Kant’s philosophy focuses on the power and limits of reason. Twoquestions are central. In his theoretical philosophy, Kant askswhether reasoning can give us metaphysical knowledge. In particular,can reason ground insights that go “beyond”(meta) the physical world, as “rationalist”philosophers such as Leibniz and Descartes claimed? In his practicalphilosophy, Kant asks whether reason can guide action and justifymoral principles. “Empiricist” philosophers claimed thatonly feelings can motivate us to act; reason cannot. In Hume’sfamous words: “Reason is wholly inactive, and can never be thesource of so active a principle as conscience, or a sense ofmorals” (Treatise, 3.1.1.11).

Against rationalist metaphysics, Kant claims that reasoning facesstrict limits. Reason cannot give us knowledge of God or a worldbeyond the senses; reasoning falls into contradiction and confusion ifit does not respect these boundaries.

Against the empiricist account of motivation and morality, Kant arguesthat reason has a vital power. Reason enables us to act on principlesthat we can share with other rational beings. In a world of limits,reason reveals human freedom.^[1]

This entry has the following structure. The first section sets out therole that reason plays in Kant’s account of knowledge andmetaphysics. This focuses on theCritique of Pure Reason or“firstCritique” (1781, second edition 1787). Thesecond section examines his moral philosophy. This focuses on hisCritique of Practical Reason or “secondCritique” (1788). Reflecting Kant’s own works andmost of the secondary literature, these two sections are relativelyindependent. The third section therefore considers the relationsbetween theoretical and practical reason. Since Kant rarely discussesreason as a whole, his general view is a matter of interpretation:Onora O’Neill has given the most prominent account. Theconcluding remarks stress the philosophical interest of her unifiedinterpretation.

• 1. Theoretical reason: reason’s cognitive role and limitations
  • 1.1 Reason and empirical truth
  • 1.2 Reason in science
  • 1.3 The limits of reason
  • 1.4 Reason’s self-knowledge
• 2. Practical reason: morality and the primacy of pure practical reason
  • 2.1 Freedom implies moral constraint: the Categorical Imperative
  • 2.2 Moral constraint implies freedom: Kant’s “fact of reason”
  • 2.3 The primacy of (pure) practical reason
• 3. The unity of theoretical and practical reason
  • 3.1 Reason’s “common principle”
  • 3.2 The maxims of reason
  • 3.3 The public use of reason
• 4. Concluding remarks
• Bibliography
  • Primary sources
  • Secondary literature
• Academic Tools
• Other Internet Resources
• Related Entries

1. Theoretical reason: reason’s cognitive role and limitations

In the first half of theCritique of Pure Reason, Kant arguesthat we obtain substantive knowledge of the world through twocapacities: sensibility and understanding. Empirical judgments dependonboth sensory experienceand concept formation.Kant emphasizes the solidity of empirical knowledge gained thisway.

In the next large section, Kant discusses “theoreticalreason”—especially philosophical reasoning. The“Transcendental Dialectic” attacks philosophical effortsto gain knowledge of a “transcendent” world, that is, aworld beyond that revealed by the senses. “Dialectic,”says Kant, is “alogic of illusion” (A293/B349).Here, reason appears mainly as a source of empty or false ideas.

Many readers fall away before the final section of the book, whereKant reviews his overall method—his own use of “purereason.” The closing “Transcendental Doctrine ofMethod” considers reason in terms of its“discipline,” “canon,”“architectonic,” and “history.”

This makes it easy to interpret theCritique of Pure Reasonas simply critical of reason—as dismissing its claims to give usknowledge. But if this were all that Kant intended, how should weunderstand his own philosophical reasoning? Kant certainly wants toshow the limits of reason. But he also needs to show how reason canplay a constructive role. He does this in at least three ways: herelates reason to empirical truth (§1.1 below); he explores reason’s role in scientific enquiry (§1.2 below); and he explains the benefits of appreciating reason’slimits (§1.3 below).^[2]

In addition, when we engage in philosophical reasoning, we ought tounderstand the capacity that we are using. As Kant puts it, reasonmust take on “the most difficult of all its tasks, namely, thatof self-knowledge” (Axi). The firstCritique beginsthis task, but does not complete it (§1.4 below).

1.1 Reason and empirical truth

Reason has an important role in our pursuit of knowledge. Kant writes:“the law of reason to seek unity is necessary, since without itwe would have no reason, and without that, no coherent use of theunderstanding, and, lacking that, no sufficient mark of empiricaltruth…” (A651/B679).^[3] Kant says relatively little to develop this claim, and the issue hasnot attracted much commentary. (But cf. Walker 1989: Ch. 4; Guyer andWalker 1990;Kant’s theory of judgment, §1.3.)^[4]

However, the basic idea is clear from his text. We form judgmentsabout the world around us all the time. We see a hand in front of usand judge it to exist; after a dream, we judge ourselves to have beendreaming and the dream’s contents to be illusory; we see the sunrise and assume that it orbits the earth. (Obviously this lastjudgment is false: see§1.2 below.) Kant devotes much effort to show how all these judgments relyoncategories. Fundamental concepts such as cause and effectstructure all our judgments. A belief that conforms to thesecategories meets the “formal” conditions of truth.However, unless we are fundamentally confused about something, all ourbeliefs meet these conditions.^[5] So how do we decide which of our beliefs are true, and whichmistaken?

Kant begins by noting that onlyjudgments can be mistaken:“It is correctly said that the senses do not err; yet notbecause they always judge correctly, but because they do not judge atall” (A293). For example, there is no error involved indreaming, however incoherent or fantastical the dream may be. But ifsomeone got confused and supposed that a dreamed-of event had reallyhappened, then she would be making a judgment—a false one. SoKant claims, “error is only effected through the unnoticedinfluence of sensibility on understanding, through which it happensthat the subjective grounds of the judgment join with the objectiveones” (A294). In the example, someone confuses a subjectiveground of judgment (“I had this dream”) with anobjective one (“these events took place”). AsKant puts it in theProlegomena:

The difference between truth and dream… is not decided throughthe quality of the representations that are referred to objects, forthey are the same in both, but through their connection according torules that determine the combination of representations in the conceptof an object, and how far they can or cannot stand together in oneexperience. (4:290)

How does reasoning help here? In the famous “Refutation ofIdealism” (added to the second edition of theCritique ofPure Reason), Kant writes: “Whether this or that putativeexperience is not mere imagination [or dream or delusion, etc.] mustbe ascertained according to its particular determinations and throughits coherence with the criteria of all actual experience”(B279).

To see what Kant means, consider a simple example. Suppose that ourdreamer believes they have won a lottery, but then starts to doubtthis belief. To decide whether it is true, they must ask how far sucha belief connects up with their other judgments, and those of other people.^[6] If it fails to connect up (they check the winning numbers, perhaps,and find no match with the actual ticket), the dreamer must concludethat the belief was false.

We can also see the logic here in terms of Kant’s claim thatreason is “the faculty of the unity of the rules ofunderstanding under principles” (A302/B359). One fundamentalprinciple of experience is that we inhabit a single world in space andtime. Therefore, all true judgments must find a place within a single,unified experience of the world. Reason seeks consistency. It helpsseparate factual mistakes from genuine knowledge by referring to thisunifying condition.

1.2 Reason in science

The principle of reasoned unity also applies to scientific judgmentsand theories. However, it takes a more complex form, since scienceseeks universal laws.

Kant claims that reason is “the origin of certain concepts andprinciples” (A299/B355) independent from those of sensibilityand understanding. Kant refers to these as “transcendentalideas” (A311/B368) or “ideas of [pure] reason”(A669/B697). He also defines reason as a “faculty ofprinciples” (A299/B356). But what are these concepts andprinciples, and can they be justified? Since Kant also argues that“ideas of reason” often lead to error and contradiction,these are serious questions.

One sort of error arises when we claim knowledge of objects beyondsensory experience, such as God or the soul. Another sort arises whenwe form “transcendental ideas.” We may try toconceptualize the ultimate basis of everything that exists, such asthe universe as a whole. Kant refers to these as “worldwholes” or cosmological ideas. As discussed in a moment (§1.3), when we claim objective knowledge about these, we fall intocontradictions or “antinomies.” For example, Kant holdsthat we can argue equally well for the opposing claims: that theuniverse has a beginning in timeand that it does not.

Scientific enquiry assumes that the world forms a well-ordered,systematic unity where all events can be subsumed under causal laws.This is not just the idea that events have causes—an inescapableassumption that Kant refers to as a category of the understanding.Rather, the claim is that universal laws underlie all particularrelations of cause and effect. This is what Kant has in mind when hespeaks of the “unity of reason” in the firstCritique. (See e.g. A302/B359, A665/B693, A680/B780.)

As an illustration, consider Copernicus’s hypothesis that theearth orbits the sun. This hypothesis contradicts our everydayperceptions—perceptions that we normally rely on withoutquestion. Historically, Galileo’s observations were a turningpoint. The telescope gave a much more extensive picture of planets andstars, casting doubt on our everyday perspective. Although a newscientific tool is involved (the telescope), the reasoning is quiteordinary. In the face of inconsistent appearances, we must decidewhich perspective to endorse—just as we may judge that animpression is merely an optical illusion, or a dream is only a dream.^[7]

For Kant’s account of scientific reasoning, Newton is moreimportant (Bxxii n; cf.§1.4 below). Newton’s laws of gravity unify Copernicus’shypothesis and Galileo’s observations—and much morebesides (A663/B691). The laws are universal: they extend to allheavenly objects, not only the sun’s motion relative to theearth.

But we can never experience all events. However extensive ourexperiences may be, and however many people’s experiences wedraw on, experience is finite. As such, it can never justify theprinciple that laws are genuinely universal, or that these laws willcontinue to hold in the future.^[8] Kant argues that reason is justified in adopting these principles(and others). However, he makes a subtle distinction. These principlesshould guide scientific inquiry—in Kant’s term, they are“regulative.” But they do not provideknowledgeof the world—as Kant puts it, they are not“constitutive.” (See e.g., Buchdahl 1992; Friedman 1992c;Kant’s theory of judgment, §4.2.)

For Kant, the categories of the understanding are the paradigm exampleof constitutive ideas. For example, the category of causationstructures all our knowledge. We do not perceive cause and effectdirectly. Instead, experience is only possible because weassume events have causes. (SeeKant and Hume on causality for more detail.) Although we can make mistakes about specificcauses, we also have the following knowledge about cause and effect:weknow that every event has a cause. The category of causeand effectconstitutes experience—it enables us toexperience the world.

By contrast, regulative principles govern our investigations, withoutoffering any guarantee as to what we will find. Science seeks thegreatest possible completeness and systematicity (cf. Guyer 1989 &2006, Abela 2006, Mudd 2017). Although scientists often pursue highlyspecialized inquiries, science also has aregulative goal. Itmust integrate all findings under the most all-encompassing laws.Newton’s laws have been modified by general relativity, forexample, pointing to the open-ended, “regulative” natureof this quest. Likewise, scientists are still looking for asatisfactory way to integrate general relativity with quantum mechanics.^[9]

We cannot know in advance how far science will succeed, or that natureisconstituted as wholly law-like. This would represent a“constitutive principle,” a “cosmological”knowledge claim which goes beyond any possible experiences. Instead,the principle of seeking law-like unity represents a“maxim” orregulative principle of reason(A666/B694). Rather than determining what science must find, suchprinciples guide scientific practice.

Kant’s account of science, and especially the role of“teleological” or purposive judgment, is further developedin theCritique of the Power of Judgment. See Guyer 1990,Freudiger 1996, and Nuzzo 2005, as well asKant’s aesthetics and teleology (§3). On Kant’s account of science more generally, seeWartenberg 1992, Buchdahl 1992, Friedman 1992b & 2013, Mudd 2016and Breitenbach 2018. On reason and science, see Neiman 1994: Ch. 2and Grier 2001: Ch. 8. The entry onKant’s philosophy of science considers Kant’s view of the natural sciences, especiallyphysics.

1.3 The limits of reason

This is the best-known point, and is considered in detail in the entryonKant’s critique of metaphysics. Kant demolishes a series of supposed proofs of the existence of God(“The Ideal of Pure Reason”) and the soul (“TheParalogisms”).

He also demonstrates that it is equally possible to prove somejudgments about “world wholes” as it is to prove theiropposites. “The Antinomies” contain arguments that spacemust be unboundedand that it must be bounded, that theremust be an absolutely first causeand that there cannot be(the problem of freedom in the famous “ThirdAntinomy”).

These sections have always been regarded as the most convincing partsof the book. Mendelssohn spoke for many of Kant’s contemporariesin calling him the “all-destroyer.” These contradictionsshow the limits of metaphysical knowledge.

Kant’s intentions are not destructive, however. Philosophersmust understand the capacity that they rely on—that is,reason (§1.4). Theologians and metaphysicians have often claimed knowledge thatreason cannot deliver. This has led to empty battles that inviteskepticism about reason. After all, if reasoning leads to inconsistentclaims, how can we rely on it? By contrast, Kant aims to clear theground: reason has limits but it is not powerless; philosophicalreasoning can justifysome claims. These include the claimsdiscussed above (§1.1 and§1.2) as well as reason’s practical role (§2 below). Gava & Willaschek (forthcoming) usefully stress thisaspect of the firstCritique.

In the Doctrine of Method (the last, least-read part of the firstCritique) Kant refers to the biblical story of Babel(Genesis, Ch. 11). The literal meaning of “Babel”is “confusion”—God punished human beings’attempt to build “a tower that would reach the heavens”(A707/B735) by giving them many different languages. Since they wereunable to understand one another, they could no longer cooperate insuch hubristic ventures.^[10] Again and again, reason returns to some simple ideas with toweringimplications—the immortal soul, God, freedom. Worse, it cooks up^[11] more or less convincing proofs of these.

Since these ideas reach beyond experience of a shared world, peoplelack a shared way to test them. Perhaps they emptily repeat otherpeople’s words without real understanding; quite possibly, theycome up with conflicting versions of these ideas. They talk past oneanother; they might as well speak different languages. Most likely,they will fall into conflict, or find peace only by submitting to anunreasoned authority. In metaphysics, Kant refers to “theridiculous despotism of the schools” (Bxxxv).^[12] In practical life, however, despotism is far from ridiculous: it isthe brutal last resort for securing order when people adoptconflicting ideas and pursue conflicting goals. Kant often alludes toHobbes, who holds that peace is only possible if an unaccountablesovereign “overawes” every member of society.^[13] On many interpretations, Kantian reason aims to build intersubjectiveorder and avoid the dangers of Babel-like hubris, conflict anddespotism (Saner 1967, O’Neill 1989, Neiman 1994).

In the Preface to the second edition of theCritique, Kantmakes a famous claim: “I had to denyknowledge in orderto make room forfaith” (Bxxx). Human beings cannothave knowledge of the world as a whole. They cannot know entities thattranscend this world, such as the immortal soul or God. We cannotexperience these things through our senses; reason cannot supply suchknowledge. However, Kant argues that knowledge isnot theprimary end of reason. Onlyour role within the world“necessarily interests everyone” (A839n/B867n). Kantrejects a “scholastic” or knowledge-oriented notion ofphilosophy. Instead, he offers a “cosmic” orworld-oriented one (A838/B866; cf. Ypi 2021: Ch. 1 and Ferrarin2015).

Kant proposes three questions that answer “all the interest ofmy reason”: “What can I know?” “What must Ido?” and “What may I hope?” (A805/B833). We haveseen his answer to the first question: we can only know the world asrevealed through the senses. Kant does not answer the second questionuntil theGroundwork of the Metaphysics of Morals, four yearslater. (Arguably, he sees no need to answer the question in this form,since he is confident that people have long known how they ought to act.^[14]) But the firstCritique does include some observations onhope—that is, faith in God and a future world. Kant argues thatknowledge of these things is not only impossible—it would alsocorrupt practical reasoning. Rather than doing good for its own sake,we would be motivated by external incentives—eternal punishmentand heavenly reward. Kant later calls this “heteronomy,”as opposed to “autonomy”—our own personal commitmentto morality. Despite this, Kant argues that we have legitimate reasonsto hope for God and immortality. We must also have confidence in ourfreedom to act morally. He connects and develops these claims in thesecondCritique, as discussed below (§2.3).

Since “reason” is a mental capacity, it may seem strangeto speak of it having “needs” or “interests.”The basic idea is that there are preconditions for successfullyexercising this capacity. For finite human beings, reason is nottransparent or infallible, as some rationalist philosophers seemed tothink. We maythink we are reasoning, when actually we arecooking up false rationalizations and self-deceptions. We may think weare reasoningwell, when we argue toward transcendent truths,such as the existence of God or a future world. So reason has an“interest” in appreciating its own limits, if it is to bevalid. As Kleingeld puts it, reason “needs to present itself toitself in the process of gaining clarity about its own workings”(1998a: 97)—above all, the principles that it must give toitself. As the next section discusses, this means that Kant viewsreason as essentiallyself-reflexive.

1.4 Reason’s self-knowledge

The firstCritique argues that there has, so far, been noreal progress in metaphysics. In the second edition Preface, Kantproudly proclaims that his book has finally put metaphysics on“the sure path of a science” (Bvii; cf. Axiii). So wemight ask: what is the relation of metaphysics—or philosophicalreasoning more generally—to areas of human enquiry that generatecertainty (geometry and mathematics) and expand knowledge (science ingeneral)?

Kant insists that mathematics cannot provide a model for philosophy.“Mathematics gives the most resplendent example of pure reasonhappily expanding itself without assistance from experience”(A712/B740). But metaphysics cannot follow this course.^[15] This is not simply a rhetorical point, since many of Kant’spredecessors had tried to do exactly this—Spinoza’sEthics is one example, Christian Wolff’s philosophyanother (see Gava 2018). Kant’s basic argument is thatmathematicians are justified inconstructing objects oraxioms a priori, because they can work with pure intuitions—likea line or the form of a triangle, for example—rather than beingrestricted only to the analysis of concepts. (See the entry onKant’s philosophy of mathematics.) Philosophers cannot follow this sort of procedure because they haveno right to assume any a priori intuitions or axioms aboutmetaphysical entities. Attempts that rely on such claims have onlyproduced “so many houses of cards” (A727/B755).

The empirical sciences also offer an unpromising model formetaphysics. In the first place, Kant has argued that experiencecannot reveal metaphysical entities. We could neverknow, forinstance, that we are free: like everything else we can know, humanconduct is open to causal explanation. Second, experience cannotgenerate the sort of necessity Kant associates with metaphysicalconclusions. Experience revealscontingent facts: it cannotshow that such-and-suchmust be the case. (Kant’sposition here is complicated, since he also holds that scientific lawshave the metaphysical quality of necessity. Genuine laws areuniversal, and not mere generalizations or rules of thumb.But this metaphysical claim requires philosophicaljustification—thus Kant’s short book,MetaphysicalFoundations of Natural Science (1786). It is not discovered byinvestigating the world.)

None of these points, however, deters Kant from using the imageryof science and experiment to describe his own philosophical efforts.These metaphors are especially prominent in the Preface to the secondedition of theCritique, where he writes:

Reason, in order to be taught by nature, must approach nature with itsprinciples in one hand, according to which the agreement amongappearances can count as laws, and, in the other hand, the experimentthought out in accord with these principles—in order to beinstructed by nature not like a pupil, who has recited to him whateverthe teacher wants to say, but like an appointed judge who compelswitnesses to answer the questions he puts to them. (Bxiii)

In other words, reason, as “[self-]appointed judge,” doesnot sit by and merely observe whatever comes along. It activelyproposes principled accounts of the phenomenon itinvestigates—that is, law-likehypotheses. Then itdevisesexperiments to confirm or disprove these.

Kant sees grounds for optimism in this picture. The natural sciencesinvestigate the infinite scope of the empirical world. In contrast,philosophy only considers “what reason brings forth entirely outof itself… as soon as [its] common principle has beendiscovered” (Axx). The Transcendental Dialectic of the firstCritique offers one application of this idea. Kant insiststhat there are only three transcendental ideas—the thinkingsubject (or soul), the world as a whole (the universe, or the entiretyof time and space), and a being of all beings (that is, God)(A334/B391). As such, metaphysics is prone to three fundamentalmistakes, which spring “from the nature of reason” itself(A339/B397; for more detail, seeKant’s critique of metaphysics). Philosophers have made many efforts to reach these ultimate or“unconditioned” objects. But these efforts always fail andthey always involve the same basic mistakes.

Whether or not Kant is correct in this, his own picture ofphilosophical reasoning may seem puzzling. He suggests that reasonmust conduct an experimentupon itself—an idea that ishard to make sense of. In addition, although he talks aboutreason’s “common principle,” he does not explainwhat this may be.

Kant’s “experiment” may be less puzzling if weconsider his hypothesis about everyday knowledge. He makes an analogywith Copernicus’s suggestion that the earth orbits the sun (Bxvif). Only if we take account of the earth’s motion can weunderstand what we see in the heavens: we need to appreciate how ourown (relative, moving) standpoint affects our observations.Kant’s parallel hypothesis is that our experience depends on thestandpoint and capacities of the observer. Our knowledge is not likeGod’s (albeit more confused and limited), as rationalistphilosophers sometimes suggested. Human knowledge has a structure ofits own, and its own limits too.

The first main sections of theCritique explain and defendKant’s philosophical hypothesis. We must take account of ourhuman standpoint in order to explain the “a prioristructure” of our experience—for example, how it isunified in a single order of space and time. The hypothesis issupported partly because the main alternative fails. Thenon-Copernican or non-Kantian view assumes a “singlestandpoint.” It does not distinguish between everyday knowledgeand metaphysical speculation “beyond the bounds ofexperience” (Bxix). This leads to all sorts of problems. Forexample, we might suppose that immortal souls could interact withmaterial objects and mortal human beings. Kant satirized this view inhis earlier book,Dreams of a Spirit-Seer Elucidated by Dreams ofMetaphysics (1766). His stated target was a religious visionarycalled Emanuel Swedenborg. But the subtitle reveals Kant’sphilosophical point: metaphysicians have been dreaming, too. They havenot appreciated the structure and limitations of the human standpoint.So their arguments have created “an unavoidable conflict ofreason with itself” (Bxviii n). The “Antinomies”mentioned above (§1.3) explain the basic contradictions that result.

Even if Kant’s arguments succeed, this still leaves awkwardquestions about philosophical reasoning. Kant is not just concernedwith our capacities for everyday knowledge. As the book’s titlesays, he means to “critique” or examine reason. Butreasoning is the only means we have to do this. Does it make sense tothink of reason as criticizing or justifying itself? Might it evenpoint to something distinctive about our capacity to reason, that itcan examine itself in this way?

§3 below discusses the most thorough reply to these question in theliterature, by Onora O’Neill. To anticipate briefly:Kant’s metaphors—of reason’s experiment upon itself,as well as legal metaphors about testimony and examination^[16]—point to a general problem: reason’sself-knowledge (cf. Axif).

Kant assumes that we have a capacity to reason and that “tophilosophize [is] to exercise the talent of reason…”(A838/B866). But since metaphysical reasoning has often gone wrong, hecannot merely assert the supremacy of reason. As he exclaims in thesecond edition Preface:

how little cause have we to place trust in our reason if in one of themost important parts of our desire for knowledge it does not merelyforsake us but even entices us with delusions and in the end betraysus! (Bxv)

Kant’s question, then, is how we mightdefend reasonfrom various doubts^[17] and how we mightdiscipline reason without beggingquestions—for instance, by invoking claims that are themselvesopen to doubt (cf. O’Neill 1989: Ch. 1, 1992, 2004 & 2015).This is the central task ofcritique (cf. Bxxxv):reason’s examination of itself. This should establishreason’s powers and limits. It should give an account ofreason’s “common principle” and its authority.

2. Practical reason: morality and the primacy of pure practical reason

The firstCritique only hints at the form of Kant’smoral theory would take.^[18] The account of practical reason in theGroundwork of theMetaphysics of Morals (1785) andCritique of PracticalReason (1788) is radically new. Kant now presents the supremeprinciple of practical reason—the Categorical Imperative. It isanimperative: it represents a command for human beings, whohave needs and inclinations and are not perfectly rational. It iscategorical or unconditional: it must always guide ouraction.

Kant’s first formulation of this principle runs as follows:“act only according to that maxim through which you can at thesame time will that it become a universal law” (4:421). (Kantalso gives other versions of this Imperative, which he sees asequivalent: seeKant’s Moral Philosophy, §5–§9.) Kant argues that this principleis implicit in common human reasoning. Although it requires muchphilosophical effort to articulate this principle, we rely on itwhenever we make moral judgments.

Alongside the Categorical Imperative, Kant also mentions the principleof “hypothetical imperatives.” As he puts it,“Whoever wills the end also wills (in so far as reason hasdecisive influence on his actions) the indispensably necessary meansto it that is in his control” (4:417; cf. 5:19f^[19]). Following Hume, many philosophers hold that practical reasoning isessentially instrumental. In this case, practical reason only givesconditional or “hypothetical” instructions. These have an“if—then” structure. If someone hasparticular ends or inclinations,then they should adoptsuitable means. (This is the idea ofinstrumental rationality; see alsoKant’s moral philosophy, §4;Kant and Hume on morality, §3.)

In Kant’s view, this is just a matter of consistency. In histerms, the principle of hypothetical imperatives isanalytic—to pursue an endis to employ“the indispensably necessary means” (4:417; cf. Korsgaard1997, Newton 2017, Pollok 2017: Ch. 8). Someone who fails to take upnecessary means is not, in fact, pursuing that end. At best, they arehoping or wishing. At worst, they are involved in a practicalcontradiction—a sure sign of irrationality. However, they canovercome this contradiction by abandoning the end. Seen as mereconsistency, rationality cannot require more.

By contrast, the categorical imperative issynthetic (4:420,447). It provides an end that every rational agentmust adopt.^[20] For Kant, the requirement to adopt universalizable maxims means thatwe must respect all rational beings as“ends-in-themselves” (4:428). For Kant, reason—andonly reason—is the source of this unconditional demand.

2.1 Freedom implies moral constraint: the Categorical Imperative

Alongside the derivation of his supreme moral principle, the mostdifficult questions about Kant’s view of practical reason centeron its relation to freedom. Although the broad outlines areconsistent, Kant’s views on this topic seem to shift more thanother aspects of his critical thought. (SeeKant’s moral philosophy, §10, for a brief sketch, and Allison 1990 for a full account.)This section and the following§2.2 focus on Kant’s radical claim that “freedom andunconditional practical law [that is, the Categorical Imperative]reciprocally imply each other” (5:29f). Freedom implies thatpractical reason can be pure (non-instrumental, unconditional), andhence that we are subject to the Categorical Imperative. Oursubjection to morality implies that we must be free.

Kant’s argument that freedom implies the Categorical Imperativeis brief (seeCritique of Practical Reason, 5:19–30).If I am free to step back from my inclinations, those inclinationscannot provide a definitive reason to act. Inclinations motivate, butthey do not compel. If someone asks me why I did something,inclinations may explain my action—the actionhelped me achieve something I wanted. But there is still an openquestion: should I have acted on those inclinations?

An example may help here. Pushing you out of the way might satisfy mydesire to get to the front of the queue. The desire motivates me; wemight say that it provides a “reason for me.” For otherpeople, my desire onlyexplains my rudeness. It does notsupply others with a reason to endorse my behavior; it does notjustify my doing this. Compare the entry onreasons for action: justification, motivation, explanation: “Because I wanted to” may be a “motivatingreason” for me; it may be an “explanatory reason”for other people. But it is not a “justificatory reason.”Kant accepts that desires are relevant to practical reasoning. Hisclaim is that they need not, and should not, determine our reasoning.Simply put, my inclinations do not matter more than otherpeople’s.

How can we find reasons that others should also accept? Kant claimsthe only possibility is to look to “the mereform ofgiving universal law” (5:27). There is, he says, “only asingle categorical imperative…act only according to thatmaxim through which you can at the same time will that it become auniversal law” (4:421). This principle is categorical (noif-s or but-s!); it is an imperative (wemust follow it!).When our action passes this test, we have done our duty. (In fact, thefirst section of theGroundwork argues that this principlefollows just by analyzing the idea of duty. For more detaileddiscussion seeKant’s moral philosophy, §2–§5.)

Kant’s emphasis on duty may sound old-fashioned; the idea ofjustifying your conduct is more familiar, especially when we thinkabout practicalreasoning. However, duty and justificationare inseparable. To claim that you have done your duty is to hold thatyour actions arejustified. Other people should endorse youractions as the right thing to do. If they resent or frustrate youractions, they are acting unreasonably—not you.

But why does Kant think that “the mereform ofgiving universal law” can capture the idea of duty, or revealwhat sort of action is justified?

Two points are central. First, the “form of… law”refers to an overall policy or principle. Without this, my actionswould be random and make no sense, even to me. Second, Kant refers to“giving universal law.” Not just any policy or principlewill do. A justification is meant to provide other people with reasonsto endorse my conduct—common sense already tells us that manypolicies won’t do this. In fact, Kant argues that there is onlyone. My policy must be to act on principles that everyone else canalso adopt. This is what he means by “universal law.”

To give a better sense of what he means, Kant notes some otherprinciples thatseem law-like, but go beyond the “mereform of universal law”—and thereby fail to be justifiable.In the secondCritique (5:39ff), he discusses six suchcandidates. Here is one example to illustrate.

Kant sees Epicurus as advocating the following policy: always followyour inclinations, wherever they might lead. This might soundlaw-like—it directs action, just as ordinary laws tell you topay your taxes or drive on the left. It may even sound like a law thata free agent could adopt. While it deprives you of the freedom to actin ways youdon’t want, it preserves something thatsounds much more worthwhile: the freedom to do whatever youdo want. However, it goes beyond the “form oflaw” because it tells you to submit to merely subjectivefactors—your inclinations, whatever they may be. This raises twoproblems.

First, the policy is not actually law-like in practice—myinclinations change and conflict with one another. For example,short-term inclinations often undermine longer-term ones; oftenit’s impossible to get everything we wish for. This isn’tso much freedom, more a recipe for confusion. To act coherently, Ineed to discipline and rank my inclinations. In other words, my willwould contradict itself if I tried only to follow inclinations. Iwould not really have a will—in Kant’s terms, there is acontradiction in evenconceiving this policy.

Second, it is not a policy that everyone can follow. If they did, theresults would be utterly chaotic; they would defeat anyone’sattempts to satisfy their inclinations. In Kant’s terms, thereis a contradiction inwilling such a policy. So far as we caneven imagine this, we see straightaway that each person’swilling would predictably fail. (Kant’s moral philosophy, §5, explores these two contradictions, in conception and inwilling.)

In more abstract terms: such a policy is unreasonable because it givesauthority to particular features of one particular agent—worse,to features that aren’t even stable or mutually consistent. InKant’s words:

it is requisite to reason’s lawgiving that it should need topresuppose onlyitself, because a rule is objectively anduniversally valid only when it holds without the contingent,subjective conditions that distinguish one rational being fromanother. (5:21)

No principle is truly law-like unless it abstracts from anagent’s particular motivations and situation. Every otherpossibility looks for substantive guidance from outside of reasonitself—just as a hypothetical imperative only guides youif you already want to pursue that end. Kant refers to thisbasic difficulty asheteronomy—that is, reasoningdirected from the outside. This external authority might be myinclinations, or religious dogma, or even a political authority. Butit is merely assumed or imposed; it cannot justify itself.Heteronomous reasoning does not address the fundamental problem: tofind principles that are entitled to guide everyone’s acting andthinking. In the word Kant has made so famous, reasoning must beautonomous—that is, self-governed.

There have been many doubts whether the mere “form of law”can really provide concrete guidance. In reply, Kantians argue that itrepresents a substantial constraint: we must avoid all ways ofthinking and acting that cannot be followed by all. (For discussionseeinter alia O’Neill 1989: Ch. 5; Herman 1993;Allison 1990: Ch. 10 §II.) If this is true, then the autonomy ofreason reveals the positive sense of freedom at the heart ofKant’s practical philosophy (cf. Brandom 1979). That is, reasonenables us to act in ways that donot rely on“contingent, subjective conditions that distinguish one rationalbeing from another” (5:21). It enables us to justify our conductto one another. It empowers us to build communities around principlesthat everyone can share.

2.2 Moral constraint implies freedom: Kant’s “fact of reason”

As well as claiming that freedom implies the Categorical Imperative,Kant claims that moral obligation implies freedom.

Kant insists that “nothing in appearances can be explained bythe concept of freedom” (5:30). Morality, he claims,“exists in the sensible world [the world as known through thesenses and by science] but without infringing on its laws”(5:43). Every action counts as an event in the world of everydayobjects (or “appearances,” in Kant’s terminology).We must take every event to be caused—in terms of everydayexplanations as well as in sciences such as physics and neuroscience.Considered as events in theobjective world, human actions donot provide us with a warrant for freedom.

Instead, it is to our consciousness orsubjectivity that Kantturns:

Ask [someone] whether, if his prince demanded, on pain of…immediate execution, that he give false testimony against an honorableman who the prince would like to destroy under a plausible pretext, hewould consider it possible to overcome his love of life… Hewould perhaps not venture to assert whether he would do it or not, buthe must admit without hesitation that it would be possible for him. Hejudges, therefore, that he can do something because he is aware thathe ought to do it and cognizes freedom within him, which, without themoral law, would have remained unknown to him. (5:30; cf. 5:155f)

As Kant also says, “the moral law, and with it practical reason,[have] come in and forced this concept [freedom] upon us”(5:30). In the next section, Kant refers to this as a “fact ofreason”: “Consciousness of this fundamental law may becalled a fact of reason because one cannot reason it out fromantecedent data of reason” (5:31; cf. 5: 6, 42f, 47, 55, 91,104).

This “fact” has caused considerable controversy, forseveral reasons. Kant is not altogether clear about what he takes thisfact to demonstrate. He has argued that facts about human beingscannot provide the basis for morality—morality must be given apriori, that is, independently of experience. Further, Kant speaks of“cognizing the moral law.” But he is well aware that hisown account of this “law” is quite new. Not least, this“fact, as it were” does not feature in his earliertreatise, theGroundwork, and does not appear again.

One school of thought—which includes many influential Kantscholars, and is sympathetically represented in Allison 1990 (Chs. 12and 13)—sees a fundamental change in Kant’s thought here.Part III of theGroundwork seems to give a“deduction” (justification) of freedom. But in the secondCritique Kant realizes that his own premises do not allowsuch an argument. So he stops argument short by appealing to asupposedly indubitable fact.

Other commentators emphasize the clear continuities between the twobooks. In particular, Kant relies on common moral consciousness inboth. Łuków 1993 emphasizes the parallel betweenAchtung (“respect” or “reverence” formorality) and the fact of reason. Kant refers to respect in all hisethical writings: it is the only feeling “self-wroughtby a rational concept [= the moral law]” (4:401n). As such, itclearly resembles what he now calls “the sole fact of purereason” (5:31). (See also O’Neill 2002 and Timmermann2010.) Moreover, Kant uses the Latin wordfactum. Arguably,this is better translated asdeed rather than fact. In otherwords, Kant points to anact of reason as well as thefact that it creates (Kleingeld 2010).

In any case, Kant’s line of thought in the long passage quotedabove is fairly clear: We all (most of us) recognize that there aresituations where we ought to do something, even though it will cost usdearly. In other words, we feel that we are subject to anunconditional moral imperative. When we acknowledge this“ought,” we show our belief in the possibility of actingon it. This shows us to be free—not merely in the negative sensethat we could go against our inclinations, but also in the positivesense that we can do our duty (4:446). Only our moral awarenessreveals this freedom.

Clearly, this line of thought is not immune to criticism. For example,our feeling of moral constraint might be explained in terms of aFreudian super-ego. But it shows why Kant thought that moralawareness—unlike any other sort of experience—gives us apractical certainty of our freedom. While this is not knowledge in theempirical or scientific sense, it provides a basis for practicalreasoning: “a fact [Faktum] in which pure reason in usproves itself actually practical” (5:42). And if Kant is rightthat only the Categorical Imperative reveals our capacity to act inways that we can justify to others, then we can see why he claims,“freedom and unconditional practical law reciprocally imply eachother” (5:29f).

2.3 The primacy of (pure) practical reason

Kant does not give a complete account of the relation betweenpractical reason and theoretical reason. However, the secondCritique does include an important section, “On theprimacy of pure practical reason in its connection with speculativereason” (5:119–121). (See Gardner 2006 and Willaschek2010.)

We have already seen two reasons why Kant must give some priority to“pure practical reason.” First, assume that Kant iscorrect—against rationalist philosophers—that theoreticalreason cannot have insight into the supersensible. This means thatreason cannot have access to any transcendent authority (such as Godon many traditional conceptions) that might command thought oraction.

Second, Kant says that practical reason can either be“pure”—i.e., “proceed from a prioriprinciples”—or it can proceed “from empiricaldetermining grounds” (5:90). Desires and inclinations areempirical, for Kant. Reasoning that helps us fulfill them is“hypothetical”—it applies only if you have thosedesires. For Kant, merely following your desires represents“heteronomy,” especially when they conflict with morality.On this view, reason is just aninstrument to help youfulfill your desires. As Hume put it, in an especially horriblemetaphor, reason would be “the slave of the passions”(Treatise, 2.3.3.4). It cannot issue its own commands.^[21]

These two points suggest that neither theoretical nor instrumentalreasoning can supply authoritative reasons for action. If there aresuch reasons—as Kant’s “fact of reason”supposes—thenonly pure practical reason can supplythem.

Now, however, Kant makes a stronger claim. He argues that purepractical reason has “primacy” even on the home territoryof theoretical reason. That is, pure practical reason should guidesome of ourbeliefs, as well as our actions.

Kant defines primacy as “the prerogative of the interest of oneinsofar as the interests of others is subordinated to it”(5:119). He gives at least three considerations to support this“prerogative” of practical reason.^[22]

One point just reflects Kant’s talk of reason’s“interests.” “[A]ll interest,” says Kant,“is ultimately practical and even that of speculative reason isonly conditional and is complete in practical use alone”(5:121). In this sense, to accept the metaphor of“interests” is to see reasoning as ultimately practical.(Cf. Kant’s notion of philosophy as “world-oriented”rather than “scholastic”—§1.3 above. See also Neiman 1994: Ch. 3; Kleingeld 1998a; Rauscher1998.)

Second, practical reason can be “pure” or independent from“pathological conditions,” that is, our inclinations. Asjust noted, Kant holds that practical reason can be “pure”or “a priori.” That is, it need not be governed byour inclinations. By contrast, theoretical reason goes wrong whenit seeks knowledge by itself—for example, in merely“speculative” proofs that God exists. Theoreticalreasoning can only gain knowledge through sensibility andunderstanding. Practical reason is independent in a way thattheoretical reasoning can never be.

A third line of thought is less abstract, and rests on the specificinterests that Kant attributes to these two forms of reasoning. Whentheir interests conflict, we must decide which are to come first.

The “interest” of theoretical reason consists in expandingour knowledge and avoiding error. As Kant has argued, reason can helpbroaden scientific knowledge, but it cannot justify speculativemetaphysical claims. However, practical reason has a conflictinginterest—in beliefs that lie beyond the bounds of knowledge.Kant writes:

But if pure reason of itself can be and really is practical, as theconsciousness of the moral law proves it to be [cf.§2.2 on the “fact of reason”], it is still only one and thesame reason which, whether from a theoretical or a practicalperspective, judges according to a priori principles; it is then clearthat, even if from the first [theoretical] perspective its capacitydoes not extend to establishing certain propositions affirmatively,although they do not contradict it,as soon as these samepropositions belong inseparably to the practical interest of purereason[, then theoretical reason] must accept them. (5:121)

In other words, even if it is finally “one and the samereason,” reason seems divided. Used practically, reason has aninterest in “certain propositions”—specifically, thereality of freedom, immortality and God. But theoretical reasoningsays these cannot be proven. The practical interest has greaterweight, because it is rooted in morality.

Here, Kant introduces the idea of a “postulate.” This is“atheoretical proposition, though one not demonstrableas such, insofar as it is attached inseparably to an a prioriunconditionally validpractical law” (5:122). For Kant,there is only one unconditional law: the Categorical Imperative or“moral law.” He claims that three postulates are“attached” to it: “immortality, freedom consideredpositively (as the causality of a being insofar as it belongs to theintelligible world), and the existence of God” (5:132).

Freedom has a special status—as discussed in§2.2, the secondCritique’s “fact of reason” ismeant to establish this. But why does he think we must“postulate” God and immortality?^[23] (See also the entry onKant’s philosophy of religion.)

As we have seen, Kant holds that we should act out of“respect” for morality—for the sake of duty.Morality must have priority over my inclinations. (Remember:inclinations canmotivate andexplain actions, butthey are not enough tojustify action.) This meansthat there is no guarantee that acting well will lead to my ownhappiness, or even other people’s. This creates aconflict—in Kant’s terms, a“dialectic”—between happiness and morality. Whilemorality is the onlyunconditional good for human beings,Kant does not deny that happiness is important. It is the natural andnecessary end of every human being (cf. 4:415); duty requires us tomake others’ happiness our end.^[24]

Kant expresses the combined importance of morality and happiness as“the highest good”:^[25]

virtue (as worthiness to be happy) is thesupremecondition of whatever can even seem to us desirable and hence ofall our pursuit of happiness… and is therefore thesupreme good. But it is not the whole and complete good forfinite rational beings; for this,happiness is also required,and that not merely in the partial eyes of a person who makes himselfan end, but even in the judgment of an impartial reason [in otherwords, this is not about my subjective wish to be happy, but rather anobjective judgment that happiness is the natural end for human beings,just as goodness is our moral end—GW]… happinessdistributed in exact proportion to morality (as the worth of a personand his worthiness to be happy) constitutes thehighest goodof a possible world. (5:110)

The argument here is bold but dubious. We must think of moral activityas resulting in happiness. Yet human agency cannot achieve this:“I [or even all of us acting together—GW] cannot hope toproduce this [highest good] except by the harmony of my will with thatof a holy and beneficent author of the world” (5:129). So, Kantargues, we must “postulate” that God exists. We must alsopostulate immortality. Only this enables us to hope that we willfinally come close to virtue, hence become worthy of happiness.

3. The unity of theoretical and practical reason

We have seen one link between theoretical and practical reason. Inanswer to the question, “What may I hope?” Kant claimsthat practical reason has “primacy.” Theoretical reasonmay accept the postulates of God, freedom and immortality “as aforeign possession handed over to it” (5:120). Theoreticalreason can only show that these things are not impossible. Ourpractical interests—at root, our moral obligations—demandthat we believe in them. To cite Kant’s famous words again:“I had to denyknowledge in order to make room forfaith” (Bxxx).

Kant’s arguments for freedom are more extensive and compelling.His arguments for God and immortality find little favor amongcontemporary authors. Perhaps the strongest support comes from JensTimmermann, who claims: “the principle that unifies the spheresof theoretical and practical reason… is the assumption of awise and benevolent God who has created a teleological world thatcoheres with morality” (2009: 197). (On teleology in theCritique of the Power of Judgment, see Guyer 1989, Freudiger1996;Kant’s aesthetics and teleology, §3. Outside that work, see also Wood 1970, Kleingeld 1998b.)

Even if Kant did believe this—in his weaker moments, perhaps^[26]—such a position lacks wider philosophical resonance. Most contemporaryphilosophers assume that the world does not “harmonizewith” morality in this way—or at any rate, that it is ahuman task to foster such harmony, for example, by building decentsocieties and just institutions. Reath 1988 argues that Kant sometimesdeploys a more defensible, “secular” or this-worldly,notion of the highest good. Kleingeld 1995 and Guyer 2000a & 2000balso offer interesting discussions.

Onora O’Neill has made the leading attempt to uncover the unityof Kantian reason and to relate it to contemporary philosophicalconcerns (1989 and subsequent essays). This section will focus on hercentral claim concerning the unifying role of the CategoricalImperative, and the main bases for this claim in Kant’s texts.O’Neill’s interpretation of Kantian reason enjoysconsiderable respect among Kant scholars, although there arerelatively few critical responses to her work. Engstrom 1992 and Wood1992 offer early reviews; Westphal 2011 and Cohen 2014, 2018 take upher account, as does Bagnoli 2022: Ch. 4; see furtherconstructivism in metaethics, §2 (especially §2.3).^[27]

3.1 Reason’s “common principle”

In the original Preface to the firstCritique, Kant suggestedthat reason has a “common principle”: “Nothing herecan escape us, because what reason brings forth entirely out of itselfcannot be hidden, but is brought to light by reason itself as soon asreason’s common principle (gemeinschaftliches Prinzip)has been discovered” (Axx). Unfortunately, neither edition oftheCritique considers what this principle might be.

Kant also raises this principle in his works on practical reason,without giving a clear account. In the Preface to theGroundwork, Kant explains why the book is not entitled aCritique of Pure Practical Reason:

[A critique of pure practical reason] is not of such utmost necessityas [a critique of pure theoretical reason], since human reason, evenin the commonest understanding, can easily be brought to a highmeasure of correctness and accuracy in moral matters, whereas in itstheoretical but pure use it is totally and entirely dialectical [i.e.,a source of illusion]… I require that the critique of a purepractical reason, if it is to be complete, also be able to present itsunity with speculative reason in a common principle; because in theend there can be only one and the same reason, which must differmerely in its application. (4:391)

In the secondCritique, Kant compares the book’sstructure with the firstCritique and comments: “suchcomparisons [are] gratifying; for they rightly occasion theexpectation of being able some day to attain insight into the unity ofthe whole rational faculty (theoretical as well as practical) and toderive everything from one principle—the undeniable need ofhuman reason, which finds complete satisfaction only in a completesystematic unity of its cognitions” (5:91). Kant’s tone isconfident. But once again, he has postponed an account of “theunity of the whole rational faculty.” (Prauss 1981 argues thatKant failed to achieve this insight, in part because he did notappreciate how cognitive success is a fundamentally practical goal.Förster 1992 discusses Kant’s reflections on this topic inhis final manuscript, theOpus Postumum.)

However, as Onora O’Neill points out in a celebrated essay(1989: Ch. 1), Kant’s claims about practical reason imply afurther claim about reason’s “common principle.”Kant has argued that the Categorical Imperative is the supremeprinciple of practical reason. He has also argued that practicalreason has primacy over theoretical reason. It follows, therefore,that the Categorical Imperative is the supreme principle ofreason.

To be sure, Kant never states this conclusion explicitly. But thereare reasons for thinking that this ought to have been his view, and insome places he comes very close to this claim.^[28]

The clearest passage is a footnote (!) in Kant’s essay,“What is it to Orient Oneself in Thinking?” (1786):

To make use of one’s own reason means no more than to askoneself, whenever one is supposed to assume something, whether onecould find it feasible to make the ground or the rule on which oneassumes it into a universal principle for the use of reason. (8:146n)

This bears an obvious parallel with the first formulation of theCategorical Imperative—“act only according to that maximthrough which you can at the same time will that it become a universallaw” (4:421). Kant now says:think only in accordancewith that maxim that could be a universal law.^[29]

To rephrase the point: Thinking and judging are activities. If theCategorical Imperative is truly “categorical” then itapplies toall our activities—“theoretical”as well as “practical.”

Other strands of Kant’s thought support this interpretation. Themost important are his “maxims of reason” and his accountof the “public use of reason.”

3.2 The maxims of reason

Kant sets out three “maxims of reason” or “maxims ofcommon human understanding.” They appear twice in his publishedwritings, in relation to both acting and thinking.^[30] These maxims are closely related to the Categorical Imperative. (Fordiscussion, see O’Neill 1989: Ch. 2 & 1992, and Neiman 1994:Ch. 5.)

In his last published work, theAnthropology (1798), Kantpresents the maxims in apractical context, as guidelines forachieving some degree of wisdom:

Wisdom, as the idea of a practical use of reason thatconforms perfectly with the law, is no doubt too much to demand ofhuman beings. But also, not even the slightest degree of wisdom can bepoured into a man by others; rather he must bring it forth fromhimself. The precept for reaching it contains three leading maxims:(1) Think for oneself, (2) Think into the place of the other [person](in communication with human beings), (3) Always think consistentlywith oneself. (7:200; cf. 228f)

The maxims also appear in theCritique of the Power ofJudgment (1790). Here, Kant relates them to thetheoretical use of reason. A famous section describes humanbeings’ capacity for asensus communis or“community sense”:

a faculty for judging that… takes account (a priori) ofeveryone else’s way of representing in thought, in orderasit were to hold its judgment up to human reason as a whole andthereby avoid the illusion which, from subjective private conditionsthat could easily be held to be objective, would have a detrimentalinfluence on the judgment. (5:293)

That is, the maxims are precepts for judging in accordance with“reason as a whole” and avoiding the distortions thatarise from “subjective private conditions.”

Kant describes (1) thinking for oneself as the maxim ofunprejudiced thought; its opposite is passivity in thought,leading to prejudice and superstition. Kant also labels this“heteronomy,” clearly recalling his interpretation ofreason and morality in terms of autonomy (4:433).^[31] (2) To think in the place of everyone else is the maxim ofenlarged orbroad-minded thought. And (3) always tothink in accord with oneself is the maxim ofconsistentthought (5:294).

The last maxim may sound more straightforward, but Kant emphasizes itsdifficulty. Consistency “can only be achieved through thecombination of the first two and after frequent observance of them hasmade them automatic” (5:295). Consistency does not just involvegetting rid of obvious contradictions in explicit beliefs. It alsorequires consistency with regard to the implications of ourbeliefs—and these are often not apparent to us. To achieve thissort oflaw-likeness in thought requires genuine attempts (1)to judge for oneself and the determination (2) to expose one’sjudgments to the scrutiny of others. In other words, a person must tryto make themselves: (1) really theauthor of their judgments;(2)accountable to others; and (3)consistent intheir claims. We musttake responsibility for ourjudgments.

The maxims support the thesis that theoretical and practical reasoninghave a unified structure. In particular, they flesh out theimplications of the Categorical Imperative. The imperative demandsrespect for “the mere form of law”—consistency,law-likeness, and autonomy in thought and action. Equally, it demandsrespect for other people as fellow reasoners and“ends-in-themselves” (Russell 2020, Lenczewska 2021). As amatter of thought, reasoning requires us to discipline our judgmentsso that others can follow them. As a matter of practical wisdom,reasoning requires us to discipline our actions so that others canadopt the same guiding principles.

3.3 The public use of reason

Anyone concerned with Kantian reason and politics must consider hisfamous essay, “What is Enlightenment?” (1784). Among manydiscussions, see O’Neill 1989: Ch. 2, 1990 & 2015: Ch. 3;Velkley 1989; Deligiorgi 2005; Patrone 2008.

As we have seen, Kant’s first maxim of reason is to think foroneself. He says this is the way to achieve “liberation fromsuperstition,” and equates this with “enlightenment”(5:294). His second maxim asks us to “think into the place ofothers.” Communication is essential for this. As he writes:“…how much and how correctly would we think if we did notcommunicate with others to whom we communicate our thoughts, and whocommunicate theirs with us!”^[32]

In “What is Enlightenment?” Kant places both maxims in apolitical context. He demands that we “have the courage to useourown reason”:

Enlightenment is the human being’s emergence from hisself-incurred immaturity. Immaturity is the inability to useone’s own understanding [= reason^[33]] without the guidance of another. This immaturity isself-incurred if its cause is not lack of understanding, butlack of resolution and courage to use it without the guidance ofanother.Sapere aude! [Dare to be wise!] Have courage to makeuse of yourown understanding [= reason]! is thus the mottoof enlightenment. (8:35)

Kant’s main concern is not with individuals, however.Enlightenment is a collective endeavor. It therefore has cultural andpolitical aspects. For enlightenment, Kant says, “nothing isrequired but… the least harmful… freedom: namely,freedom to makepublic use of one’s reason in allmatters” (8:36).

This “least harmful freedom” is not the freedom to actpolitically. It is not even freedom of speech in the now-familiarsense that covers many forms of personal expression. Instead, it isthe freedom to reason “as a scholar before the entirepublic of theworld of readers” (8:37).

As we might expect, Kant contrasts public reasoning with privatereasoning. However, the distinction has an unfamiliar meaning. ForKant, employees reason privately. He mentions civil servants, militaryofficers, or clerics in an established church. In contemporary terms,these are all more or less “public” roles. However, theseemployees are bound by the policies, procedures and dictates of theirorganization. They reason about how to act—but only to judge thebest way of fulfilling tasks laid down by their employers.

We can draw a parallel with an individual who reasonsinstrumentally—about the best way to satisfy their inclinations,for example. Instrumental reason decides the best means to achievepre-given ends. In Kant’s terms, it isheteronomous—directed by a source outside reason itself.

Similarly, these employees are directed by state or church. Theiroffices may be “public,” but their reasoning isprivate—deprived of full publicity. The employee“carries out another’s commission” (8:38). Hisreasoning is not free: it is “impermissible to argue; instead,one must obey” (8:37). The employee does not abandon his ownmoral judgment—Kant is careful to add that he must resign if hecould not “in conscience [continue to] hold his office”(8:38). But within that office, he acts as a “passivemember” (8:37) of the commonwealth.

By contrast, the public use of reason is active and autonomous. Itdoes not follow directions; it is accountable to all. When a personreasons publicly, he thinks for himself and speaks as a member of“the society of citizens of the world” (8:37). Outside ofhis post, an employee may reason freely, and may even criticizegovernment policies or religious teachings.

The same applies to the ordinary citizen. He must obey—forexample, by paying taxes—and may not speak in a way thatencourages others to disobey. But he “does not act against theduty of a citizen when, as a scholar, he publicly expresses histhoughts about the inappropriateness or even injustice of[governmental] decrees” (8:38).

This creates a stark contrast between roles: the obedient citizen oremployee, as opposed to the “scholar” who makes“public use of reason.” But Kant sees no inconsistency.Peaceful co-existence requires that citizens do not undermine, letalone rebel against, existing governments—that risks chaos andeven civil war. In Kant’s words, “morally practical reasonpronounces [an] irresistibleveto: there is to be nowar” (6:354). As citizens of a given state, we must act aspolitical institutions demand. As citizens of the world, however, wemay criticize those demands and advocate reform.^[34]

Kant’s views invite criticism. In his day, only a few citizenscould read and publish their views; hardly any women could. So hisposition may seem elitist and sexist. It implies a gulf between theoryand practice—between what employees and citizens may believeought to be done, and what they must do (“obey”!).“The least harmful freedom” may seem not just harmless butpowerless. Freedom to publish is only a small part of the freedomsessential to modern democracy. Hence Kant’s views may seemantiquated. Certainly, they fall far short of democratic citizenshipas we understand it.Kant’s social and political philosophy (§4 and §6) discusses these issues.

For this entry, the key point is this. Kant equates reason with fullpublicity. “To use one’s own reason” is to make asincere attempt to addressall “citizens of theworld.” Judgments and principles areonly reasonable tothe extent that they can be accepted by all. Our actions must respectexisting boundaries and governments, because the duty to pursue andmaintain peace is categorical. But reason itself aspires touniversality. So as citizens of the world, we have another categoricalduty: to see ourselves as accountable to every human being, and henceto improve existing institutions so that they recognize the claims ofall. For reasoning to befully public, citizens andnon-citizens, rulers and ruled must discuss as equals—despitetheir inequalities in practice.

A famous passage from theCritique of Pure Reason expressesthis idea, linking reason, critique, and autonomy:

Reason must subject itself to critique in all its undertakings, andcannot restrict the freedom of critique through any prohibitionwithout damaging itself and drawing upon itself a disadvantageoussuspicion. For there is nothing so important because of its utility,nothing so holy, that it may be exempted from this searching reviewand inspection, which knows no respect for persons [i.e. no personbears more authority than any other—GW]. On this freedom reststhe very existence of reason, which has no dictatorial authority, butwhose claim is never anything more than the agreement of freecitizens, each of whom must be able to express his reservations,indeed even his veto, without holding back. (A738f/B766f, translationmodified)

In the term used by several contemporary Kantians, this procedureconstitutes reason. (See Herman 2007: Ch. 10, Korsgaard 2008,Reath 2013, Bagnoli 2017; Møller 2020 focuses on Kant’slegal metaphors.) Reason is the only unconditional (that is,non-heteronomous) authority for thinking and acting.

4. Concluding remarks

No one doubts that belief and action are subject to demands ofrationality. Theoretical reasoning aims at knowledge of the worldaround us (perhaps also knowledge of ourselves). Practical reasoningaims to make a difference in the world (perhaps also a difference inourselves).

This distinction hides a larger question: whatis reason, exactly?^[35] Rationality clearly involves the capacity to make logical and causalinferences: to draw out the consequences of our beliefs; to judge whatconsequences are likely to follow from different possibilities foraction. Rationality clearly requires consistency. Our beliefs shouldnot contradict one another, or have contradictory implications. Weshould use means that support our goals, rather than defeat them. Weshould avoid contradictory goals.^[36] Beyond this, however, it is not obvious that “reason” isa unitary capacity or that “rationality” is a unitaryrequirement. If truth seems self-evident as a criterion of thinking,the criteria for rational action are much harder to make out.

If we accept the basic lesson of the firstCritique, it iseven harder to discern the criteria of rational action. Againstrationalist philosophers, Kant argues that theoretical reason cannotdiscover realities beyond empirical experience. (Recall the literalmeaning of metaphysics:beyond the physical world.) So wecannot know any sort of moral authority, be it God or revelation, thattells us how to think and act. Traditionally, truth and goodness stoodtogether. Mendelssohn called Kant the “all-destroyer”partly because his critique undercut this traditional pairing.Experience can disclose (empirical) truth, but not goodness.

Perhaps, then, we can only reason practically in order to satisfypre-existing feelings or goals. These factors can motivate us toreasoning, but they are not themselves open to rational justification.Hume’s empiricist account of morality illustrates the point.Reason is “the slave of the passions.” We can only hopethat our fellow human beings feel sympathy, and will put reason towork in its cause. Although we may disapprove, we cannot say it isirrational to feel or pursue more anti-social“passions” such as spite or envy or pride. In Hume’sown words, “’Tis not contrary to reason to prefer thedestruction of the whole world to the scratching of my finger…’Tis as little contrary to reason to prefer even my ownacknowledg’d lesser good to my greater…”(Treatise 2.3.3.6).

Since Kant has cut off (theoretical) knowledge of goodness,doesn’t consistency drive him to similar skepticism? Since hisaccount is structured by a distinction between theoretical andpractical reason, how can he show their unity?

This entry has suggested that Kant’s account of reason isconstructive (not skeptical) and unified (not divided). In thought andin action, reasoners must look for principles that others can alsoadopt:

To make use of one’s own reason means no more than to askoneself, whenever one is supposed to assume something, whether onecould find it feasible to make the ground or the rule on which oneassumes it into a universal principle for the use of reason.(“What is it to Orient Oneself in Thinking?” 8:146n)

In theoretical enquiry, we seek knowledge of the world and explore thelimits of knowledge; we think and philosophize. We hope to reachjudgments that are valid for all and draw on principles that all canrely on.

In practical reasoning, we decide how to act, individually and withothers. Kant often emphasizes duty, but this entry has stressed therelated idea ofreasoning as justification. If an action isjustified, other people should endorse it, rather than resent orcriticize it. If their circumstances are similar, they should actsimilarly. Even if their circumstances differ, the same underlyingprinciples should guide their actions. If we can share principles, wecan justify our conduct and find a basis for cooperation.

O’Neill (2000) situates the Kantian account against threealternatives. (i) Theinstrumental account of reason remainspopular. Following Hume, rational choice theory andconsequentialism see practical reasoning as a tool. Reason works out how to achievegiven ends (cf.practical reason, §4;reasons for action: internal vs. external;instrumental rationality;philosophy of economics, §5). (ii) Thecommunitarian account sees reason asembedded within shared traditions. Rationality is what a historicalcommunity takes it to be—for example, its usual modes ofargument and accepted authorities (cf. MacIntyre 1988;communitarianism). (iii) Theperfectionist account sees reason as an individualcapacity to discern moral truth or goodness. This view resembles therationalism which Kant opposed, except that it stresses divinerevelation or moral intuition, rather than abstract thought (cf.moral non-naturalism, §3).

Arguably, all three accounts fail in providing reasoned justificationto some audiences. (i) The instrumental reasoner cannot justify theiraction to people who do not endorse their goal. Instrumental reasonsexplain action in terms of the actor’s desires or ends.They can make sense of the action, but they do not give the audiencereasons to endorse it. Explanation is not justification. (ii) For thecommunitarian, their tradition defines what beliefs and practicescount as reasonable. If someone disagrees, such a reasoner can saylittle except: this is how I have been taught; this is how we dothings round here. This sort of reasoning cannot justify itself tooutsiders, nor to insiders who dissent. (Recall Kant’s freecitizens who “must be able to express [their] reservations,indeed even [their] veto.”) (iii) The perfectionist believesthat they can intuit how it is good to act or to be. But what can theysay to someone who has different “intuitions,” ornone—except perhaps to condemn them as blind or confused?

On the interpretations advanced by Saner, O’Neill, Neiman andothers, Kant was aware of these options and rejected them all. We sawabove (§1.4) that Kant characterizes reason in terms of a self-reflexiveprocedure. Reason is autonomous—that is, it rejects all externalauthorities. Reason has authority only because it rests on on-goingcritique. Openness to critique means giving up any mode of thinking oracting that cannot be adopted by all.

To put the same point in more concrete terms: Kant ties reasoning tojustification. His account stresses the limits of many everydayjustifications. Consider the modes of reasoning just mentioned. (i)“Because I wanted to” is sometimes a perfectly goodreason—so long as I don’t infringe onothers’ legitimate claims. (ii) “Because that’s howwe usually do things” is sometimes a perfectly goodreason—so long as our practices don’t endanger orhumiliate or exploit. (iii) “Because it felt like the rightthing to do” may be a perfectly good reason—so longas my “intuitions” reflect a moral sensibility thatfits my circumstances.

The Kantian point is this: Don’t forget the proviso,“so long as…” In the wrong circumstances,desires, customs and intuitions can lead us astray. Reasoning isincompleteunless we can also justify giving weight to thedesires, customs or intuitions that make something seem reasonable tous. By contrast, reasoning is complete when it meets the followingtest: Everyone could think and act on the same principles that nowguide my thought and action. To judge whether our thinking and actingmeet this demanding test, we must consider objections from differentpoints of view. As Kant says, each person “must be able toexpress [their] reservations.” This is an on-going process, withno fixed or final endpoint. Only this active process can makereasoning fully “public,” “autonomous” and“universal(izable).”

The Kantian view does not assume that we are hemmed in by ourinterests and inclinations—as many instrumental accounts do. Itdoes not ask us to rely on what others already accept—as thecommunitarian account does. It does not suppose that we can intuitwhat everyone should accept—as the perfectionist account does.Not least, it does not suppose that reason can teach us about God orany other moral authority beyond everyday experience—as manyrationalist philosophers supposed. Kant’s account sees humanbeings as able to step back from their inclinations, habits andintuitions. Reasoning is not the voice of God, but rather a capacityto heed the voices of those we share a world with. Kant asks us to usethis ability to seek principles that all can think and live by, and toorganize our lives together on this basis.

This account depends on a particular interpretation of Kant’stexts. It is abstract, ambitious, and complex in its ramifications.But if it succeeds, it captures two powerful aspects of Kant’sphilosophy: a universalism that transcends self and communityboundaries; a modesty that respects the limits of human insight.

Bibliography

Primary sources

Kant’s works are cited by volume and page numbers of theAkademie edition ofKants gesammelte Schriften(Berlin, 1902–). The only exception is theCritique of PureReason, cited by the standard A and B pagination of the first(1781) and second (1787) editions respectively. TheGroundwork is printed inAkademie volume 4 and theCritique of Practical Reason in volume 5; unless otherwisenoted, references beginning with “4:” are to theGroundwork and those beginning with “5:” to thesecondCritique. TheAkademie pagination is found inthe margins of all modern translations. Apart from the 2011 edition oftheGroundwork, translations are from the standard CambridgeEdition of the Works of Immanuel Kant:

• Critique of Pure Reason, translated/edited by P. Guyer& A. Wood, Cambridge: Cambridge University Press, 1997.
• Practical Philosophy, translated/edited by M. Gregor,Cambridge: Cambridge University Press, 1996. [Includes “What isEnlightenment?”, theCritique of PracticalReason, and the Metaphysics of Morals.]
• Groundwork of the Metaphysics of Morals (German-EnglishEdition), translated/edited by M. Gregor & J. Timmermann,Cambridge: Cambridge University Press, 2011.
• Religion and Rational Theology, translated/edited by A.Wood & G. di Giovanni, Cambridge: Cambridge University Press,1996. [Includes “What Does it Mean to Orient Oneself inThinking?” andConflict of the Faculties.]
• Lectures on Ethics, translated/edited by J. B. Schneewind& P. Heath, Cambridge: Cambridge University Press, 1998.
• Theoretical Philosophy 1755–1770, translated/editedby D. Walford, with R. Meerbote, Cambridge: Cambridge UniversityPress, 1992. [IncludesDreams of a Spirit-Seer.]
• Theoretical Philosophy after 1781, translated/edited byH. Allison, M. Friedman, G. Hatfield & P. Heath, Cambridge:Cambridge University Press, 2002. [Includes theProlegomena.]
• Anthropology, History, and Education, translated/editedby G. Zoller, R. Louden, M. Gregor & P. Guyer, Cambridge:Cambridge University Press, 2007. [Includes theAnthropology.]
• Critique of the Power of Judgment, translated/edited byP. Guyer & E. Matthews, Cambridge: Cambridge University Press,2000.

Secondary literature

• Abela, P., 2006, “The Demands of Systematicity: RationalJudgment and the Structure of Nature,” inA Companion toKant, G. Bird (ed.), Oxford: Blackwell, pp. 408–422.
• Allison, H., 1990,Kant’s Theory of Freedom,Cambridge: Cambridge University Press.
• –––, 2004,Kant’s TranscendentalIdealism, revised and expanded edition, New Haven: YaleUniversity Press.
• Ameriks, K., 2003,Interpreting Kant’s Critiques,Oxford: Clarendon Press.
• Bagnoli, C., 2017, “Kant in Metaethics: The Paradox ofAutonomy, Solved by Publicity,” inThe Palgrave KantHandbook, M. Altman (ed.), London: Palgrave Macmillan, pp.355–77.
• –––, 2022,Ethical Constructivism(Cambridge Element), Cambridge: Cambridge University Press.
• Brandom, R., 1979, “Freedom and Constraint by Norms,”American Philosophical Quarterly, 16(3): 187–196.
• Breitenbach, A., 2018, “Laws and Ideal Unity,” inLaws of Nature, W. Ott & L. Patton (eds.), Oxford: OxfordUniversity Press, pp. 108–121.
• Buchdahl, G., 1992,Kant and the Dynamics of Reason: Essays onthe Structure of Kant’s Philosophy, Oxford: Basil Blackwell(Chs. 7 & 8).
• Cohen, A., 2014, “Kant on the Ethics of Belief,”Proceedings of the Aristotelian Society, 114(3):317–334.
• –––, 2018, “Kant on Science andNormativity,”Studies in History and Philosophy ofScience, 71(October): 6–12.
• Deligiorgi, K., 2005,Kant and the Culture ofEnlightenment, Albany NY: State University of New YorkPress.
• Engstrom, S., 1992, Review of Onora O’Neill,Constructions of Reason,Ethics, 102(3):653–655.
• Ferrarin, A., 2015,The Powers of Pure Reason: Kant and theIdea of Cosmic Philosophy, Chicago: University of ChicagoPress.
• Förster, E., 1992, “Was darf ich hoffen? Zum Problemder Vereinbarkeit von theoretischer und praktischer Vernunft beiImmanuel Kant,”Zeitschrift für philosophischeForschung, 46(2): 168–185.
• Freudiger, J., 1996, “Kants Schlußstein: Wie dieTeleologie die Einheit der Vernunft stiftet,”Kant-Studien, 87(4): 423–435.
• Friedman, M., 1992a, “Causal Laws and Foundations of NaturalScience,” inThe Cambridge Companion to Kant, P. Guyer(ed.), Cambridge: Cambridge University Press, pp. 161–199.
• –––, 1992b,Kant and the ExactSciences, Cambridge, MA: Harvard University Press.
• –––, 1992c, “Regulative andConstitutive,”Southern Journal of Philosophy,30(Supplement): 73–102.
• –––, 2013,Kant’s Construction ofNature: A Reading of the Metaphysical Foundations of NaturalScience, Cambridge: Cambridge University Press.
• Gardner, S., 1999,Kant and the Critique of Pure Reason,London: Routledge.
• –––, 2006, “The Primacy of PracticalReason,” inA Companion to Kant, G. Bird (ed.), Oxford:Blackwell, pp. 259–274.
• Gava, G., 2018, “Kant, Wolff and the Method ofPhilosophy,” inOxford Studies in Early ModernPhilosophy (Volume VIII), D. Garber & D. Rutherford (eds.),Oxford: Clarendon Press, pp. 271–303.
• Gava, G. & M. Willaschek, forthcoming, “TheTranscendental Doctrine of Method of the Critique of PureReason,” inThe Kantian Mind, S. Baiasu & M.Timmons (eds.), London & New York: Routledge.
• Gelfert, A., 2006, “Kant on Testimony,”BritishJournal for the History of Philosophy, 14(4): 627–652.
• Grier, M., 2001,Kant’s Doctrine of TranscendentalIllusion, Cambridge: Cambridge University Press.
• Guyer, P., 1989, “The Unity of Reason: Pure Reason asPractical Reason in Kant’s Early Conception of theTranscendental Dialectic,”Monist, 72(2):139–167; reprinted as Ch. 2 of hisKant on Freedom, Law andHappiness, Cambridge: Cambridge University Press, 2000.
• –––, 1990, “Reason and ReflectiveJudgment: Kant on the Significance of Systematicity,”Noûs, 24(1): 17–43.
• –––, 1997, Review of Susan Neiman,The Unityof Reason,The Philosophical Review, 106(2):291–295.
• –––, 2000a, “Freedom as the Inner Value ofthe World,” Ch. 3 of hisKant on Freedom, Law andHappiness, Cambridge: Cambridge University Press, pp.96–125.
• –––, 2000b, “The Unity of Nature andFreedom: Kant’s Conception of the System of Philosophy,”inThe Reception of Kant’s Critical Philosophy, S.Sedgwick (ed.), Cambridge: Cambridge University Press, pp.19–53.
• –––, 2006, “Bridging the Gulf:Kant’s Project in the Third Critique,” inA Companionto Kant, G. Bird (ed.), Oxford: Blackwell, pp.423–440.
• Guyer, P. & R. Walker, 1990, “Kant’s Conception ofEmpirical Law,”Proceedings of the Aristotelian Society(Supplementary Volumes), 64: 221–258.
• Herman, B., 1993,The Practice of Moral Judgment,Cambridge MA: Harvard University Press.
• –––, 2007,Moral Literacy, CambridgeMA: Harvard University Press.
• Hieronymi, P., 2021, “Reasoning First,” inRoutledge Handbook of Practical Reasoning, R. Chang & K.Sylvan (eds.), Abingdon: Routledge, pp. 349–65.
• Kleingeld, P., 1995, “What Do the Virtuous Hope for?Re-reading Kant’s Doctrine of the Highest Good,” inProceedings of the Eighth International Kant Congress (Volume1), H. Robinson (ed.), Milwaukee: Marquette University Press, pp.91–112.
• –––, 1998a, “The Conative Character ofReason in Kant’s Philosophy,”Journal of the Historyof Philosophy, 36(1): 77–97.
• –––, 1998b, “Kant on the Unity ofTheoretical and Practical Reason,”Review ofMetaphysics, 52(2): 311–339.
• –––, 2010, “Moral consciousness and the‘fact of reason’,” inKant’s Critique ofPractical Reason: A Critical Guide, A. Reath & J. Timmermann(eds.), Cambridge: Cambridge University Press, pp. 55–72.
• –––, 2016, “Kant on ‘Good’,the Good, and the Duty to Promote the Highest Good,” inTheHighest Good in Kant’s Philosophy, T. Höwing (ed),Berlin: De Gruyter, pp. 33–49.
• Klemme, H., 2014. “Is the Categorical Imperative the HighestPrinciple of Both Pure Practical and Theoretical Reason?”Kantian Review, 19(1): 119–126.
• Korsgaard, C., 1996,Creating the Kingdom of Ends,Cambridge: Cambridge University Press.
• –––, 1997, “The Normativity ofInstrumental Reason,” inEthics and Practical Reason,G. Cullity & B. Gaut (eds.), Oxford: Oxford University Press, pp.215–254; reprinted as Ch. 1 of her 2008, pp. 27–68.
• –––, 2008,The Constitution of Agency:Essays on Practical Reason and Moral Psychology, New York: OxfordUniversity Press.
• Lenczewska, O., 2021, “Becoming Pluralists: Kant on theNormative Features of Pluralistic Thinking,”KantYearbook, 13(1): 107–28.
• Łuków, P., 1993, “The Fact of Reason:Kant’s Passage to Ordinary Moral Knowledge,”Kant-Studien, 84(2): 204–221.
• MacIntyre, A., 1988,Whose Justice? Which Rationality?,London: Duckworth.
• Mikalsen, K. K., 2010, “Testimony and Kant’s Idea ofPublic Reason,”Res Publica, 16(1): 23–40.
• Møller, S., 2020,Kant’s Tribunal of Reason:Legal Metaphor and Normativity in the Critique of Pure Reason.Cambridge: Cambridge University Press.
• Mudd, S., 2016, “Rethinking the Priority of Practical Reasonin Kant,”European Journal of Philosophy, 24(1):78–102.
• –––, 2017, “The Demand for Systematicityand the Authority of Theoretical Reason in Kant,”KantianReview, 22(1): 81–106.
• Neiman, S., 1994,The Unity of Reason: Rereading Kant,New York: Oxford University Press.
• Newton, A., 2017, “The Analytic Proposition UnderlyingKantian Hypothetical Imperatives,”Kant-Studien,108(4): 543–67.
• Nuzzo, A., 2005,Kant and the Unity of Reason, WestLafayette, IN: Purdue University Press.
• O’Neill, O., 1984, “Transcendental Synthesis andDevelopmental Psychology,”Kant-Studien, 75(2):149–167.
• –––, 1989,Constructions of Reason,Cambridge: Cambridge University Press.
• –––, 1990, “Enlightenment as Autonomy:Kant’s Vindication of Reason,” inThe Enlightenmentand its Shadows, P. Hulme & L. Jordanova (eds.), London:Routledge, pp. 184–199.
• –––, 1992, “Vindicating Reason,” inThe Cambridge Companion to Kant, P. Guyer (ed.), Cambridge:Cambridge University Press, pp. 280–308; reprinted as Ch. 1 ofO’Neill 2015.
• –––, 1996,Towards Justice and Virtue,Cambridge: Cambridge University Press.
• –––, 2000, “Four Models of PracticalReason,” in herBounds of Justice, Cambridge: CambridgeUniversity Press, pp. 11–28; German version, 1994, “VierModelle praktischer Vernunft,” inVernunftbegriffe in derModerne, H. F. Fulda & R. Horstmann (eds.), Stuttgart:Klett-Cotta, pp. 586–606.
• –––, 2002, “Autonomy and the Fact ofReason in theKritik der praktischen Vernunft,30–41,” inImmanuel Kant, Kritik der praktischenVernunft, O. Höffe (ed.), Berlin: Akademie Verlag, pp.81–97.
• –––, 2004, “Kant: Rationality as PracticalReason,” inThe Oxford Handbook of Rationality, A. Mele& P. Rawling (eds.), Oxford: Oxford University Press, pp.93–109; reprinted as Ch. 2 of O’Neill 2015.
• –––, 2015,Constructing Authorities: Reason,Politics and Interpretation in Kant’s Philosophy,Cambridge: Cambridge University Press.
• Patrone, T., 2008,How Kant’s Conception of ReasonImplies a Liberal Politics: An Interpretation of the “Doctrineof Right”, Lewiston NY: Edwin Mellen Press.
• Pollok, K., 2017,Kant’s Theory of Normativity:Exploring the Space of Reason, Cambridge: Cambridge UniversityPress.
• Prauss, G., 1981, “Kants Problem der Einheit theoretischerund praktischer Vernunft,”Kant-Studien, 72(3):286–303.
• Rauscher, F., 1998, “Kant’s Two Priorities ofPractical Reason,”British Journal for the History ofPhilosophy, 6(3): 397–419.
• Reath, A., 1988, “Two Conceptions of the Highest Good inKant,”Journal of the History of Philosophy, 26(4):593–619.
• –––, 2013, “Formal Approaches toKant’s Formula of Humanity,” inKant on PracticalJustification: Interpretive Essays, S. Baiasu & M. Timmons(eds.), Oxford: Oxford University Press, pp. 201–228.
• Rescher, N., 2000,Kant and the Reach of Reason:Studies in Kant’s Theory of Rational Systemization,Cambridge: Cambridge University Press.
• Ripstein, A., 2021,Kant and the Law of War, New York:Oxford University Press.
• Ruffing, M., 2015, “Kants ‘Stimme derVernunft’—Analyse einer unauffälligenMetapher,” inKant and the Metaphors of Reason, P.Kauark-Leite, G. Cecchinato, V. de Araujo Figuereido, M. Ruffing &A. Serra (eds.), Hildesheim: Olms, pp. 195–204.
• Russell, F., 2020, “Kantian Self-Conceit and the Two Guisesof Authority,”Canadian Journal of Philosophy, 50(2):268–83.
• Saner, H., 1967/73,Kant’s Political Thought: ItsOrigins and Development, E. Ashton (trans.), Chicago: ChicagoUniversity Press.
• Schadow, S., 2022, “Acting for a Reason. What Kant’sConcept of Maxims Can Tell Us about Value, Human Action, and PracticalIdentity,” inKant’s Theory of Value, C. Horn& R. Dos Santos (eds.), Berlin: De Gruyter, pp. 65–88.
• Sticker, M., 2021, Rationalizing (Vernünfteln).Elements in the Philosophy of Immanuel Kant, Cambridge:Cambridge University Press. doi:10.1017/9781108625661
• Stoddard, E., 1988, “Reason on Trial: Legal Metaphors in theCritique of Pure Reason,”Philosophy andLiterature, 12(2): 245–260.
• Timmermann, J., 2009, “The Unity of Reason: KantianPerspectives,” inSpheres of Reason, S. Robertson(ed.), Oxford: Oxford University Press, pp. 183–198.
• –––, 2010, “Reversal or retreat?Kant’s Deductions of Freedom and Morality,” inKant’s Critique of Practical Reason: A Critical Guide,A. Reath & J. Timmermann (eds.), Cambridge: Cambridge UniversityPress, pp. 73–89.
• Velkley, R., 1989,Freedom and the Ends of Reason: On theMoral Foundations of Kant’s Critical Philosophy, Chicago:University of Chicago Press.
• Walker, R., 1989,The Coherence Theory of Truth: Realism,Anti-Realism, Idealism, London: Routledge.
• Wartenberg, T., 1992, “Reason and the Practice ofScience,” inThe Cambridge Companion to Kant, P. Guyer(ed.), Cambridge: Cambridge University Press, pp. 228–248.
• Westphal, K., 2011, “Kant’s Moral Constructivism andRational Justification,” inPolitics and Metaphysics inKant, S. Baiasu, S. Pihlström & H. Williams (eds.),Cardiff: University of Wales Press, pp. 28–46.
• Willaschek, M., 2010, “The Primacy of Practical Reason andthe Idea of a Practical Postulate,” inKant’s Critiqueof Practical Reason: A Critical Guide, A. Reath & J.Timmermann (eds.), Cambridge: Cambridge University Press, pp.168–196.
• Wood, A., 1970,Kant’s Moral Religion, Ithaca NY:Cornell University Press.
• –––, 1992, Review of Onora O’Neill,Constructions of Reason,Philosophical Review,101(3): 647–650.
• Ypi, L., 2021,The Architectonic of Reason: Purposiveness andSystematic Unity in Kant’s Critique of Pure Reason, Oxford:Oxford University Press.

Academic Tools

	How to cite this entry.
	Preview the PDF version of this entry at theFriends of the SEP Society.
	Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO).
	Enhanced bibliography for this entryatPhilPapers, with links to its database.

Other Internet Resources

• Das Bonner Kant-Korpus, full text of the Akademie edition of Kant’s works, volumes1–23 (in German).
• Kant-Gesellschaft, the German Kant Society, publishers ofKant-Studien.
• North American Kant Society, comprehensive website with current Kant-related activities.
• U.K. Kant Society.

Related Entries

autonomy: in moral and political philosophy |constructivism: in metaethics |Kant, Immanuel: aesthetics and teleology |Kant, Immanuel: and Hume on causality |Kant, Immanuel: and Hume on morality |Kant, Immanuel: and Leibniz |Kant, Immanuel: critique of metaphysics |Kant, Immanuel: moral philosophy |Kant, Immanuel: philosophy of mathematics |Kant, Immanuel: philosophy of religion |Kant, Immanuel: philosophy of science |Kant, Immanuel: social and political philosophy |moral motivation |practical reason |practical reason: and the structure of actions |rationalism vs. empiricism |rationality: instrumental |reasoning: moral |reasons for action: agent-neutral vs. agent-relative |reasons for action: internal vs. external |reasons for action: justification, motivation, explanation

Acknowledgments

For comments on this entry over the years, my thanks to Carla Bagnoli,Graham Bird, Tatiana Patrone, Alison Stone, Lea Ypi, and the refereesfor this Encyclopedia, including R. Lanier Anderson and PaulGuyer. For additional assistance my thanks to Alix Cohen, SebastianGardner, Katharina Kraus, Onora O’Neill, Margit Ruffing, andJens Timmermann. My grateful thanks, too, to Nick Bunnin, fororganizing the Chinese philosophy summer school which gave me theopportunity to lecture on this topic, and to Diana Diaconescu forinvaluable research assistance with the most recent (2023)revisions.