity as an end in itself enjoins us to recognize the unconditional value of freedom.

We are now in a position to isolate the relation between, on one hand, Kant's monism and, on the other hand, Kant's conception of freedom. The only way not to subordinate your own or another's freedom to set, systematize, and pursue ends is to strive always to act in accordance with a system of ethical principles. For illustration, consider Kant's own example.Footnote ⁹² Suppose I make a false promise to you even though I could not rationally permit you to make one to me if you were in my shoes. In doing so, I fail to act on a system of principles, for I am acting on a maxim that I refuse to extend to you. What could justify this exceptional treatment for myself? The most plausible difference seems to be that I am assuming that your rational agency—your capacity to recognize and make choices in light of practical reasons—matters less than mine does, and that this justifies subordinating, indeed using, your agency in service of whatever end I aim to realize through my deceptive promise. That assumption expresses disrespect for your freedom because it presupposes that your ability to recognize and respond to reasons matters less than mine does and that this justifies subordinating your agency to my contingent end.

The demand for systematicity among our principles thus serves an ideal of equal freedom: we must each strive to act in accordance with principles that afford all persons similar privileges and subject them to similar burdens. Any contradiction that might arise within or between maxims signals the impermissibility of a maxim, not because logical inconsistencies are in themselves immoral, but rather because acting on a maxim that is inconsistent with others I would apply to myself when in another's place expresses indifference or contempt for the value of that other person's freedom. This is why Kant insists that the unity reason seeks is the intrinsic means to the end of humanity; it is why “reason relates every maxim of the will” not to satisfy any contingent interest, but “rather out of the idea of the dignity of a rational being.”Footnote ⁹³ The categorical imperative to “act only on that maxim through which you can at the same time will that it should become a universal law” expresses the formal condition of respecting humanity, which is the coherence of universally applicable principles.Footnote ⁹⁴

Monism's role in what Kant calls ethical “autonomy” fits into this picture. Kant famously distinguishes autonomy from heteronomy. The latter involves the subordination of the will to principles whose ground is some contingent material end such as pleasure, or an empirical idea of happiness as the fullest satisfaction of our inclinations, or the will of someone else or of some god.Footnote ⁹⁵ In short, for Kant, the heteronomous agent is governed by Berlin's axiomatic values. The autonomous person, by contrast, treats humanity as an end in itself not merely by acting in accordance with the moral law (which could be done by accident), but by actively striving to comply with it for the sake of the moral law.Footnote ⁹⁶ The dignity of our self-governance, therefore, inheres in our capacity to act on a well-integrated system of principles for the sake of our freedom. As Kant writes, autonomy is “the principle of every human will as a will universally legislating through all its maxims.”Footnote ⁹⁷ Autonomous agents strive to act from a set of principles that comprise a harmonious, lawful whole, not from atomistic, colliding principles considered one by one in isolation.

Where does this leave us? It may seem that we are no longer in the territory of “asymptotic monism,” but rather in the grip of yet another form of axiomatic monism that is confident about its basic principles and how to apply them. After all, Kant identifies an ultimate, unconditional moral end—the freedom our humanity gives us—and tells us what respecting this end requires. We might be tempted to conclude that Kant, just as Berlin warns, endorses basic axioms and decision procedures.

We should resist this temptation. Although freedom is indeed the fundamental end of Kantian ethics, as we have seen there is a crucial sense in which it is not well described as an “end” at all, at least not in the familiar sense of a state of affairs to be effected or produced. Freedom is not an end state or destination or metric for practical reasoning. Rather, freedom reflects the distinctive problem that conscious agents must confront by virtue of their very capacity for practical reasoning. This is the problem, indeed the necessity, not merely of deciding what to do, but of deciding what is the best thing to do given our beliefs about the world and the various possible actions we may be inclined toward. We respect our freedom to the extent that we remain always prepared to revise and reorder possible ends suggested by inclination, rather than resting on brute desire, prejudice, dogma, or axiom. To respect freedom is to accept that there is no contingent end so desirable that it may be exempted from reason's inspection, revision, or abandonment.

An ideal such as this, which demands that we take no end as axiomatic, is not itself well described as an axiom. Kant's asymptotic image is more illuminating. Just as a curve may come arbitrarily close to a line as it races to infinity, asymptotic monism holds that moral actions and judgments realize the ideal of respect for freedom only by approximation. The intrinsic—indeed constitutive—means of this respect is to strive to act upon a coherent scheme of maxims. But since this ideal of systematicity can only be approached in the limit, complete knowledge and realization of the ideal is possible only for what Kant calls a “holy will,” and no human being is sufficiently knowledgeable, intelligent, or divine to achieve this.Footnote ⁹⁸ Under Kant's monism, unlike Berlin's, morality's questions never cease.

IV. Kant's Monism and Juridical Freedom

In Kant's thought, the relation between ethical monism and ethical freedom parallels the relation between legal monism and juridical or political freedom. In order to understand this parallel, we must understand Kant's distinction between rightful conduct and ethical conduct.Footnote ⁹⁹ Rightful conduct depends only on the external consistency of many agents' deeds and interactions in physical space. Since the free movements of our bodies and property can collide, we must regulate the movement of our bodies and property in a way that preserves everyone's freedom. Conduct that conforms to such regulation is rightful. Conduct can be rightful even if it is not performed autonomously from the motive of duty because it involves only the outer form of an action, namely its external consistency with other people's deeds and belongings, or what Kant calls its “legality.” Since the legality of conduct concerns only the outer form of actions, it can in principle be compelled coercively by some other person or by a state. Ethical conduct, by contrast, depends on the internal consistency of the maxims on which an agent acts. Since ethical conduct concerns the inner form of one's motives necessary for autonomous action, it may not be compelled by someone else because compulsion may only direct outer conduct. Kant divides his moral system into two subsystems that correspond to this distinction between Right and ethics.Footnote ¹⁰⁰ The first part is “the system of the doctrine of right,” which concerns duties of Right, or juridical duties, that others can establish and enforce. The second part is the “system of the doctrine of virtue (Ethica) which treats duties which cannot be so given,” namely ethical duties.Footnote ¹⁰¹

Kant introduces his Universal Principle of Right (UPR)—the most abstract principle in his system of Right—as the principle that “an action is right if it can coexist with everyone's freedom in accordance with a universal law, or if on its maxim the freedom of choice of each can coexist with everyone's freedom in accordance with universal law.”Footnote ¹⁰² There is an important analogy between the UPR and the categorical imperative (CI). Whereas the CI, as we have seen, is an ethical principle that requires us to act in accordance with maxims that comprise a unified, systematic whole that applies equally to all agents in relevantly similar circumstances, the UPR demands that external laws authorizing coercion similarly comprise a system of equal, reciprocal restraints on outer freedom so that no person's outer freedom is restricted by laws which do not similarly restrict the outer freedom of others.Footnote ¹⁰³ The UPR's demand for equal, reciprocal restraints on outer freedom thus mirrors the CI's demand that agents organize their inner maxims under universal laws that could univocally guide all agents in similar circumstances.

Although they are distinct imperatives, both the CI and UPR share a common justification in the same fundamental value of humanity—our capacity freely to set and pursue ends—and this common justification eliminates the potential for conflict between the ethical and juridical spheres.Footnote ¹⁰⁴ Both imperatives, and hence the ethical and juridical systems of duties each underwrites, arise in response to different potential incompatibilities with humanity. In the ethical case, inner freedom is threatened or disturbed when one chooses to make it his or her maxim to subordinate the moral law to some contingent end. It is precisely because of the inner, ethical character of heteronomy that external coercion may not be imposed to mitigate it; autonomy must come from within. Moreover, since only a person's acts, not his or her intentions or maxims, can restrict the outer freedom of another, then insofar as coercive laws are needed to enforce outer freedom they cannot be concerned with ethical autonomy that derives from the purity of inner motives. So, contrary to Berlin's suggestion, because of the inner character of this dimension of freedom there can be no conflict between coercive laws and inner autonomy. Moreover, the Kantian state does not parade as a means of positive self-liberation while overriding the ethical self-understanding of its subjects. Since the juridical domain, in contrast to the ethical domain, is a domain of coercively enforceable external duties, the state's actions are indifferent to the ethical motives of its subjects because the problem it aims to solve is not that the citizen is in some way morally deficient or rationally “divided against himself.”Footnote ¹⁰⁵ Kant is explicit that the problem of establishing a state must be “soluble even for a nation of devils,” because “the problem is not the moral improvement of human beings.”Footnote ¹⁰⁶ Rather, the problem is to bring the outer form of individuals' actions under a system of public laws that enables them to interact peacefully under conditions of equal juridical freedom. Kant's view on the relation between freedom and coercion simply does not support Berlin's concern, which is recently stressed by Gustavsson, that force may masquerade as a means of positive self-mastery.Footnote ¹⁰⁷ Positive ethical liberation is never coercion's legitimate purpose. “Kant's rational judge” does not send you to prison because, “like a child, a savage, an idiot, you are either not ripe for self-direction, or permanently incapable of it.”Footnote ¹⁰⁸ Rather, you are sent to prison only if your acts in some way threaten or disturb the juridical freedom of others.Footnote ¹⁰⁹

This does not mean that the Kantian state may never justify coercion on the basis of a judgment about an individual's rational capacities when that individual disagrees with the judgment. Kant is clear, for instance, that parents have an enforceable right against their children to “manage and develop” them to ensure that their capacity to understand and take responsibility for their actions fully matures.Footnote ¹¹⁰ This right does not depend on the children's consent because the state judges that children are not in a position to consent to this at a young age. Similar judgments of responsibility are ordinary fixtures of criminal law's framework of guilt and punishment, which assumes that criminal defendants must have a substantial capacity to understand their conduct and conform it to the law, and that courts can have the authority to decide whether they do. These practices are not based on an ideal of moral perfection, nor do they demand that individuals achieve ethical autonomy in Kant's sense. They also do not assume that these judgments are easy to make, and it is surely possible to complicate the picture by introducing borderline cases. Nevertheless, states must make at least some judgments of this kind. Even Berlin's theory rests on a judgment about which capacity the liberal state should be in the business of protecting, namely the capacity for choice between incommensurable values; this is why Berlinian negative liberty applies to human beings and not to beings that lack the capacity to choose.

Unlike the ethical case, in the juridical case the threat to freedom is not the inner threat of a person herself subordinating the moral law to a contingent incentive, but rather the potential of each person's power to act to be subjected to that of another person. This subjection occurs, not when someone subordinates the moral law to some contingent incentive, nor, as I describe below, when someone's acts or ends are frustrated by others, but rather when another interferes with one's capacity to set ends with one's rightful means, either by usurping one's ends through fraud or coercion, or by destroying one's means through injury to one's body or theft or destruction of one's rightful holdings.Footnote ¹¹¹ It is because of the external, juridical character of this type of domination that coercion may be deployed in order to deter or repel it. Kant explains the idea of coercive enforcement in terms of the “hindering of a hindrance to freedom”: those who coercively hinder the external freedom of others may be coercively hindered from doing so.Footnote ¹¹² Since the ideal of equal external freedom, expressed by the UPR, serves both to justify and limit the use of coercion, it cannot justify uses of coercion that diminish individual outer freedom.

In both the ethical and juridical cases, the solution to these different threats to freedom is a systematic, coherent system of either inner laws (in the ethical case) or external laws (in the juridical case). Whereas ethical laws subject the will to an impartial ethical principle—namely, the categorical imperative's demand for inner systematicity of maxims—that subordinates contingent incentives, juridical laws are needed to subject and direct external coercion to an impartial legal principle—namely, the UPR's demand for outer systematicity of actions—that can render coercion consistent with juridical freedom. What emerges is the legal analog of Kant's ethical monism, which we have already considered. Kant's legal monism, like his ethical monism, imposes on political actors who act in and on behalf of the rest of the community a duty to act from legal principles that can serve as universal laws for all. Whereas Kant's ethical monism demands that any particular principle we each embrace must cohere with our own personal ethical convictions considered as a whole, Kant's legal monism demands that officials act in the name of the community only on principles that cohere with the community's institutionally defined public political morality, a morality that is embedded within the community's legal record interpreted as a whole.

Kant's argument for a monistic legal order of this kind flows from his conception of juridical freedom. Kant defines juridical freedom as independence from having one's power to act directed by another for another's purpose.Footnote ¹¹³ This is not the Berlinian conception of negative liberty as an “area within which man can act unobstructed by others.”Footnote ¹¹⁴ Rather, a person is independent of another only if she, not someone else, determines which ends she will pursue with her rightful means. I am not unfree, on this view, if I am obstructed from stealing from or assaulting others because another's property and person are not among my rightful means. Nor am I unfree simply because another person's acts or omissions limit my options; the person who frustrates my end by purchasing the last quart of milk at the store simply changes the environment in which I must act but neither usurps my means nor sets my ends for me.Footnote ¹¹⁵ The distinctly interpersonal character of this formulation follows from the idea of a person's humanity, which we have seen is the capacity to set and pursue purposes with the means one takes oneself to have. A person's right to external freedom is one's entitlement to use one's rightful means for setting and pursuing ends consistently with other people's capacity to do the same.

Certain types of interpersonal relationships threaten juridical freedom. Since public officials claim to and usually do in fact possess the power to make coercively enforceable arrangements for citizens without their consent, the juridical relationship between state and subject poses one such threat to independence. The unavoidable, nonconsensual, and essentially coercive character of the relationship between state and subject can be made rightful only if officials making arrangements for subjects act to ensure the continuing independence of those for whom they make arrangements. In particular, the state's actions can be rightful only if authorized by an “omnilateral will,” which Kant ultimately conceives in terms of a public institutional perspective: a system of public law. In order to constrain discretionary and potentially arbitrary exercises of political power that might violate independence, Kant's conception of juridical freedom requires a legal system that places reciprocal limitations on everyone's use of freedom so that no person may bind others in ways that they themselves cannot also be bound. That requirement implies that law must authorize all political action in one voice, not a plurality of inconsistent voices, and that there must be unique legal answers to all questions that might arise concerning persons' juridical relations to one other. As Arthur Ripstein expresses this point, “the only way that any particular application of concepts governing interpersonal interaction can be consistent with the freedom of everyone is if there is a single, public interpretation, provided by a public authority authorized to speak on behalf of everyone.”Footnote ¹¹⁶

A helpful route to this conclusion is through a further parallel to Kant's ethical system.Footnote ¹¹⁷ Kant famously argues that ethical freedom has both a negative and a positive aspect. One is internally free when one's will is, negatively, not determined fundamentally by sensual drives or inclinations and, positively, is determined by the moral law's demand that we act from maxims that have the form of universal law.Footnote ¹¹⁸ Analogously, Kant holds that juridical freedom also has both a negative and positive aspect. One is juridically free when the outer form of one's power to act is, negatively, independent from domination by the unilateral choice of other persons and, positively, dependent upon the universal public lawgiving of the juridical state, which Kant calls a “condition of public lawful freedom.”Footnote ¹¹⁹ As a state of universal public lawgiving, the juridical state must strive to legislate, apply, and enforce public laws defining the private relations between individuals in a way that can be rationally accepted from every citizen's point of view. Positive dependence on public law is not, as Berlin warns, a perfectionist ideal of self-mastery that would license despotism under the rhetoric of freedom. It is rather the other side of negative juridical freedom: one is negatively free from coercive dependence on another private actor when, and only when, her entitlements to use her power to act, including her property, are determined not by any private person but by an impersonal system of public law. One cannot be negatively free without being positively dependent on public law because one's entitlement to use objects consistent with everyone else's entitlement to do the same must be determined somehow, and if it is determined by any private individual or group, rather than by law, then this creates a condition of dependence, not freedom.

Kant's Idea of the Original Contract (IOC) plays a critical role in both the construction and interpretation of a condition of public lawful freedom. The purpose of Kant's contract argument is not to suggest that the state results from an actual, voluntary agreement between private contracting parties.Footnote ¹²⁰ It is rather meant to express the fundamental normative conditions under which the exercise of public power can be consistent with individual freedom.Footnote ¹²¹ The IOC binds officials to give laws “in such a way that they could have arisen from the united will of a whole people.”Footnote ¹²² The notion of possible acceptance in this formulation is best understood as the requirement that government's actions and decisions express a coherent, consistent, and intelligible conception of equal freedom, one that everyone subjected to it has reason to endorse. For Kant, this ideal must guide all acts of legislation, judgment, and coercion not merely for identifying or assessing deficiencies in the existing legal system, but also for understanding and interpreting existing law's requirements. This means that the parts of existing legal systems must be thought of as conditioned by the ideal form of government: just as the parts of an organism can be judged to be what they are and function as they do in relation to the concept of a well-functioning version of the organism, so too government's discrete decisions and judgments must strive to conform to its internal ideal of a system of universal public law expressed in the IOC. Laws can violate the IOC, therefore, not only because they are substantively very unjust or procedurally unfair, but also if they are unsystematic, arbitrary, or otherwise inconsistent with the community's own conception of justice and fairness as expressed in their institutional history. In that way, one demand of the IOC is that governments strive for systematicity and consistency in the principles that underpin its laws in order to render the relationship between ruler and subject consistent with each person's freedom.

The demand for consistency with the community's own standards of justice and procedural fairness does not imply that these standards operate as fixed axioms within a Kantian legal system. They would serve as axioms only if their content were fixed in such a way that immunized them from revision and interpretation. Conceptions of law and democracy that identify the content of a public political will with a particular kind of axiom (an original founder's intention, or the plain meaning of a text, or a factually agreed upon basic norm, or the decision of a statistical majority, for example) do indeed furnish such fixed, merely contingent, and conflicting grounds of political institutions. Contingent axioms like these disrupt the rational unity of the system by introducing arbitrary fixed points into the community's normative character. On the Kantian view, by contrast, a truly omnilateral will is governed by laws that comprise a systematic whole, always subject to reinterpretation and reform. The conception of legal reasoning that emerges is not mechanical deduction from axiomatic premises. Apparent conflicts between the community's more concrete conceptions of freedom and equality, for instance, can be resolved by reflecting on the purposes of the entire system of Right.Footnote ¹²³ If, for example, in the context of a law prohibiting hate speech, the basic principles of equal citizenship seem to collide with freedom of expression, then any limitation on the latter right must be justified in terms of the more basic right to independence that underlies both. On this view, efforts to reconcile these principles must assume that they can be refined in light of their shared presupposition that we are free and equal. Reconciling different aspects of equal freedom involves adjusting each aspect holistically under the idea that they can be brought progressively into a unified doctrinal whole.

Kant argues that public institutions with the authority to legislate, reform, interpret, and enforce the positive public expression of equal freedom are needed in order to protect private citizens from the unilateral legislation, interpretations, and enforcement of other private citizens.Footnote ¹²⁴ But would not the very existence of such authorities simply reintroduce a kind of political axiom into Kant's system by tying the content of equal freedom to the positive—as in empirical—acts and decisions of public officials?Footnote ¹²⁵ No, it would not. For it does not follow from the fact that a public authority's judgment must replace each person's private judgment about the demands of equal freedom that what that authority actually decides can be identified or interpreted without reference to the idea of equal freedom upon whose meaning the authority is supposed to pronounce. We must not confuse the need, which Kant certainly accepts, for positive official activities to establish, interpret, and enforce law with the very different requirement that we must understand and judge those activities according to the doctrine of legal positivism, according to which the meaning of those activities is simply a matter of brute social fact. Kant's argument for a system of public Right asserts the moral need for law that is established by the empirical activities of a coercive sovereign state. But it does not follow merely from that requirement that the content of these standards is simply a matter of social fact. The need for public authorities is fully compatible with—indeed requires—a nonpositivistic conception of law which treats the products of public legislation and judgment as a dynamic system of public principles that is to be reformed and judged holistically in light of its parts and purposes.Footnote ¹²⁶

Conclusion: The “Dangers” of Monism and Pluralism

I have argued, following Kant, that treating individual freedom as an end in itself, rather than as a mere means to particular ends we happen to have, requires—just is—striving to act on a unified scheme of principles. Although we realize this unity only asymptotically, striving to constrain our actions by the demands of unity expresses a kind of impartiality and freedom from blind prejudice, arbitrary choice, and unchecked self-interest. Berlin derides the monistic aspiration to coherence as symptomatic both of “a deep and incurable metaphysical need” and of “an equally deep, and more dangerous, moral and political immaturity.”Footnote ¹²⁷ He compares “monism,” which he characterizes in the axiomatic way I have described, to a theoretical outlook that is “responsible for the slaughter of individuals on the altars of the great historical ideas.”Footnote ¹²⁸ Berlin's worries about axiomatic monistic ideologies, which fueled some of the twentieth century's worst violations of individual liberty, were understandable, especially at the time he wrote: at the end of the Second World War and the start of the Cold War. But his concerns were overstated and unfair, for Berlin insufficiently acknowledges certain political dangers of pluralism itself.

As long as states exist, the coercive imposition of some values and ways of life is unavoidable. As I have argued, value pluralism seems to have the distinctive moral disadvantage of ruling out the possibility that public decisions about values and ways of life could be made on principle, in a way that is rationally intelligible to those who do not accept them but are nevertheless coercively bound by them. That possibility alone—of rationally intelligible, rather than arbitrary normative decision-making—has enormous social and political value. Kant's asymptotic monism allows and insists on it. Pluralism implies that it is often impossible because there is often nothing substantive to argue about, nothing beyond the power of brute choice to recommend any political value or decision over any other. If Berlin's pluralism—the idea that sometimes there simply is no unique answer—were correct, then it should dissolve the burden of hard decisions; there should be no uncertainty at all, but rather perfect certainty that such problems are insoluble and continued reflection pointless. If that is the case, then pluralism can be an excuse for not confronting the hard questions Kant's asymptotic monism demands we confront. The discourse of pluralism may thereby risk distorting and inhibiting intellectually responsible thought and argument about the demands of our values, laws, and socials institutions. On this view, the idea of “moral pluralism” is not merely a conceptual confusion, but also a moral mistake.