Preface 1787

This is an important shift in Kant’s presentation. The 1781 preface was largely announcing the problem and defending the project. In the 1787 preface, Kant begins by giving us a criterion for what counts as genuine science. The key word is “certainty,” but not merely certainty in the psychological sense of feeling sure. Kant is interested in a method that produces an “undeviating” or “steadfast” progress, where the investigators know what path they are on, why they are on it, and can build upon results already secured. His test for metaphysics is therefore almost historical and methodological: look at what has happened to metaphysics. Has it accumulated secure results? Have its practitioners agreed upon a method? Can later thinkers proceed from established foundations rather than continually beginning again? Kant’s answer is plainly no. Metaphysics repeatedly prepares itself with elaborate arguments and then, when it approaches the goal, discovers that the road is uncertain. It retreats, tries another path, and produces another system. Hence the wonderful image of metaphysics “groping about in the dark.” The problem is not simply that metaphysicians disagree. The deeper problem is that they have not discovered the method that would make cumulative progress possible. Kant therefore says that even if the Critique does nothing more than indicate “the path” reason must travel, it will have performed an enormous service. Notice the modesty hidden inside the ambition: Kant is willing to abandon some of the traditional aims of metaphysics if that is what is required to put reason onto a genuinely scientific path.

The comparison with science is therefore not accidental. Kant is writing after the extraordinary success of Newtonian physics, and one of his central philosophical questions is: why has mathematics and natural science achieved stable progress while metaphysics has remained a battlefield? The question is not “Why are scientists smarter than philosophers?” It is “What is structurally different about the kinds of cognition involved?” Kant wants to understand what makes a science capable of securing its method. This is one reason the Critique will repeatedly return to mathematics and physics as models. Geometry does not continually reinvent the meaning of proof. Mathematics develops through determinate procedures. Newtonian physics has principles that allow researchers to derive and test consequences. Metaphysics, by contrast, keeps discovering that its own supposed principles are contestable. Kant therefore wants to discover whether metaphysics can undergo something like a scientific revolution—not by turning metaphysics into physics, but by discovering the conditions under which its own claims can be justified.

Then comes one of the most famous passages in the entire preface: logic. Kant says that logic has advanced along the “sure course of science” from the earliest times, and that since Aristotle it has apparently been unable to advance any further because it has reached completion. The German says that logic has “seit Aristoteles keinen Schritt rückwärts hat tun dürfen,” traditionally rendered as having made no backward step, while the sense of the passage is that logic has apparently remained complete since Aristotle. This claim has generated enormous historical debate, and we should be cautious about taking it literally as an accurate history of logic. Kant is not writing a history of formal logic in the modern sense. He is using “logic” in the inherited Aristotelian and post-Aristotelian sense of a formal science concerned with the forms of thought. His point is that once logic abstracts completely from the objects being thought about, it reaches an extraordinary degree of certainty. It does not ask whether the object is a tree, a number, a planet, or God. It asks whether a conclusion follows validly from premises, whether a judgment has a certain logical form, whether thought conforms to its formal rules. Because the content is bracketed, the science can be extraordinarily stable.

This is why Kant becomes almost hostile toward attempts to “enlarge” logic by adding psychology, metaphysics, or anthropology. He gives examples such as imagination and wit, the origin of knowledge, different kinds of certainty, idealism, skepticism, prejudices, their causes and remedies. His complaint is not that these are uninteresting subjects. Quite the opposite: they are legitimate subjects, but they belong to different sciences. If you introduce psychology into logic, you are no longer doing pure logic. If you investigate the historical origin of knowledge, you have moved into another discipline. If you investigate the causes of prejudice, you are doing anthropology or psychology. Kant says that when sciences are allowed to “run into one another,” they are not thereby enriched; they are “disfigured.” This insistence on disciplinary boundaries is fundamental to the entire critical project. Kant is constantly asking: what kind of question is this, what faculty is responsible for answering it, what are the conditions of its legitimacy, and what happens when one domain illegitimately trespasses into another?

Here we should notice something subtle about Kant’s use of “formal.” Logic abstracts from the content of cognition and concerns itself with the form of thought as such. If I say “All A are B; C is A; therefore C is B,” logic does not care what A, B, and C stand for. The argument can concern dogs, numbers, planets, or imaginary creatures. The validity is formal. This is why Kant says logic applies whether cognition is a priori or empirical, regardless of its origin or object. Logic does not investigate whether the concepts being used successfully refer to objects. It investigates whether thought conforms to the formal laws of thought. This distinction will become enormously important when Kant turns from general logic to transcendental logic. General logic abstracts from all objects; transcendental logic cannot do that, because it wants to investigate those a priori concepts and principles through which objects can be thought. That is going to be one of the major transitions in the Critique.

And this is where Aristotle becomes particularly important for our larger project. Kant’s remark about logic reaching completion with Aristotle is an invitation to look backward. Aristotle’s Organon established a systematic investigation of forms of reasoning, especially the syllogism, and developed a vocabulary of logical relations that dominated Western philosophy for centuries. Kant inherits this tradition but asks a new question: if general logic tells us about the formal functions of judgment, can we use those functions to identify the fundamental concepts through which the understanding thinks objects? This is the path that leads to Kant’s “table of judgments” and eventually his “table of categories.” The famous Kantian categories—unity, plurality, totality; reality, negation, limitation; substance, causality, community; possibility, existence, necessity—are not simply a revised list of Aristotle’s categories. Kant derives his table from the forms of judgment. That is a major transformation. Aristotle’s categories are fundamentally categories of what can be predicated of beings; Kant’s categories become pure concepts of the understanding that structure objects of possible experience. We should keep that distinction very much alive as we read.

The next sentence gives us Kant’s explanation of why logic achieved its success so early: “the narrowness of its field.” This is a beautiful phrase because Kant is reversing what might ordinarily seem like a weakness. We might think a science is impressive if it covers an enormous territory. Kant says that logic’s very narrowness is what allows its certainty. Logic can abstract away from all objects and deal only with the understanding’s own forms. It is like a closed laboratory in which the variables have been drastically reduced. Once we remove the question of what the objects are, we can study the formal operations of thought with extraordinary precision. But the moment reason has to deal with objects, the difficulty changes completely. Now we have to ask not only whether our thinking is internally consistent but whether our concepts legitimately apply to something given outside thought.

This is the crucial distinction: logical validity is not the same thing as objective validity. I can construct a perfectly valid argument from premises that have no legitimate application to experience. Logic can tell me that if A implies B and A is true, then B follows. But logic by itself cannot tell me whether A actually corresponds to an object of experience. This is exactly the problem Kant will confront with metaphysics. A metaphysical argument can be internally impeccable and still fail to establish that its concepts have legitimate objective application. The fact that I can think a concept consistently does not mean that there is an object corresponding to it. This is why Kant’s Critique has to go beyond general logic. He needs a science that can investigate the relation between the pure forms of cognition and objects of possible experience.

This is the point where the word “propædeutic” becomes significant. Kant calls logic a “propædeutic,” from the Greek προπαιδευτικός, meaning preliminary instruction or preparation. Logic is the vestibule of the sciences. The vestibule is not the building itself, but you have to pass through it to enter. Logic teaches us how to judge correctly at the formal level, but it does not provide substantive knowledge about objects. Kant says that “real, substantive knowledge” belongs to the sciences properly so called—the objective sciences. This creates an architectural sequence that is going to become important: general logic gives us the formal conditions of correct thought; transcendental philosophy investigates the a priori conditions under which objects can be cognized; the particular sciences then provide determinate knowledge within their legitimate domains.

But we should notice a potential tension already appearing. Kant says logic concerns thought regardless of its object, while the sciences concern objects. Yet the Critique itself is going to investigate pure reason and understanding. So where does it fit? It cannot simply be ordinary logic, because ordinary logic abstracts from objects. But it cannot be empirical science either, because Kant is investigating a priori conditions. This is precisely why Kant needs the concept of “transcendental logic.” Transcendental logic will occupy the strange middle ground between formal logic and empirical science. It will investigate the a priori conditions under which objects can be thought and experienced. That is one of the central innovations of the Critique.

There is also a fascinating historical irony in Kant’s statement about Aristotle. Modern logic, of course, would eventually make enormous advances after Kant—Boole, Frege, Russell, mathematical logic, model theory, and so on. So Kant’s historical claim was plainly not correct if interpreted as a prediction about logic as a whole. But it tells us something about what Kant meant by logic in 1787. He had not yet imagined the mathematical transformation of logic that would occur in the nineteenth and twentieth centuries. For him, the essential achievement of logic was the formal analysis of judgment and inference. What matters philosophically is not whether his historical prediction survives; it is the structural point he is making: a discipline becomes scientifically secure when its object and method are clearly delimited. That principle remains important even if the particular history he gives us does not.

And this brings us back to the deepest concern of the preface. Kant is asking why metaphysics has not achieved the kind of stability that logic possesses. The answer cannot simply be “because metaphysicians are careless.” The problem is that metaphysics has been trying to deal with objects while lacking an adequate account of the conditions under which those objects can legitimately be thought. Logic can abstract from objects entirely; metaphysics cannot. So metaphysics needs something that logic alone cannot provide: an investigation into the relationship between the forms of cognition and objects. That is the territory of the Critique. The question will therefore become: what must the mind contribute in order for there to be an object of experience at all?

There is a beautiful progression here that I think is worth holding onto as we proceed. General logic asks: “Given these thoughts, what follows correctly?” Transcendental philosophy asks: “What makes objective thought possible in the first place?” Empirical science asks: “What are the determinate objects and laws we discover within experience?” Traditional dogmatic metaphysics often tried to leap directly from pure concepts to ultimate reality, bypassing the critical question of legitimacy. Kant’s intervention is to insert the missing investigation. Before reason can claim knowledge of objects, reason must investigate its own conditions of legitimate object-directed cognition.

And this is precisely why the Second Edition is so important. Kant is sharpening the architecture of the First Critique after having had six years to think about the reception and difficulties of the 1781 edition. The 1787 preface is more explicitly concerned with the scientific status and method of the Critique. The famous slogan that will soon appear—“thoughts without content are empty, intuitions without concepts are blind”—belongs to this same architectural concern. Neither pure thought nor raw givenness is sufficient for human knowledge. Knowledge requires their synthesis. That is the problem Kant is now preparing us to understand.

So as we enter the Second Edition, I would keep three questions running simultaneously in the background. First: what makes a science genuinely scientific rather than merely speculative? Second: what is the precise difference between formal logic and transcendental logic? And third: how can a priori concepts legitimately apply to objects? The first question concerns method, the second concerns the structure of cognition, and the third is ultimately the great problem of the Critique. Kant thinks that if he can answer all three, he can explain both the extraordinary success of mathematics and natural science and the persistent failure of traditional metaphysics. And that is the real ambition of the book: not merely to tell metaphysics to stop making certain claims, but to discover why some forms of knowledge can advance securely while others continually lose their way.

This is one of the most important passages in the entire preface because Kant is now giving us his model for what a successful science looks like. He is asking us to compare metaphysics with mathematics and physics and, more specifically, to ask what happened when those sciences discovered their “sure course.” The crucial distinction in the opening sentence is between two ways in which a priori cognition can relate to its object. Kant says it can either “determine the conception of the object,” where the object itself is supplied from elsewhere, or “establish its reality.” The first is theoretical rational cognition; the second is practical rational cognition. The German distinction is between theoretische and praktische Erkenntnis der Vernunft. This is a very early appearance of the division that will eventually become fundamental to Kant’s entire philosophy. Theoretical reason asks what we can know about objects; practical reason concerns what reason can determine as something to be done, and therefore concerns the actuality of freedom, morality, and ultimately the conditions under which the good can be realized. But Kant immediately gives us a methodological rule: in either case, the a priori element must be isolated and investigated first. We must distinguish what reason contributes independently of experience from what comes from other sources. Otherwise, “irremediable confusion” results. This is essentially the critical method in miniature: separate the ingredients before asking what knowledge is possible.

The phrase “must determine the conception of the object” needs particular attention. Kant is not saying that pure reason simply invents the object. He means that in theoretical rational cognition, reason can determine what an object must be like according to a priori principles, while the actual object may have to be supplied through some other source. Mathematics is the clearest example. When the geometer constructs a triangle, the concept of the triangle is not obtained by simply looking at nature and generalizing from observed triangles. The geometer works with a priori conditions of construction. Physics, however, has a more complicated status. Its laws contain a priori elements—most famously principles such as causality and the mathematical formulation of nature—but physics also depends on empirical observation. Thus Kant says mathematics is “purely a priori,” whereas physics is “partially so.” This is the seed of his famous conception of synthetic a priori knowledge. The astonishing question Kant is approaching is: how can there be knowledge that is both necessary and informative about objects? Pure mathematics seems to give us necessity without relying on particular observations, while physics seems to give us universal laws about nature that are not simply analytic definitions. Kant wants to understand how this is possible.

Then he turns to mathematics and gives us one of the most famous philosophical stories in the Critique: the Greek revolution in mathematical thinking. Kant says that mathematics had entered “the sure course of science” among the Greeks, but he immediately qualifies this by saying that this road was not simply waiting there ready to be discovered. Mathematics had to discover how to construct its own method. Logic had an easier path because logic “has only to deal with itself.” Mathematics has to deal with objects, or at least with objects constructed in intuition. This is why Kant’s story of Thales matters so much. The traditional story attributes to Thales the demonstration that the angles opposite the equal sides of an isosceles triangle are equal. Kant says that whether the actual discoverer was Thales is uncertain, but the philosophical significance of the story is what matters. Something changed in the way mathematical knowledge was conceived. The mathematician stopped merely contemplating a figure as something given and instead began to construct the object according to a rule.

The word “construction” is absolutely central here. Kant’s German is “Konstruktion.” For Kant, mathematical cognition is not merely conceptual cognition. If I have the concept of a triangle, I can think of the general characteristics of triangles, but that alone does not give me mathematical cognition in Kant’s full sense. Mathematical cognition requires that the concept be exhibited in intuition through construction. I construct a particular triangle, not because that particular triangle is itself the object of ultimate interest, but because the construction exhibits the general rule contained in the concept. This is why Kant will later say that philosophical cognition proceeds through concepts, while mathematical cognition proceeds through the construction of concepts. That distinction is enormously important. The mathematician does not merely analyze what is already contained in a concept. The mathematician generates an intuition according to the concept and then discovers necessary relations in that constructed intuition.

This also clarifies the extraordinary sentence that the first mathematician had to “produce these properties, as it were, by a positive a priori construction.” Kant is not saying that the mathematician creates the properties arbitrarily. Quite the opposite. The mathematician constructs under constraints. He cannot simply attribute any property he likes to the triangle. Kant says the mathematician must “not attribute to the object any other properties than those which necessarily followed from that which he had himself, in accordance with his conception, placed in the object.” This is a remarkably subtle formulation. The mathematician contributes something to the object, but what follows from that contribution is not arbitrary. If I construct an isosceles triangle according to the relevant definition, certain properties follow necessarily. The necessity comes from the structure of the construction. The object of mathematical cognition is therefore not simply passively observed. It is constructed under a priori rules.

This is one of the places where Kant’s relation to Aristotle and Leibniz becomes fascinating. Aristotle’s geometry and logic inherited from the ancient Greek tradition certainly supplied Kant with the model of rigorous demonstration, but Kant’s conception of mathematical construction is also a departure from purely conceptual rationalism. Leibniz, for example, placed enormous emphasis on conceptual analysis and logical necessity. Kant thinks conceptual analysis alone cannot explain the character of mathematical cognition. The proposition that the angles of an isosceles triangle are equal is not merely obtained by unpacking the concept “isosceles triangle.” Something has to be constructed in intuition. This is part of Kant’s larger argument that cognition requires two distinct capacities: concepts and intuitions. Concepts provide rules; intuitions provide something given or exhibited. Mathematics is powerful because it combines the a priori structure of concepts with construction in pure intuition.

This is also where space is beginning to enter the picture, even though Kant has not yet explicitly given us the Transcendental Aesthetic. Geometry, for Kant, is possible because space is not merely an empirical generalization extracted from observations of physical objects. Space is a pure form of human sensibility. When the geometer constructs a triangle, the construction occurs in pure spatial intuition. This will eventually lead Kant to the extraordinary claim that geometry yields synthetic a priori cognition because spatial intuition itself is an a priori form through which objects can be given to us. So this little story about Thales is actually preparing the ground for one of the deepest claims in the Critique: the necessity of geometry tells us something about the structure of human cognition.

We should also be careful about Kant’s historical claim. The modern historical picture of mathematics is much more complicated than Kant’s account suggests. The attribution of the isosceles-triangle theorem to Thales is traditional, and Diogenes Laertius is a late source rather than direct historical evidence from Thales himself. Kant knows this to some degree—he explicitly says “Thales, or whatever may have been his name.” So the story is functioning philosophically rather than as a precise history of mathematics. Kant is constructing a thought experiment about a “revolution” in mathematical method. He wants us to imagine the moment when the mathematician stopped asking, “What does this figure appear to have?” and started asking, “What follows necessarily from a construction made according to a rule?” That is the conceptual revolution he cares about.

And then Kant gives us the famous phrase “royal road.” Mathematics has found a royal road to science, but it found it through construction. This phrase will matter when we compare it to physics. Physics, Kant says, took much longer to enter the highway of science. Here he invokes Francis Bacon. Again, Kant is not claiming that Bacon single-handedly invented experimental science. He explicitly says others were already on the right track and that Bacon “imparted fresh vigour” to the pursuit. What matters is the methodological revolution: nature should no longer be approached as something before which the investigator simply waits passively for revelations. The investigator must actively interrogate nature according to principles.

This is where Kant’s famous “revolution in the way of thinking” is beginning to emerge. The mathematical revolution involved constructing the object according to a priori rules. The scientific revolution in physics involves something analogous: the investigator approaches nature with questions determined by reason and subjects the resulting answers to experimental testing. Nature is not simply allowed to dictate its laws through passive observation. The scientist constructs an experimental situation and asks nature a question. This is an extraordinarily important idea in Kant. He will later say that reason must approach nature “not in the character of a pupil who listens to everything the teacher chooses to say, but of an appointed judge who compels the witness to answer questions which he himself has formulated.” That is the mature form of the same thought appearing here. Science progresses when reason actively structures its encounter with objects.

And this gives us a very interesting parallel between mathematics and physics. In mathematics, reason constructs an object in pure intuition. In physics, reason constructs an experimental situation in empirical intuition. In both cases, cognition does not consist in passively copying what is given. The knower contributes a structure or question to the encounter. But the two sciences differ because mathematics can remain entirely within pure a priori construction, while physics must ultimately consult experience. Kant is therefore not an idealist in the simplistic sense that “the mind makes everything up.” The empirical world still matters enormously. The scientist cannot simply decide what nature will do. Reason constructs the question and the experimental conditions, but nature supplies the answer.

This is why the sentence “I shall confine myself to the empirical side of natural science” is interesting. Kant is about to discuss the revolution in physics, and he is deliberately setting aside the pure mathematical or metaphysical foundations of physics for the moment. He wants to point toward the experimental transformation associated with Bacon, Galileo, Torricelli, and others. The famous examples will come in the next paragraphs. Galileo’s inclined plane and Torricelli’s atmospheric-pressure experiment are especially important because Kant interprets them as cases where the investigator actively interrogates nature rather than merely observing it. The scientist goes to nature with a conceptual framework and an experimental question.

This is also where Kant’s “Copernican revolution” is being prepared, even though he has not yet used the famous phrase. The revolutionary thought will be: perhaps instead of assuming that our cognition must conform to objects, we should investigate whether objects of experience conform, in some respect, to the conditions of our cognition. In mathematics, we construct the object according to the rules of our intuition. In natural science, we approach nature with concepts and experimental procedures that determine the questions we can ask. Kant is going to transpose this methodological insight into transcendental philosophy. Instead of asking how our cognition passively conforms to independently given objects, ask what must be true of objects as experienced if they are to be objects for a cognition structured in this way.

There is an important difference, however, between Kant’s mathematical and physical examples. In mathematics, the construction can be a priori. In physics, empirical material remains indispensable. That distinction is going to prevent us from turning Kant into a crude subjectivist. Kant is not saying that because the scientist structures experience, nature is merely a projection of the mind. He is saying something more precise: the objects we can experience are necessarily encountered through certain forms and concepts supplied by our cognitive faculties. The empirical content remains something we receive; the formal conditions under which it can become an object of knowledge are supplied by us. This is the beginning of the famous Kantian picture in which sensibility gives us intuitions while understanding supplies concepts.

There is also a fascinating connection to Hume here. Hume had forced Kant to confront the problem of causality: experience gives us sequences of events, but where does the necessity of causal connection come from? Kant’s response will eventually be that causality is not learned by observing repeated conjunctions. Rather, causality is one of the a priori concepts through which the understanding organizes objective experience. But this passage shows why Kant can make that claim without abandoning empiricism altogether. Physics still requires experience. What reason supplies is the form under which empirical events can be understood as lawful objects. Kant is therefore trying to preserve the empirical content of science while explaining its necessary structure.

And there is a beautiful continuity with the passage we just read on logic. Logic achieves certainty by abstracting from objects and examining the formal laws of thought. Mathematics achieves certainty by constructing objects in pure intuition according to a priori concepts. Physics achieves scientific progress by bringing a priori principles into an active, experimental relation with empirical objects. Metaphysics now has to ask: what are the a priori principles that make this entire relation between cognition and objects possible? That is the enormous question Kant is about to take up.

So I would mark this passage as one of the major turning points of the book. Kant has now given us three models of scientific progress: logic, mathematics, and physics. Logic succeeds because it deals only with the formal structure of thought. Mathematics succeeds because it learns to construct its objects a priori in intuition. Physics succeeds because reason actively interrogates nature according to principles and tests the consequences experimentally. Kant’s question is whether metaphysics can undergo a comparable revolution. Can reason stop simply speculating about objects beyond experience and instead investigate the conditions under which objects can be known at all? If it can, then metaphysics might finally leave its “groping about in the dark” and enter a genuinely scientific path.

And the phrase I would carry forward from this entire passage is not actually “a priori.” It is “construct.” Kant is telling us that genuine knowledge is not simply passive reception. The mathematician con-structs; the physicist experimentally interrogates; and, at the transcendental level, Kant will ask what the knowing subject must contribute to make objective experience possible. That is the bridge from the scientific revolution to the Copernican revolution. The great question will become: if cognition contributes something to the constitution of experience, what exactly does it contribute—and how far does that contribution legitimately extend?

This is the passage where Kant’s entire project begins to come into focus. He has just told us that mathematics found its “royal road” through construction, and now he gives us the corresponding story for physics: Galileo, Torricelli, and Stahl. But notice what Kant thinks the scientific revolution actually consisted in. It was not simply the discovery of new facts. It was a transformation in the relationship between reason and nature. The scientist ceased to regard nature as something before which reason must passively wait, and instead began to approach nature with a question already structured by reason. This is what Kant means when he says that “reason only perceives that which it produces after its own design.” The German is particularly striking: “Die Vernunft sieht nur ein, was sie selbst nach ihrem Entwurfe hervorbringt.” Vernunft sieht ein is stronger than simply “sees.” It means something like “understands,” “grasps,” or “sees into.” And Entwurf is a design, draft, or projected plan. So Kant’s claim is not merely that scientists manipulate experiments. It is that cognition becomes scientific when reason has a prior structure or design according to which it can interrogate what is given.

The phrase “after its own design” can easily be misunderstood as saying that the scientist invents reality. Kant does not mean that. The experiment does not make nature obey whatever we happen to want. Rather, the scientist determines the conditions under which a particular question can be answered. Galileo does not simply watch falling bodies and collect impressions. He constructs an experimental situation in which a mathematical relation can be isolated. Torricelli does not simply observe atmospheric phenomena. He sets up a situation in which the pressure of the atmosphere can be measured against a known column of mercury. Stahl’s chemical experiments, as Kant understands them, similarly involve the controlled addition and subtraction of elements. In each case, nature is still providing the empirical answer. But reason determines the form of the question and the conditions under which that answer can become intelligible.

This is why the pupil/judge metaphor is so powerful. “Reason must approach nature,” Kant says, “not, however, in the character of a pupil, who listens to all that his master chooses to tell him, but in that of a judge, who compels the witnesses to reply to those questions which he himself thinks fit to propose.” The German is even more forceful: reason must not “allow itself to be led around” by nature but must proceed with principles of judgment according to fixed laws. Nature becomes the witness, reason becomes the judge. But we should not read this as human arrogance toward nature. A judge does not invent the testimony. The witness still has to answer. Rather, the judge determines which questions must be asked in order for the testimony to become legally meaningful. That is Kant’s model of scientific inquiry. Nature gives us appearances, but reason supplies the form of interrogation through which those appearances can become knowledge.

This is already very close to the transcendental method. Kant is not yet asking, “What are the laws of nature?” He is asking something deeper: “What must be true of our cognition for the experience of nature as lawful to be possible?” This is the transition from the scientific revolution to the Copernican revolution. Galileo and Newton revolutionized our understanding of nature by changing how we interrogated it. Kant wants to perform an analogous revolution in philosophy by changing how we interrogate cognition itself. Instead of asking only whether our cognition conforms to objects, he will ask whether the objects we can experience necessarily conform to certain conditions of cognition. That is the radical move that the Critique will make.

The sentence about accidental observations is particularly important: “For accidental observations, made according to no preconceived plan, cannot be united under a necessary law.” Here Kant is making a distinction between a sequence of observations and a law. Suppose I observe that every time A occurs, B follows. I can accumulate a thousand observations and still not possess necessity simply by virtue of accumulation. What turns observations into scientific knowledge is the principle that allows them to be understood as instances of a law. This is where Kant is quietly confronting the empiricist problem that Hume had made famous. Hume had argued that repeated experience can never itself reveal necessary connection. We observe conjunction, succession, and regularity, but necessity is not something given in the sensory impressions. Kant agrees with the basic difficulty. But rather than conclude that necessity is merely a psychological habit, Kant asks whether the necessity comes from the structures through which the understanding organizes experience. That will eventually lead us to causality as one of the categories.

So when Kant says that “only the principles of reason can give to concordant phenomena the validity of laws,” he is making a much stronger claim than “science uses theories.” He is saying that a lawful nature, as an object of scientific cognition, requires principles supplied by reason. Empirical observations alone do not generate necessity. There has to be an a priori contribution. This is one of the main reasons Kant thinks physics contains both empirical and a priori elements. The empirical content comes from observation; the lawful form in which those observations can constitute objective knowledge involves principles that are not themselves derived from individual observations.

This also helps explain the earlier distinction between mathematics and physics. In mathematics, reason can construct its object a priori. In physics, reason cannot construct nature itself, but it can construct an experimental situation and interrogate nature according to mathematical and conceptual principles. The two sciences therefore share something fundamental: neither is simply passive. Mathematical cognition actively constructs its object in pure intuition; physical cognition actively constructs experiments through which empirical nature can be questioned. In both cases, scientific reason is productive rather than merely receptive.

And now Kant makes the decisive turn: “We come now to metaphysics.” The mood changes immediately. Metaphysics is “a purely speculative science.” It is “entirely independent of the teachings of experience.” It deals with “mere conceptions,” unlike mathematics, which deals with concepts applied to intuition. And then comes the devastating phrase: “in it, reason is the pupil of itself alone.” Notice how perfectly this reverses the metaphor from physics. In natural science, reason became the judge and nature became the witness. In metaphysics, there is no empirical witness. Reason is both the investigator and the thing being investigated. It asks questions of pure reason using pure reason alone. This is precisely why metaphysics is so dangerous. There is no external experiment that can settle many of its disputes.

The German expression here, “die Vernunft ist … der Schüler seiner selbst,” is striking because it exposes the peculiar circularity of traditional metaphysics. Reason tries to discover truths beyond experience using concepts that are generated independently of experience. But how can it determine whether those concepts legitimately apply to objects beyond possible experience? It has no empirical tribunal available to it. This is why Kant says metaphysics has not yet found the “sure path of science.” The problem isn’t simply that metaphysicians are stupid or careless. The structure of the discipline itself creates the difficulty: it wants knowledge where the ordinary means of validating knowledge are unavailable.

The description of metaphysics as “the oldest of the sciences” is also important. Kant is not dismissing metaphysics as a foolish adolescent mistake. Quite the opposite. Metaphysics is ancient, persistent, and apparently ineradicable. Even if civilization were destroyed and all the empirical sciences disappeared, metaphysical questions would eventually return. Why? Because they arise from reason itself. Human beings inevitably ask about the ultimate ground of things, the totality of the world, God, freedom, the soul, necessity, possibility, and the unconditioned. This connects directly to the 1781 passage you gave earlier: reason is compelled to ask questions that it cannot simply decline. Metaphysics is therefore not an accidental intellectual hobby. It is generated by the nature of reason.

And this creates Kant’s central predicament. If metaphysics is impossible, then why does reason generate an apparently irresistible desire for it? If metaphysical questions were simply meaningless, we could dismiss them. But Kant does not think we can. Reason naturally seeks the “unconditioned.” Whenever we encounter something conditioned, reason asks for its condition; then it asks for the condition of that condition, and so on. Eventually reason seeks a condition that is itself unconditioned. This movement is not an optional philosophical preference. It is built into reason’s activity. That is why Kant’s eventual solution cannot simply be, “Stop asking metaphysical questions.” He has to explain why reason asks them and precisely where the legitimate use of reason ends.

This is why the language of “groping,” “standing still,” “retracing its steps,” and “abandoning the path” returns. Kant is testing metaphysics against the criteria established at the beginning of the preface. Does it have a secure method? No. Does it accumulate permanent results? No. Do its practitioners agree? No. Instead, metaphysics becomes “an arena specially adapted for the display of skill or the exercise of strength in mock-contests.” The German word translated as “mock-contests” is Scheingefechte: sham battles, apparent battles, combats that look real but accomplish nothing. This is a wonderfully precise diagnosis. Metaphysical systems can be extraordinarily sophisticated and intellectually impressive while nevertheless failing to produce stable possession of the territory they claim to conquer. One philosopher proves one position; another proves its opposite; centuries pass; the fundamental dispute remains.

This is where Kant’s “antinomies” will eventually become crucial. In the Transcendental Dialectic, he will show that pure reason can generate equally compelling arguments for contradictory positions: the world has a beginning in time and the world has no beginning; everything is composed of simple parts and nothing is composed of simple parts; freedom exists and everything occurs according to natural necessity; there is a necessary being and there is no necessary being. The point is not merely that philosophers happen to disagree. Kant wants to demonstrate that the structure of pure reason itself can generate these conflicts when it exceeds the limits of possible experience. The metaphysical battlefield is therefore not an accident. It has a systematic source.

And now Kant asks the question that will drive the rest of the Critique: “Why is it that, in metaphysics, the sure path of science has not hitherto been found?” There are two possibilities. Perhaps the path does not exist. Or perhaps philosophers have simply been looking in the wrong direction. Kant strongly resists the first possibility because it would make the deepest aspirations of reason strangely pointless. Why would nature give us a faculty that irresistibly seeks answers to the most important questions only to have that faculty endlessly deceive itself? This is a theological and philosophical question at once. Kant is unwilling to conclude that reason is fundamentally fraudulent.

But notice how carefully he formulates the alternative. He does not say, “Metaphysics must be possible because we desire it.” Desire is not evidence. Instead, he asks whether there are “indications” that can guide a renewed investigation. This is precisely where the scientific revolutions of mathematics and physics become models. Perhaps metaphysics has been approaching its objects incorrectly. Perhaps the problem lies not in reason’s desire for knowledge but in its conception of the relationship between cognition and objects.

This is the moment immediately before Kant’s most famous philosophical proposal. The question is going to become: what if we stop assuming that cognition must conform to objects and instead investigate whether objects, insofar as they are objects of experience, must conform to the conditions of our cognition? Kant will call this analogous to the Copernican revolution. Copernicus realized that certain astronomical problems became intelligible when the observer’s standpoint was transformed. Kant proposes a similar reversal at the level of knowledge.

There is an especially beautiful connection here to the phrase from the previous passage: “reason only perceives that which it produces after its own design.” In physics, this meant that reason approaches nature with principles and questions. Kant now wonders whether philosophy itself has failed because it has not applied this insight to cognition. Traditional metaphysics assumes that our concepts must conform to things as they are in themselves and then wonders why pure reason cannot secure knowledge of those things. Kant’s experiment is to reverse the relation. Let us suppose that the objects of possible experience must conform to certain structures supplied by our cognition. Then perhaps the necessity of mathematics, the universality of natural laws, and the limits of metaphysical knowledge can all be explained at once.

This is why I would resist reading “Kant is saying the mind creates reality” into this passage. That is too crude. His claim will be much more precise: the mind contributes certain forms and concepts that constitute the conditions under which something can appear as an object of possible experience. The thing in itself is not thereby created by us. What is constituted is the object as experienced and known by finite human cognition. Kant is trying to explain how objective knowledge can be possible without requiring us to pretend that we have direct intellectual access to reality independently of the conditions of our sensibility and understanding.

There is also an important connection to Aristotle here. Aristotle’s metaphysics asks about being qua being and about the principles and causes of beings. Kant inherits the ancient question of ultimate principles but shifts the locus of investigation. Before asking about the ultimate structure of being, Kant asks what makes cognition of objects possible. In a sense, he turns metaphysics back upon the knower. Likewise, with Leibniz, Kant inherits the idea that reason can discover necessary truths through pure concepts, but he becomes suspicious of the assumption that pure concepts automatically give us knowledge of objects beyond experience. Kant’s critical question is therefore: what is the legitimate objective use of a concept? A concept may be perfectly coherent and yet lack an object that can be given in experience.

This is also why the phrase “mere conceptions” should not be treated as an insult to concepts. Concepts are indispensable. The problem is that metaphysics, according to Kant, tries to make concepts do the work that in mathematics is accomplished through intuition and in physics through experience. A concept by itself does not guarantee an object. Later Kant will formulate the principle in a sentence that should remain almost like a mantra for our reading: “Thoughts without content are empty, intuitions without concepts are blind.” Pure concepts without something that can be given to them do not yield cognition of an object. But raw intuition without conceptual organization does not yield intelligible objective knowledge either.

So this passage gives us the basic diagnosis of the whole book. Science succeeds when reason discovers the proper relationship between its own principles and what is given to it. Logic succeeds by restricting itself to the formal conditions of thought. Mathematics succeeds by constructing objects in pure intuition. Physics succeeds by interrogating nature through rationally designed experiments. Metaphysics fails because it tries to obtain knowledge from pure concepts while bypassing the conditions under which objects can actually be given. Kant’s task will therefore be to determine exactly what pure reason can legitimately know, exactly where it overreaches, and why it nevertheless cannot stop generating the questions that take it beyond those limits.

And I think the deepest sentence here is the last question: “what indications do we possess to guide us in a renewed investigation?” Kant is telling us that the answer will not come from simply reading more metaphysical books. We have to examine the successful sciences and ask what methodological transformation made their success possible. The revolution in mathematics and the revolution in physics become clues for a revolution in metaphysics. The philosopher must become, in a sense, the experimentalist of reason itself. Instead of asking reason to tell us what lies beyond all possible experience, Kant will ask reason to submit itself to an experiment: what happens if we change our assumption about the relation between cognition and objects? The entire Critique is the unfolding of that experiment.

Yes. This is the decisive passage. We have reached what is usually called Kant’s “Copernican revolution,” although notice that Kant himself does not call it a “Copernican revolution” here. He calls it an “experiment” (Versuch) modeled on Copernicus. That word is worth preserving because Kant is presenting his entire transcendental philosophy as a kind of experiment in the standpoint from which we approach cognition. Everything we have read in the preface has been preparing for this moment. Mathematics succeeded because it discovered that the mathematician must actively construct according to a priori rules. Physics succeeded because reason approaches nature with principles and interrogates it experimentally rather than simply receiving whatever nature happens to show. Metaphysics has failed because it has assumed the opposite relationship: that cognition must conform to objects. Kant now asks whether the whole problem might be reversed. Instead of beginning with the assumption that our cognition must conform to independently conceived objects, suppose we ask whether objects, insofar as they are objects of possible experience, must conform to the structure of our cognition. The crucial German sentence is: “Bisher nahm man an, alle unsere Erkenntnis müsse sich nach den Gegenständen richten.” Literally: “Up to now it has been assumed that all our cognition must conform itself to the objects.” Then comes the reversal: “Man versuche es daher einmal, ob wir nicht in den Aufgaben der Metaphysik damit besser fortkommen, daß wir annehmen, die Gegenstände müssen sich nach unserem Erkenntnisse richten.” “Let us therefore try once whether we might not get further with the problems of metaphysics by assuming that the objects must conform to our cognition.” That “let us try” is important. Kant is not announcing an arbitrary dogma. He is proposing a methodological hypothesis and asking us to see what follows from it.

The first thing we need to clarify is what Kant means by “objects conforming to our cognition.” He does not mean that you personally invent a tree by thinking about it. He does not mean that the external world is imaginary. And he does not mean that human beings consciously manufacture reality. That interpretation would miss the precision of the argument. Kant is asking about objects as they can be known by finite human beings. An object of experience is never encountered without the conditions of our sensibility and understanding. Before an object can appear to us, it must already be received within the forms through which we can receive anything at all. Later Kant will argue that these forms are space and time. Then, once something is given through sensibility, it must be synthesized and thought through concepts supplied by the understanding. Thus, “the object conforms to cognition” means that whatever can count as an object for us must conform to the a priori structures through which we are capable of experiencing and thinking objects. The object in itself may be something we cannot know directly. The object as experienced is necessarily structured according to the conditions of our cognitive faculties.

This is why Kant says that the old assumption makes a priori knowledge almost impossible. If objects exist entirely independently of the structure of cognition, and cognition must simply conform itself to those objects, then how could we possibly know something about them before encountering them? If I have never encountered an object, how could my concepts somehow tell me in advance what properties that object must possess? This is the fundamental difficulty. Kant’s solution is not to deny objects but to distinguish between objects as they may be independently of our mode of cognition and objects as they are necessarily given to us within experience. Once that distinction is made, a priori knowledge becomes intelligible. We can know a priori those features that belong to objects insofar as those objects must conform to the conditions under which we can experience them. This is why Kant’s argument ultimately depends upon the distinction between appearance and thing in itself.

The Copernicus analogy is therefore extremely carefully chosen. Before Copernicus, the apparent motion of the heavens was explained by assuming that the celestial bodies moved around the observer. Copernicus reversed the standpoint and asked what would happen if we treated the observer as moving while the stars remained comparatively fixed. The point was not simply to deny what appears to the senses. The apparent motion remained. What changed was the explanatory standpoint. Kant wants to perform an analogous reversal in epistemology. Instead of asking, “How must our cognition change itself in order to conform to objects?” he asks, “What must objects be like, insofar as they are objects for cognition, in order for our cognition to have the necessary structures that it does?” This is the philosophical equivalent of moving the observer.

But there is an even more precise connection to Copernicus. Copernicus did not simply switch one dogma for another. He changed the reference frame in which the phenomenon was understood. Kant’s revolution likewise changes the reference frame of epistemology. The traditional picture treats the subject as a passive spectator confronting an independently constituted object. Kant’s picture treats the subject as possessing a priori forms and concepts that condition how objects can appear and be known. The “object” of experience is therefore relational: it is something given to a subject under determinate conditions of cognition. Kant is not saying that there is no reality outside the subject. He is saying that what we can legitimately call an object of experience is inseparable from the conditions under which it can be experienced.

Now Kant narrows the issue to “intuition.” This is another word we need to become very comfortable with. In Kant, an intuition (Anschauung) is not merely a vague hunch or an intuitive feeling. It is a singular, immediate mode of representation through which an object is given. Concepts, by contrast, are general representations through which objects are thought. The basic division is therefore: sensibility gives us objects through intuitions; understanding thinks objects through concepts. Kant will eventually summarize the relationship in the famous formulation: “Without sensibility no object would be given to us, and without understanding no object would be thought.” Here, however, he is focusing first on intuition. If intuition had to conform entirely to objects as they exist independently of us, Kant says, he cannot see how we could know anything about the structure of objects a priori. But if objects must conform to the nature of our faculty of intuition, then a priori knowledge becomes possible.

This is where space and time are waiting just around the corner. Kant has not yet argued for them here, but the logic is already visible. Suppose space is not something we learn by comparing external objects, but rather the a priori form through which external objects can appear to us at all. Then geometry can have necessary a priori validity for objects of possible experience because any such object must appear spatially. Likewise, if time is the a priori form of inner sense, then temporal relations can be known a priori because anything we experience must occur within time. This is the foundation of Kant’s answer to the question of how synthetic a priori cognition is possible.

But Kant knows that intuition alone is insufficient. This is why the paragraph suddenly makes another turn: “Now as I cannot rest in the mere intuitions, but—if they are to become cognitions—must refer them, as representations, to something, as object…” This sentence is enormously important. Kant is saying: even if we solve the problem of intuition, we haven’t yet solved cognition. A sensory representation by itself does not amount to knowledge. It has to be referred to an object. We need to be able to say, in effect, “this appearance is of something.” But how do we determine that object? Through concepts. And now the Copernican experiment has to be repeated at a second level.

The first question was: what if intuition conforms to the subject rather than the object? The second is: what if concepts conform to the subject rather than the object? Kant says there are again two possibilities. Either the concepts through which we determine objects conform to the objects independently of our cognition, or the objects of experience conform to our concepts. If we choose the first option, we are thrown back into the same problem. How could we know a priori that our concepts correspond to objects before those objects are given? But if we choose the second option, the problem becomes intelligible. Objects of experience must conform to the concepts that constitute the conditions of possible experience.

This is the crucial point where Kant’s “categories” are beginning to emerge. The understanding possesses a priori concepts—such as unity, plurality, substance, causality, possibility, necessity, and so forth. These are not empirical generalizations. We do not first observe thousands of causal sequences and then abstract the concept of causality. Rather, causality is one of the concepts through which events can be constituted as objectively related in experience. This does not mean that we can arbitrarily impose any concept whatsoever upon the world. Only certain fundamental concepts have objective validity, because only certain concepts are constitutive of the possibility of experience itself. The Transcendental Analytic will be Kant’s attempt to identify these concepts and justify their objective validity.

The sentence “experience itself is a mode of cognition which requires understanding” is therefore a bombshell. Experience is not simply raw sensory input. Experience, in Kant’s technical sense, is already cognitively organized. The German is: “Denn Erfahrung selbst ist eine Erkenntnisart, die Verstand erfordert.” “For experience itself is a kind of cognition that requires understanding.” This is one of the fundamental things we have to keep in mind as we read Kant. Experience is not the same as sensation. Sensation is material given to sensibility. Experience is organized cognition of objects. The transition from sensation to experience requires synthesis and conceptual determination. That is why Kant can say that there are a priori laws of understanding that must precede objects as objects of experience.

The next sentence is perhaps the most important one for understanding transcendental philosophy: “Before objects are given to me, that is, a priori, I must presuppose in myself laws of the understanding which are expressed in conceptions a priori.” Notice the direction. Kant is not saying that the laws of understanding are learned from objects. He is saying that the possibility of experiencing objects already presupposes them. This is the transcendental move. Instead of asking for the empirical origin of a concept, Kant asks what function the concept performs in making objective experience possible. The question becomes functional rather than merely historical. Causality, for example, is not explained by saying, “We have seen one event follow another many times.” Kant’s question is: what must the understanding contribute for a sequence of appearances to be experienced as an objective event governed by a causal law?

This is also where Kant’s distinction between “appearance” and “thing in itself” becomes unavoidable. The objects that must conform to our concepts are objects of experience. Kant is not saying that reality in its absolute independence from cognition must conform to the human mind. He is saying that anything that can be an object for us in experience necessarily conforms to the forms and concepts through which experience is possible. The famous Kantian distinction is therefore not simply between “real” and “unreal.” Appearances are empirically real. They are the actual objects of experience. The thing in itself is not another mysterious object hiding behind every appearance like a second world; it is the object considered apart from the conditions under which it can be experienced by us. Kant’s restriction is epistemological: we know objects as they appear under the conditions of our cognition, not objects as they might be independently of those conditions.

And then comes the extraordinary final move: “Now there are objects which reason thinks, and that necessarily, but which cannot be given in experience, or, at least, cannot be given as reason thinks them.” This sentence opens the door to the entire Transcendental Dialectic. Kant is acknowledging that reason necessarily thinks things that cannot be objects of possible experience. God is one example. The soul as a simple, immortal substance is another. The world considered as a complete totality is another. Freedom, in its noumenal sense, creates another particularly interesting case. Reason cannot simply stop thinking these things. They arise naturally from reason’s demand for completeness and the unconditioned. But the question is whether the categories that legitimately structure experience can be extended to these objects beyond experience.

This gives us a magnificent experimental test. Kant says that these objects will provide “an excellent test” of his new method. Why? Because if his method is right, we should discover that pure reason can legitimately use its concepts within experience but gets itself into contradictions when it treats those concepts as determining objects beyond possible experience. And that is exactly what Kant will demonstrate in the Antinomies. Reason can prove that the world has a beginning in time and that the world has no beginning in time, provided it treats the world as a thing in itself and applies the category of totality beyond experience. The contradiction is not evidence that logic itself fails. It is evidence that reason has misapplied a legitimate cognitive structure outside the domain in which it has objective validity.

The footnote makes this even clearer. Kant says that the new method has been “borrowed from the natural philosopher.” But because metaphysical objects cannot themselves be experimentally manipulated, we have to perform the experiment differently. We look at “one and the same conception” from two sides: first in relation to experience, where the object is given through the senses and understanding; second in relation to reason, where the object is isolated and considered as something transcending possible experience. This double perspective is absolutely essential. The same concept can have legitimate use in relation to possible experience and illegitimate use when detached from that domain. The problem is not necessarily the concept itself. The problem is the range of its application.

This is where Kant’s critical project becomes much subtler than simple empiricism. An empiricist might say: if you cannot experience something, stop talking about it. Kant does not say that. Reason can and must think beyond experience. What Kant denies is that merely thinking something gives us cognition of it. This distinction between thinking and knowing will be crucial. You can think the concept of an absolutely necessary being. You can think an immortal soul. You can think the world as an absolutely complete totality. But you cannot therefore claim theoretical knowledge of those objects. The concepts may have a legitimate regulative function for reason while lacking constitutive objective validity beyond experience.

And now we can see why Kant has been so insistent throughout the preface on the difference between logic, mathematics, physics, and metaphysics. Logic concerns the formal laws of thought. Mathematics constructs concepts in pure intuition. Physics combines a priori principles with empirical intuition. Metaphysics attempts to use pure concepts to reach objects that may lie beyond possible experience. Kant’s entire critical project is going to determine exactly where this last move succeeds and where it fails.

The deepest thing happening in this passage, though, is that Kant has shifted the question from ontology to conditions of intelligibility. Traditional metaphysics asks, “What is the ultimate structure of reality?” Kant asks first, “What must be true of cognition for objects to be objects for us at all?” That is the transcendental question. And the answer will be surprisingly strong: there are forms of sensibility and concepts of understanding that are not derived from experience but are conditions for experience. Consequently, some features we ordinarily think belong simply to “the world”—spatiality, temporality, causal ordering, substantiality, unity, plurality, necessity, and so forth—must be understood partly through the contribution of our cognitive faculties.

This is also where your earlier interest in the relation between Kant and Hegel, Husserl, Heidegger, and Derrida will eventually become extremely fruitful. We should postpone the full comparisons as you requested, because Kant needs to be understood on his own terms first. But notice that Kant has just made a move that later continental philosophy will repeatedly return to: the object is not philosophically intelligible independently of the conditions through which it becomes an object. Husserl will radicalize the analysis of object-givenness; Heidegger will transform the question of conditions of intelligibility into a question about Being and disclosedness; Hegel will challenge Kant’s distinction between the conditions of experience and the thing in itself; Derrida will interrogate the stability of the distinctions Kant uses to establish these boundaries. But we should earn those comparisons by first understanding what Kant himself is doing.

For now, I would put a large conceptual marker here: Kant has just proposed the reversal that governs the entire Critique. The old question was, “How can our cognition conform to objects?” Kant’s experimental question is, “What must objects be like, insofar as they are objects of experience, to conform to the necessary conditions of our cognition?” From this follows the entire architecture of the book: first sensibility and its a priori forms, then understanding and its a priori concepts, then reason and its drive toward the unconditioned, and finally the critique of reason when it attempts to extend those concepts beyond possible experience. The Critique is therefore not merely a catalog of what we can and cannot know. It is an investigation into the reciprocal relation between givenness and structure: what is received, what is contributed, and what happens when the structures that make experience possible are mistaken for knowledge of reality beyond experience.

And perhaps the single sentence to carry into the next section is Kant’s final principle: “We only cognize in things a priori that which we ourselves place in them.” German: “daß wir nämlich von den Dingen nur das a priori erkennen, was wir selbst in sie legen.” Everything now depends on what “we ourselves place in them” actually means. Kant’s answer will not be “whatever we feel like putting there.” It will be the necessary forms of sensibility and the necessary functions of understanding. The extraordinary wager of the Critique is that by discovering what the mind necessarily contributes to experience, we can simultaneously explain the possibility of a priori scientific knowledge and establish the limits beyond which reason can no longer claim theoretical knowledge.

This is where Kant tells us what the Copernican experiment actually bought him. The previous passage gave us the hypothesis: perhaps objects of experience conform to the structures of our cognition. Now Kant says, in effect, “the experiment worked.” But the result is deliberately paradoxical. It gives metaphysics a scientific foundation precisely by limiting what metaphysics can know. Kant’s great discovery is not that pure reason can finally reach farther than experience. It is almost the opposite: reason becomes more secure by discovering exactly where its legitimate reach ends. The German opening is emphatic: “Dieser Versuch hat den gewünschten Erfolg und verspricht der Metaphysik in ihrem ersten Teile, da sie sich mit Begriffen a priori beschäftigt, von denen die korrespondierenden Gegenstände in der Erfahrung gegeben werden können, den sicheren Gang einer Wissenschaft.” The “desired success” is therefore qualified from the beginning. Metaphysics can become scientific in its “first part,” where its a priori concepts relate to objects that can actually be given in experience. This qualification is going to govern everything that follows.

The distinction between the “first part” and “second part” is crucial. Kant is not simply declaring that metaphysics is impossible. He is dividing metaphysics according to the relation between its concepts and possible experience. In the first part, we have a priori concepts whose corresponding objects can be encountered in experience. This is where transcendental philosophy can establish genuine knowledge. Kant can investigate the a priori forms of sensibility and the a priori concepts of understanding because they structure objects that can actually be experienced. But metaphysics traditionally has a second ambition: it wants to know objects that cannot be given in experience—God, the immortal soul, the world as an absolute totality, and ultimately the unconditioned. Here the situation changes completely. Kant’s critical method will establish the legitimacy of a priori cognition within experience while simultaneously demonstrating the illegitimacy of using that cognition to generate theoretical knowledge beyond experience.

That is why the next sentence is so important: “For by this new method we are enabled perfectly to explain the possibility of a priori cognition.” This is the central epistemological achievement. Kant is not satisfied merely to point out that mathematics and physics contain a priori knowledge. He wants to explain how such knowledge is possible. This is the meaning of “transcendental” in Kant. A transcendental investigation asks about the conditions of possibility of cognition. Instead of merely saying, “We know that mathematics is necessary,” Kant asks, “What must be true of human cognition for necessary mathematical knowledge to be possible?” Instead of merely saying, “Nature obeys causal laws,” he asks, “What must be true of the structure of experience for causal necessity to be possible?” The Critique is therefore an investigation into possibility conditions.

And Kant says something even stronger: the method allows him “to demonstrate satisfactorily the laws which lie a priori at the foundation of nature, as the sum of the objects of experience.” This phrase “nature, as the sum of the objects of experience” needs to be handled carefully. Kant is not saying that he has deduced every particular empirical law of physics from pure reason. He means that the fundamental framework within which something can count as nature for us is structured a priori. Nature, insofar as it is an ordered field of possible experience, must conform to certain principles of understanding. This is where causality, substance, unity, plurality, necessity, and the other categories will enter. Kant can therefore explain how nature can be both empirically given and necessarily law-governed. The necessity does not come from repeatedly looking at nature; it comes from the conditions under which nature can be experienced as an objective order at all.

But then Kant announces “a surprising result”: the very investigation that secures metaphysical knowledge in its legitimate domain seems to destroy metaphysics in its traditional domain. We discover that “our faculty of cognition is unable to transcend the limits of possible experience.” And Kant immediately recognizes the apparent disaster: “yet this is precisely the most essential object of this science.” Traditional metaphysics wants to know what lies beyond experience. Kant’s critical philosophy seems to tell it that this is exactly what theoretical cognition cannot do. The success of the Critique therefore looks like the failure of metaphysics.

This tension is at the heart of Kant. He does not want to abolish metaphysics; he wants to transform it. The old metaphysics thought its task was to obtain theoretical knowledge of supersensible objects. Kant’s metaphysics will instead include a critique of the conditions and limits of such knowledge. There is a genuine science of the a priori, but it is not the science that dogmatic metaphysicians thought they possessed. Kant is effectively saying: “You were right that reason necessarily asks metaphysical questions. You were wrong about what follows from its asking them.” The questions are unavoidable; the answers cannot always take the form of theoretical cognition.

Now we reach the famous distinction: “our rational cognition a priori … has only to do with phenomena, and that things in themselves, while possessing a real existence, lie beyond its sphere.” This is one of the sentences that needs to be read with extreme precision. “Phenomena” are not illusions. Kant is not dividing reality into a real world and a fake world. Phenomena are objects as they are given under the conditions of human sensibility and understanding. They are the actual objects of experience. The thing in itself, by contrast, means the object considered independently of the conditions under which it can be experienced by us. Kant explicitly says that things in themselves “possess a real existence.” So his limitation of knowledge is not a denial of reality. It is a limitation on what our mode of cognition can determine about that reality.

This is where Kant differs from what we might loosely call subjective idealism. Berkeley, for example, famously connected being with being perceived. Kant is not doing that. He maintains empirical realism while introducing transcendental ideality. Objects of experience are genuinely real within experience; they are not inventions of the individual mind. But their mode of appearing to us is necessarily conditioned by the forms of our sensibility and understanding. Thus Kant can simultaneously say that appearances are real and that things in themselves are not cognizable by us as they are in themselves.

The next sentence explains why reason nevertheless cannot leave the noumenal realm alone: “that which of necessity impels us to transcend the limits of experience and of all phenomena is the unconditioned.” This is one of the deepest ideas in the entire Critique. Reason is not merely tempted by metaphysics because human beings are curious. There is a structural demand within reason itself. Whenever reason encounters something conditioned, it asks for the condition. If A depends upon B, reason asks what explains B. If B depends upon C, it asks what explains C. Reason continues the series and eventually demands something that is not itself conditioned by something else: the unconditioned.

The German is “das Unbedingte,” literally “the unconditioned,” from Bedingung, condition. There is an interesting linguistic resonance here: a “Bedingung” is a condition, something upon which something else depends. Reason seeks what is not itself dependent upon another condition. It wants completion. It wants the series of conditions to terminate—not merely arbitrarily, but in something that can serve as the ultimate ground. This is why Kant’s metaphysics cannot simply be dismissed. The demand for the unconditioned is generated by reason’s own operation.

Here Aristotle and Leibniz become particularly relevant, although we should keep the historical comparison subordinate to Kant’s text. Aristotle’s search for first principles and ultimate causes lies deep in the background. His causal regress raises the question of whether explanation can proceed indefinitely or whether there must be a first principle. Leibniz intensifies the issue through the principle of sufficient reason: if everything contingent requires a sufficient reason, reason seeks an ultimate ground, culminating in the necessary being. Kant inherits this rational demand but radically changes its epistemological status. He will argue that reason’s demand for the unconditioned is legitimate as a demand of reason, but it does not follow that the unconditioned can therefore become an object of theoretical knowledge.

This distinction between “reason’s demand” and “knowledge’s achievement” is absolutely central. Kant does not say, “There is no unconditioned.” He says that reason cannot find the unconditioned within the domain of objects as they are given in experience. The unconditioned is a demand generated by reason’s attempt to complete the series of conditions. The trouble begins when reason treats this demand as though it had discovered an object corresponding to it within experience—or when it tries to determine a supersensible object by applying categories whose legitimate use belongs to experience.

Now Kant gives us the actual test of the Copernican hypothesis. He says, in effect: let us take the two competing models and see what happens to the unconditioned. Model one: our cognition conforms to objects as things in themselves. Model two: objects as phenomena conform to our mode of representation. If the first produces contradiction and the second dissolves it, then the experiment has succeeded. This is extraordinary because Kant is not merely asserting transcendental idealism. He is presenting the antinomies as an internal test of the position. The contradictions of metaphysics will function like the experimental results of a scientific investigation.

This is what the footnote’s chemistry analogy makes explicit. Kant compares the procedure to the chemists’ experiment of reduction, or “synthetic process.” He is thinking of the analysis of a compound into its heterogeneous elements and then the recombination of those elements. The metaphysician similarly “separates” pure a priori cognition into two heterogeneous components: cognition of things as phenomena and cognition of things in themselves. Then dialectic recombines them in relation to reason’s necessary idea of the unconditioned. The important discovery is that harmony appears only when the distinction between phenomena and things in themselves is maintained.

The footnote is therefore almost a miniature description of the entire Critique. First, analysis: separate what had previously been confused. Then dialectic: observe what happens when reason tries to reunite the elements according to its demand for completeness. Finally, diagnosis: contradiction arises if we collapse the distinction, while contradiction disappears if we preserve it. Kant is effectively saying that the antinomies are not embarrassing failures of his philosophy. They are evidence for it.

This gives us a much richer understanding of the word “critique.” Critique does not mean criticism in the ordinary sense, as though Kant’s goal were to attack metaphysics. It means something closer to adjudication. Remember the judge metaphor from the earlier passage. Reason establishes a tribunal for itself. The Critique asks what claims reason is entitled to make, what evidence supports those claims, and where reason exceeds its jurisdiction. The metaphor of jurisdiction will become increasingly useful as we read. Kant is not trying to destroy reason; he is trying to determine its legitimate sphere of authority.

And this is why the contradiction over the unconditioned matters so much. Suppose we treat the world as a thing in itself and ask whether it has a beginning in time. Reason can produce an argument for a beginning and an argument against a beginning. Likewise with spatial limits, simple substances, freedom, and necessary being. If the world is treated as an object existing in itself that can be completely determined by pure concepts, reason becomes self-contradictory. But if the world of experience is understood as phenomenal—structured by the conditions of sensibility and understanding—then the contradiction disappears. The mistake was not necessarily in the individual arguments. The mistake was in the assumption that the totality of conditions could itself be given as an object within experience.

This is a subtle point that will matter enormously later: the “world” as an experienced totality is not itself an object of experience in the same way that a table or planet is. We can experience things within the world, but we cannot step outside the totality of possible experience and experience “the world as a whole” in the same manner. Reason nevertheless demands that totality. That tension is what produces the antinomy. Kant’s solution is not to deny the demand for totality but to recognize that the demand belongs to reason and cannot simply be converted into an empirical object.

There is therefore a strange double movement in Kant’s argument. First, he restricts cognition: “you cannot know things in themselves.” Then, almost immediately, this restriction becomes the condition for resolving metaphysical contradiction. The limitation is productive. By refusing to pretend that reason can know what it cannot know, Kant gives reason a stable domain in which its a priori claims become genuinely secure. This is why the title “Critique of Pure Reason” is so important. The Critique is not principally an attack on reason. It is reason examining its own powers.

And we should notice the remarkable phrase near the end: “we shall then be convinced of the truth of that which we began by assuming for the sake of experiment.” Kant’s method is almost self-validating in the scientific sense. He began with a hypothesis: perhaps objects of experience conform to our cognition. He then derives the consequences. The consequence is that the contradictions generated by reason’s attempt to know the unconditioned disappear once we distinguish phenomena from things in themselves. The hypothesis has therefore gained confirmation through its explanatory power. Kant’s philosophy is not empirical science, but he deliberately models its method on the revolution in natural science.

There is one point in the English translation that deserves particular attention. The final sentence says: “the unconditioned does not lie in things as we know them, or as they are given to us, but in things as they are in themselves, beyond the range of our cognition.” Read quickly, this can sound as though Kant is saying, “The unconditioned exists somewhere in the thing-in-itself world, and we simply cannot reach it.” That is too strong. The more careful point is that the unconditioned cannot be located among phenomena as objects of possible experience. Kant’s language allows him to preserve the idea of the unconditioned as something demanded by reason, but the Critique will make it increasingly difficult to turn that demand into a determinate theoretical object. This distinction becomes especially important when we reach God, freedom, and immortality.

This passage therefore gives us a map of what we are about to read. The Transcendental Aesthetic will ask: what are the a priori conditions of sensibility? The answer will be space and time. The Transcendental Analytic will ask: what are the a priori conditions supplied by understanding? The answer will involve the categories and their deduction. Together these establish the legitimate domain of a priori cognition: possible experience. Then the Transcendental Dialectic will ask what happens when reason pushes beyond that domain in pursuit of the unconditioned. The result will be the paralogisms, the antinomies, and the critique of the traditional proofs of God. And finally, Kant will have to explain what legitimate role these ideas can nevertheless play once they have been denied constitutive theoretical knowledge.

So I would hold onto three distinctions from this paragraph as we continue. First, “a priori” does not mean “mysteriously innate information about the world”; it means cognition that is independent of particular experience and has the kind of necessity Kant is investigating. Second, “phenomenon” does not mean “illusion”; it means the object as it can be given and known under the conditions of possible experience. Third, “thing in itself” does not simply mean “another object hiding behind appearances”; it means the object considered apart from our mode of cognition. If those three distinctions remain clear, an enormous amount of Kant becomes easier.

And there is one final irony worth noticing. Kant began the preface by describing metaphysics as a queen who had fallen into civil war, dogmatism, skepticism, and finally indifference. Now he has discovered the reason for the civil war: reason has been trying to exercise jurisdiction over objects beyond the conditions of its own cognition. The cure is not to abolish the kingdom but to establish its constitution. The Critique is therefore, in a very literal sense, Kant’s constitutional settlement for reason: here is what reason can legislate, here is where its legislation is valid, and here is where it must recognize a boundary. The question that will occupy us next is therefore wonderfully concrete: what exactly are the structures that reason contributes to experience? Once we enter the Transcendental Aesthetic, Kant will stop speaking at this broad programmatic level and begin showing us, step by step, what he thinks those structures actually are.

This is the point where the entire Preface suddenly opens toward the Critique of Practical Reason. The First Critique has just established a boundary: speculative reason cannot theoretically know the supersensible. But Kant immediately insists that a boundary is not the same thing as an annihilation. If speculative reason cannot cross the boundary, perhaps practical reason can approach what lies beyond it in a different way. This is the seed of the later distinction between theoretical and practical reason. And we should be very careful here, because Kant is not yet saying that morality gives us theoretical knowledge of God, freedom, or immortality. He is asking whether practical cognition supplies a different kind of rational access to what speculative cognition cannot establish. That distinction will eventually become the entire architecture of the second Critique.

The opening sentence is deliberately cautious: “after we have thus denied the power of speculative reason to make any progress in the sphere of the supersensible, it still remains for our consideration whether data do not exist in practical cognition…” The word “data” matters. Kant is not saying that practical reason gets to make up whatever metaphysical entities it wants. He is asking whether practical cognition contains something given to reason that speculative cognition does not possess. In the second Critique, that “fact” will become the moral law, the famous “fact of reason” (Faktum der Vernunft). The moral law is not inferred from experience. Nor is it a theoretical deduction about a supersensible object. Rather, we encounter ourselves as obligated by a law of practical reason. That fact gives Kant a route toward freedom that speculative reason alone could never provide.

Notice, too, the beautiful phrase “to rise beyond the limits of all possible experience from a practical point of view.” Kant is not retracting the limitation he has just established. The phrase “from a practical point of view” is doing all the work. Theoretical cognition asks, “What is the case?” Practical reason asks, roughly, “What ought I to do?” Those are different forms of rational relation to reality. Kant will insist that the inability to answer the first question about a supersensible object does not prevent the second question from generating rational commitments concerning it. This is why the First Critique cannot simply be read as a book that tells us, “You cannot know God, therefore God is irrelevant.” Kant’s actual position is much more interesting: speculative reason cannot establish God as an object of theoretical knowledge, but practical reason may have grounds for postulating God in relation to the moral vocation of rational beings.

The word “space” is particularly revealing: “Speculative reason has thus, at least, made room for such an extension of our knowledge.” In German Kant says that pure reason has made “Platz” for it. The metaphor is architectural. The Critique does not fill the supersensible realm with knowledge; it clears a space in which another legitimate use of reason might operate. This is the famous meaning behind Kant’s later remark, from the Preface to the second edition, that he had to “deny knowledge in order to make room for faith” (“den Glauben Platz zu bekommen”). But we should not import the whole later formulation prematurely. Here Kant is making a methodological point: speculative reason’s limitation does not itself constitute a prohibition against every possible rational relation to the supersensible.

And then he says something even more provocative: speculative reason “does not rob us of the liberty to fill it up, if we can, by means of practical data—nay, it even challenges us to make the attempt.” Reason itself challenges us. This is very important for understanding Kant’s attitude toward metaphysics. He does not regard metaphysical questions as pathological curiosities that philosophy should simply suppress. The demand for the supersensible remains. The Critique transforms the question from “Can speculative reason prove these objects?” into “What legitimate role can these ideas have in practical reason?” That transformation is one of the great conceptual transitions from the First to the Second Critique.

The footnote about Copernicus and Newton reinforces this methodological point. Kant says that Copernicus initially introduced his model as a hypothesis, but the subsequent laws of celestial motion confirmed the new standpoint and even led to the discovery of the invisible force of Newtonian attraction. The analogy is carefully constructed. Copernicus changed the standpoint first; the explanatory consequences came afterward. Kant is saying that his own Copernican procedure is initially introduced as an experimental hypothesis in the Preface. The Critique itself, however, will not leave it at the level of hypothesis. He promises that the argument will become “apodeictic,” necessary and demonstrative, through the analysis of space, time, and the elementary concepts of understanding.

That word “apodeictic” deserves our attention because Kant has been using it throughout the Preface. It comes from the Greek apodeiktikos, meaning demonstrative or necessary. An apodeictic judgment is not merely probable or even empirically well-confirmed; it carries necessity. Kant’s entire ambition is to determine whether such necessity is possible in philosophy. Mathematics has it. Certain fundamental principles of natural science have it. The question is whether metaphysics can possess it without pretending to know what lies beyond experience. Kant thinks it can—but only after it changes its method.

The next paragraph then gives us an exceptionally concise statement of what the Critique is and is not. “This attempt to introduce a complete revolution in the procedure of metaphysics … constitutes the aim of the Critique of Pure Speculative Reason.” The key word is “procedure.” Kant is not yet offering us a metaphysical system. He explicitly says: “It is a treatise on the method to be followed, not a system of the science itself.” This is why we should resist the temptation to read the Critique as though Kant’s primary goal were simply to tell us what reality ultimately consists of. His immediate task is methodological and juridical: determine the procedure by which pure reason can legitimately investigate its own claims.

This explains the title “Critique.” The Critique is not the completed metaphysics. It is the investigation that makes a legitimate metaphysics possible. Think of the analogy Kant himself gives elsewhere: before constructing a building, one needs to examine the ground and determine what can bear weight. The Critique is that preliminary investigation. It asks what the foundations of pure cognition are, how far they extend, and where they collapse. Only after this can one construct a “Metaphysics of Nature” or “Metaphysics of Morals.” This is why Kant’s unpublished or separately planned Metaphysic of Nature that he mentioned earlier was supposed to follow the Critique.

But Kant immediately makes an extraordinary claim about what the method can accomplish. The Critique “marks out and defines both the external boundaries and the internal structure of this science.” So Kant is not merely going to place a fence around metaphysics. He thinks he can map its internal architecture. The distinction between sensibility, understanding, and reason is not arbitrary. The entire system of pure cognition can be systematically enumerated because pure reason is itself a systematic unity.

Here we encounter one of Kant’s most important assumptions: reason is not an accidental pile of capacities. It is a structured whole. He says that “reason is, in regard to the principles of cognition, a perfectly distinct, independent unity, in which, as in an organized body, every member exists for the sake of the others, and all for the sake of each.” The German image is organic: “wie ein organisiertes Wesen,” like an organized being. This is an extraordinary image because it means that Kant’s philosophy is going to be architectural and anatomical at the same time. He wants to discover the elements of reason, but he also wants to understand their systematic relations. You cannot safely understand one principle in isolation because its function depends upon its place within the total structure.

This is one reason Kant cares so much about completeness. Earlier he said that pure reason is a “perfect unity.” Here he explains why. If reason itself is systematic, then discovering one fundamental principle should constrain what the others can be. You cannot arbitrarily add a new fundamental principle whenever convenient. If a principle genuinely belongs to pure reason, it must have a determinate place within the whole. This is also why Kant’s famous “table” of categories is not simply a random list. He believes the fundamental concepts of understanding can be derived systematically from the forms of judgment. Whether he actually succeeds in this derivation is another question—and Hegel, among others, will later challenge him—but the ambition is extraordinarily systematic.

The sentence “nothing must be attributed to the objects but what the thinking subject derives from itself” now makes sense in relation to everything we’ve just read. This does not mean that the subject invents objects. It means that when we seek a priori cognition, we cannot smuggle empirical content into our concepts. We can only identify those structures that cognition itself necessarily contributes. This is precisely the project of the Transcendental Aesthetic and Analytic. What belongs necessarily to our mode of sensibility? What belongs necessarily to our mode of understanding? Those are the legitimate sources of a priori cognition.

And then Kant says something almost audacious about metaphysics: once it enters the “sure path of science,” it can theoretically complete its work. Why? Because metaphysics is concerned with principles and the limits of their employment, and pure reason is finite in its inventory. There is no endless accumulation of empirical facts. The relevant structures are finite and discoverable. Kant therefore imagines metaphysics as a science capable of completion. This is very different from physics, where new empirical discoveries can continually expand the body of knowledge. Metaphysics, once its principles and boundaries are established, would have no new fundamental principles waiting to be discovered.

That idea is encapsulated in the Latin quotation: “Nil actum reputans, si quid superesset agendum.” It means roughly, “Count nothing as accomplished if anything remains to be done.” Kant is saying that metaphysics, because of its systematic character, is obligated to pursue completeness. If one fundamental question remains unresolved, the system is not complete. This is why he repeatedly emphasizes the totality and unity of pure reason. The Critique is not merely a collection of answers to isolated philosophical puzzles. Kant wants a map of the whole field.

And here we can begin to see something that will become extremely important when we eventually compare Kant with Hegel. Kant thinks systematic completeness is possible because the structures of pure reason can be completely enumerated and their legitimate jurisdiction determined. Hegel will inherit this extraordinary ambition but argue that Kant has not gone far enough: the supposedly fixed structures of Kantian understanding themselves generate movement and contradiction. Hegel will therefore try to turn the Kantian system into a dynamic self-development of concepts. But for now, we should resist letting Hegel rewrite Kant for us. Kant is still constructing the tribunal.

The same caution applies to Heidegger, Husserl, and Derrida. Each will inherit something from this moment, but each will reinterpret it radically. Husserl will be fascinated by the conditions under which objects are given to consciousness. Heidegger will ask whether Kant’s transcendental question has to be grounded in a more fundamental question of Being and finitude. Derrida will repeatedly examine the boundaries Kant draws between reason, experience, the conditioned, and the unconditioned. But all of those later moves depend upon understanding what Kant means when he says that reason can determine its own limits.

There is also a subtle connection back to Aristotle. Kant’s language of an “organized body” and a complete science has an Aristotelian flavor, but the project has been radically transformed. Aristotle asks about the principles of beings; Kant asks about the principles of cognition of beings. Aristotle’s metaphysics reaches toward first causes and being qua being; Kant asks whether our cognitive apparatus can legitimately make such determinations. Likewise, Leibniz’s dream of a complete rational science remains in the background. Kant shares the rationalist aspiration to systematicity but refuses to let pure concepts automatically become knowledge of supersensible objects. He wants Leibniz’s systematic rigor without Leibniz’s unrestrained metaphysical reach.

And now we can understand why Kant calls metaphysics “the fundamental science.” It is not fundamental because it gives us the deepest inventory of objects. It is fundamental because it investigates the principles and limits of reason itself. It asks what reason can legitimately claim when it operates independently of experience. In that sense, Kant has turned metaphysics reflexive. Metaphysics must first investigate its own conditions of possibility.

This gives us a beautiful progression through the Preface. At the beginning, reason is tormented by questions it cannot avoid. Then we see mathematics and physics successfully establish scientific methods. Kant asks whether metaphysics can undergo a similar revolution. He reverses the relation between cognition and objects. He discovers that objects of experience conform to a priori structures of cognition. This explains synthetic a priori knowledge. But the same discovery reveals that those structures cannot legitimately be extended beyond possible experience. Reason encounters the unconditioned and generates contradictions. The contradictions disappear when we distinguish phenomena from things in themselves. Speculative reason therefore discovers its limits. And precisely by establishing those limits, it leaves open a domain in which practical reason may operate.

That last transition is perhaps the most important thing to carry forward. Kant is not ending metaphysics; he is redistributing its possibilities. Theoretical reason is confined to phenomena. But practical reason will reopen the question of the supersensible under a different form. This is why your original instinct to begin with the First Critique was exactly right. When we eventually return to the Preface of the Second Critique, the opening claim that “freedom is the keystone of the whole system of pure reason” will no longer appear as an isolated proclamation. We will have seen the entire machinery that makes such a statement possible.

For now, I would put the whole Preface into one enormous conceptual sentence: Kant believes that reason’s failure to know the supersensible is not a failure of reason but the consequence of discovering the conditions under which reason genuinely knows anything at all; once those conditions are made explicit, metaphysics can become scientific within its proper domain, the contradictions of dogmatic metaphysics can be diagnosed, and the very boundary that restricts speculative reason can potentially become the opening through which practical reason encounters freedom, morality, and the questions of God and immortality.

And that is why the end of the Preface is not really an ending. Kant has finished telling us why the Critique is necessary. He has not yet shown us the machinery. The next task is to enter the work itself and watch him perform the promised investigation: first, what does sensibility contribute? Then, what does understanding contribute? Then, what does reason demand when it seeks the unconditioned? Those three movements will give us the actual anatomy of Kant’s revolution.

This is one of the most important passages in the entire Preface, because Kant now tells us why the limitation of reason is not merely a philosophical defeat. The Critique appears, at first, to be a negative project: it tells speculative reason, “You cannot go there.” But Kant’s great claim is that this negative restriction has a positive purpose. By preventing speculative reason from occupying the entire territory of thought, criticism preserves a space for practical reason—especially morality and freedom. In other words, Kant is not merely putting a fence around knowledge. He is protecting something that would otherwise be destroyed by an overly ambitious theory of nature. This is the point at which the First Critique begins to reveal its ethical purpose.

The opening question is almost theatrical: “But, it will be asked, what kind of a treasure is this that we propose to bequeath to posterity?” Kant anticipates the obvious objection. Suppose I spend hundreds of pages showing that human reason cannot know things in themselves. What have I actually given humanity? Why should anyone regard ignorance as a “treasure”? His answer is that ignorance, when properly determined, can itself be liberating. The critical philosopher does not simply say, “We don’t know.” He says exactly what we cannot know, why we cannot know it, and what remains possible once that boundary has been established. That is a radically different condition from metaphysical confusion. Kant’s ignorance is disciplined ignorance.

This is why he says the “primary use” of the Critique is negative. It prevents speculative reason from venturing beyond experience. But immediately he says that this negative function has a “positive value.” The key is that speculative reason, when it exceeds its limits, does not merely produce false beliefs; it can actually “contract” the use of reason. That is a startling claim. You might think that a metaphysics which tries to explain everything gives reason more freedom. Kant says precisely the opposite. A theory that insists everything must be explained according to the principles valid for objects of experience can become so expansive that it destroys other legitimate forms of rational activity.

This is where the phrase “threaten to extend the limits of sensibility … over the entire realm of thought” becomes crucial. Sensibility has a legitimate sphere: objects as they can be given to us in space and time. But if we take the principles governing such objects and declare that they govern absolutely everything that can be thought, we have silently transformed a condition of experience into a universal ontology. That is the dogmatic mistake Kant is fighting. The laws of nature are valid for appearances, but speculative reason wants to say: therefore everything that exists, without qualification, must be subject to those same laws.

And that would be disastrous for Kant because practical reason concerns precisely what cannot simply be reduced to empirical nature. Morality involves obligation. Obligation involves the possibility that I could have acted otherwise. And that possibility points toward freedom. If everything about the human being is exhaustively determined by natural causality in exactly the same sense in which physical events are determined, then the deepest notion of moral responsibility becomes difficult, perhaps impossible, to preserve. Kant’s critical distinction therefore becomes the conceptual space in which freedom can even be thought.

This is why the police analogy is more philosophically important than it initially appears. Kant says that criticizing speculative reason is like establishing a police force. The police do not produce your profession, your art, your commerce, or your happiness directly. Their positive contribution consists in preventing one citizen from violently interfering with another. Likewise, the Critique does not itself create morality. It prevents speculative reason from illegitimately interfering with practical reason. Kant’s point is constitutional rather than constructive: reason needs boundaries because different legitimate uses of reason need room to operate.

Now we arrive at the compact summary of the transcendental doctrine: “That space and time are only forms of sensible intuition.” This is the first major result of the Critique. Space and time are not properties that things possess independently of us in exactly the same way. They are forms through which objects can be given to sensibility. “Form” here means the structural condition of appearing. Anything that appears to us externally must appear spatially; anything that appears to us internally must appear temporally. Kant is therefore not asking whether space and time are “real” in an ordinary sense. He is asking what kind of reality they have and under what conditions they function.

The next clause is equally important: “we have no conceptions of the understanding, and, consequently, no elements for the cognition of things, except in so far as a corresponding intuition can be given to these conceptions.” This gives us the famous Kantian division between intuition and concept. An intuition gives us an object; a concept allows us to think what is given. Neither is sufficient by itself for ordinary human cognition. Kant will later formulate the principle memorably: “Thoughts without content are empty, intuitions without concepts are blind.” The human mind requires both receptivity and spontaneity: sensibility receives, understanding thinks.

This is why the categories cannot simply be used to generate knowledge of things in themselves. Categories such as causality, substance, unity, plurality, necessity, and so forth are legitimate only within a cognition in which an appropriate intuition can be given. They structure possible experience. They do not function as magical intellectual windows into a supersensible realm. This is one of the most important things to keep in mind as we read the Analytic. Whenever Kant asks whether a category applies somewhere, we should ask: “Where is the corresponding intuition?” If there is no possible sensible intuition, theoretical cognition has lost its footing.

And then Kant makes a distinction that will become absolutely decisive: “while we surrender the power of cognizing, we still reserve the power of thinking objects, as things in themselves.” The distinction between “thinking” and “cognizing” is perhaps the single most important distinction in this paragraph. To think something is not yet to know that something exists, nor to know what it is like. Kant can think the concept of a thing in itself without possessing theoretical cognition of such a thing. The thing-in-itself is therefore not an object that Kant claims to see behind appearances. It marks a limit-concept: the thought of an object considered independently of the conditions under which it can be experienced.

The footnote makes this distinction explicit. Kant says, essentially, that I can think whatever I please so long as I do not contradict myself. But that does not mean that whatever I can think corresponds to a real object. Here he distinguishes “logical possibility” from what he calls “real possibility” or “objective validity.” A concept can be internally non-contradictory without there being an object corresponding to it. I can think of a thousand-winged golden creature without contradiction. That does not establish its existence. So the absence of contradiction is necessary for legitimate thought, but it is not sufficient for cognition.

This is an important place where Kant’s philosophical vocabulary becomes more precise than ordinary language. When we say “I know something,” Kant wants to know what kind of justification establishes that knowledge. A concept alone does not give existence. An empirical intuition can establish actual existence through experience. A priori reasoning can establish possibility under certain conditions. But merely being able to form a coherent concept does not establish an object. This distinction will become devastatingly important when Kant later examines the ontological argument for God.

Now Kant gives us his famous example: freedom. Before the Critique, he says, we would seem to face an impossible contradiction. We want to say that the human will is free, but we also observe human actions occurring in nature and therefore apparently subject to causal necessity. If causality is universally valid for everything that exists, then the human will cannot be free. But if the will is free, then perhaps causality is not universally valid. Without the critical distinction, we seem forced into contradiction.

Notice the wording: “with regard to one and the same being.” That is the hidden source of the contradiction. We are treating the human being under one and the same description in both propositions. We say: the human being is a natural object, therefore causally determined; the human being is a free agent, therefore not causally determined. If both claims concern the same thing in exactly the same sense, they contradict one another.

Kant’s solution is not to choose one side. This is what makes the passage so powerful. He changes the sense in which the human being is being considered. As phenomenon, the human being belongs to the causal order of nature. Every observable action has antecedent conditions. Human behavior, insofar as it is an event in the phenomenal world, is subject to natural laws. But the same being, considered as a thing in itself, can be thought as free. The contradiction disappears because the two predicates are no longer being asserted of the same object under the same mode of consideration.

This is the famous Kantian “two standpoints” solution. We should be careful with the terminology. Kant is not saying that there are literally two human beings inside us, one phenomenal and one noumenal. Nor is he giving us an empirical method for detecting our noumenal self. He is saying that the same rational being can be considered under two distinct modes: as an appearance within nature and as something considered independently of its appearance. Natural causality governs the former. Freedom can be thought of with respect to the latter.

This is the point at which the Copernican revolution becomes an ethical revolution. If we had continued to assume that the categories describe things as they are absolutely and independently of our mode of cognition, causality would have become metaphysically universal. Everything—including the human will—would necessarily fall under it. But once causality is restricted to phenomena, the apparent conflict with freedom disappears. Kant has not proved freedom theoretically. He has shown that natural determinism does not disprove it.

That qualification is absolutely essential. Kant says: “I cannot, by means of speculative reason, and still less by empirical observation, cognize my soul as a thing in itself.” Therefore he cannot theoretically know that the soul is free. He cannot look into himself and discover a noumenal faculty of freedom. Nor can psychology establish transcendental freedom through observation. Freedom, in the strict Kantian sense, lies outside theoretical cognition.

But then comes the remarkable reversal: “while I cannot cognize, I can quite well think freedom.” This is the achievement of the Critique. The concept of freedom is at least thinkable without contradiction once we distinguish the phenomenal and noumenal standpoints. That may sound modest, but Kant thinks it is enormously important. Before criticism, natural necessity seemed to make freedom contradictory. After criticism, natural necessity and freedom can coexist because they belong to different senses in which the same being can be considered.

And now we reach the moral heart of the entire Preface. Kant says: “Suppose now that morality necessarily presupposed liberty.” This is a hypothetical formulation here because he has not yet established the moral law in the First Critique. But he is telling us exactly what will happen later. If morality requires freedom, and speculative reason had demonstrated that freedom is impossible, then morality would have to surrender. Theoretical determinism would destroy practical reason.

But Kant insists that speculative reason has not demonstrated that freedom is impossible. At most, it has demonstrated that freedom cannot be cognized as an object of theoretical experience. Those are completely different claims. The critical distinction therefore prevents speculative reason from making an illegitimate negative conclusion. “I cannot know freedom” does not entail “freedom is impossible.”

This is the precise meaning of Kant’s famous phrase about “denying knowledge” in order to make room for something else. He is not asking us to believe contradictory propositions. He is removing a false contradiction produced by applying phenomenal categories to the noumenal domain. Freedom becomes thinkable. Later, practical reason will do more: it will give freedom a special kind of certainty through the moral law.

This is also where the First Critique begins to point unmistakably toward the Second. The First Critique can establish that freedom is not theoretically contradictory once the critical distinction is made. But it cannot establish the reality of freedom merely by thinking it. Kant knows this. That is why the Second Critique will be necessary. The moral law will provide the practical ground through which freedom becomes known in a special practical sense. The relationship will become reciprocal: freedom is the condition of the moral law, but the moral law is how we become conscious of freedom.

And this takes us directly back to the footnote from the Critique of Practical Reason that you brought me at the beginning. Remember Kant’s distinction there between ratio essendi and ratio cognoscendi: freedom is the ratio essendi, the ground of being, of the moral law; the moral law is the ratio cognoscendi, the ground of our cognition, of freedom. We can now see where that apparently paradoxical formulation comes from. The First Critique prepares the negative condition: freedom is not ruled out by theoretical reason. The Second Critique then supplies the positive practical condition: the moral law makes us conscious of freedom.

There is therefore a beautiful progression:

Theoretical reason asks: Can I know freedom?

Kant: No—not as an object of possible experience.

Theoretical reason asks: Is freedom therefore impossible?

Kant: No. Once phenomenon and thing in itself are distinguished, freedom can be thought without contradiction.

Practical reason then asks: Do I encounter something that necessarily presupposes freedom?

Kant: Yes—the moral law.

And that is how freedom moves from being merely thinkable in the First Critique toward being practically established in the Second.

The passage also helps us understand Kant’s famous claim that “morality does not require the speculative cognition of liberty.” This is crucial. Morality does not require me to possess a metaphysical theory explaining how free will operates outside time and causality. It requires that I act as a responsible rational agent. The practical standpoint does not wait for speculative metaphysics to solve the noumenal world. Kant thinks that would reverse the proper order. The moral law addresses me immediately as an agent. Its validity does not depend upon my first possessing theoretical knowledge of the metaphysical structure that makes my agency possible.

This is one reason Kant’s moral philosophy is much stranger than a simple theory of “free will.” Freedom is not initially a mysterious psychological power to choose between alternatives. Kant’s deepest notion of freedom is autonomy: the capacity of rational will to be governed by a law that reason gives itself rather than merely being determined by external inclinations. This is where the later Formula of Universal Law and Formula of Humanity will become important. But again, we should wait. The First Critique is establishing the metaphysical space in which such a conception of autonomy will be possible.

The final sentence brings the entire argument together: “For this result, then, we are indebted to a criticism which warns us of our unavoidable ignorance with regard to things in themselves, and establishes the necessary limitation of our theoretical cognition to mere phenomena.” Kant’s ignorance is now revealed as protective. By acknowledging what we cannot know, we prevent theoretical reason from colonizing everything. Nature remains nature; moral agency remains possible; freedom is not empirically discoverable but is no longer theoretically excluded.

And this gives us a particularly beautiful way of understanding Kant’s phrase “negative” versus “positive.” The Critique is negative when it says: speculative reason cannot know the noumenal. But precisely because it says that, it is positive with respect to practical reason. It protects the possibility of freedom. It protects morality from being swallowed by mechanism. Later it will protect the rational significance of God and immortality as postulates of practical reason. The boundary is therefore not merely a wall. It is a membrane separating different legitimate uses of reason.

One last thing I would underline before we move on: Kant is not saying that there are two worlds, one phenomenal and one noumenal, and that we simply happen to live in the phenomenal one while secretly possessing access to the noumenal one. That reading will cause enormous trouble later. The distinction is first and foremost a distinction in the standpoint from which an object is considered. The same being can be considered as phenomenon and as thing in itself. The phenomenal standpoint is the domain of possible experience and causal explanation. The noumenal standpoint is the limiting standpoint required if we are to think the object independently of its appearance. Kant grants us the latter as thought, but not as theoretical cognition.

That distinction is going to be one of the load-bearing beams of everything that follows. When we eventually reach Hegel, he will attack precisely the Kantian separation between phenomenon and thing in itself. When we reach Heidegger, he will ask whether Kant’s analysis of the conditions of experience has not already brought us closer to the question of Being than Kant himself recognized. Husserl will radicalize the investigation of how objects are constituted in experience. Derrida will repeatedly return to the strange relationship between what can be thought, what can be known, and what exceeds the conditions of presence. But for now, we should let Kant’s own architecture stand.

The remarkable thing about this passage is that Kant has essentially told us why the Critique had to be written before the moral philosophy could be written. He had to first prevent nature from becoming an absolute metaphysics. Only then could freedom be thought without contradiction. And only after freedom has been secured as a possibility can practical reason take the next step and ask whether there is something within reason itself that actually requires us to regard ourselves as free. That “something” will be the moral law.

This is Kant bringing the Preface to its real climax. The earlier passages established the limits of speculative reason and showed how those limits make freedom thinkable. Here Kant goes further: he explains why the limitation of speculative reason is not merely compatible with religion and morality but, in his view, necessary for them. This is the famous passage containing “I have therefore found it necessary to abolish knowledge, in order to make room for belief” (“Ich musste also das Wissen aufheben, um zum Glauben Platz zu bekommen”). It is one of the most quoted sentences in Kant, but it is also one of the most easily misunderstood. Kant is not celebrating irrational belief, and he is not saying that whenever evidence is lacking we should simply believe whatever we like. He is making a much more precise claim: theoretical reason must surrender a kind of knowledge it cannot legitimately possess so that practical reason can legitimately relate us to ideas that theoretical reason cannot establish as objects of knowledge.

The first sentence is already setting up a parallel between God, freedom, and the soul: “The positive value of the critical principles of pure reason in relation to the conception of God and of the simple nature of the soul, admits of a similar exemplification.” Kant is referring back to the freedom argument from the previous passage. Just as criticism prevents speculative reason from proving that freedom is impossible, it also prevents speculative reason from falsely claiming either to prove or disprove certain supersensible claims. The “simple nature of the soul” refers especially to the traditional rationalist argument that the soul must be a simple, indivisible substance and therefore capable of surviving bodily death. Kant will attack this argument in the Paralogisms. But again, destroying the proof is not equivalent to proving the opposite. Kant is trying to clear the ground.

Then comes the remarkable claim: “I cannot even make the assumption—as the practical interests of morality require—of God, freedom, and immortality, if I do not deprive speculative reason of its pretensions to transcendent insight.” Notice the hierarchy. Freedom, God, and immortality are not being presented here as conclusions of speculative metaphysics. They are assumptions required from the standpoint of practical reason. This distinction will become absolutely decisive in the Second Critique, where Kant will call God, freedom, and immortality “postulates of pure practical reason.” A postulate is not a theorem. It is not something proved by theoretical demonstration. It is something that practical reason has grounds to presuppose because of what practical reason requires.

This is why Kant says speculative reason must surrender its “pretensions to transcendent insight.” “Pretension” is the important word. Kant is not saying that metaphysics is bad because it asks about God or freedom. He is saying it becomes illegitimate when it claims theoretical knowledge of these objects by extending concepts beyond the domain in which they have legitimate application. The categories—causality, substance, necessity, and so forth—are extraordinarily powerful, but they are not universally applicable intellectual instruments for describing anything whatsoever. Their legitimate use is tied to possible experience. Once we extend them beyond that sphere, Kant thinks we transform the supersensible into something phenomenal and thereby destroy precisely the distinction we needed to preserve.

This is the meaning of the extraordinary phrase “converting them into phenomena.” If I try to know God by applying the categories exactly as I apply them to empirical objects, I have effectively treated God as though God were an object within the same experiential order as other things. But then I am no longer talking about God as a supersensible reality. I have reduced the transcendent to the phenomenal. The same problem applies to freedom. If I demand empirical evidence of noumenal freedom, I am asking for freedom to appear as an object within the causal order of nature. But once it appears as an empirical event, I am no longer dealing with transcendental freedom in Kant’s sense.

Now we arrive at the famous sentence: “I must, therefore, abolish knowledge, to make room for belief.” In German: “Ich musste also das Wissen aufheben, um zum Glauben Platz zu bekommen.” The verb “aufheben” is fascinating because it carries several meanings in German: to abolish, cancel, preserve, and lift up. Kant himself is not using it in the later explicitly Hegelian technical sense, but the ambiguity is philosophically suggestive. He is abolishing a certain claim to knowledge while preserving the possibility of another rational relation to the supersensible. We should therefore resist translating the sentence into the crude formula “Kant says knowledge is bad and faith is good.” His point is almost the reverse: knowledge is valuable precisely because we need to know the limits of what knowledge can legitimately claim.

And “Glaube,” belief or faith, also requires care. Kant’s “belief” here is not simply ecclesiastical faith or psychological conviction. It is a rationally grounded practical assent. The exact character of this “belief” becomes clearer later when Kant distinguishes different kinds of assent, especially opinion, belief, and knowledge. Practical belief has a rational necessity that is different from theoretical certainty. We should therefore hear “belief” as something like rational practical commitment, not arbitrary religious credulity.

The next sentence is one of Kant’s most provocative: “The dogmatism of metaphysics … is the true source of the unbelief (always dogmatic) which militates against morality.” This deserves slow reading. Kant is attacking both sides of the traditional metaphysical conflict. Dogmatic metaphysics tries to prove God, freedom, and immortality through speculative reasoning. But when those proofs inevitably fail, the skeptic or atheist can simply turn the same speculative framework against them and claim that because they cannot be theoretically proven, they are false or meaningless. Kant’s criticism says both parties have accepted the same mistaken premise: that supersensible realities must be established or destroyed by speculative theoretical proof.

That is why Kant calls unbelief “always dogmatic.” The point is not that every atheist is psychologically dogmatic. The philosophical point is that the assertion “God does not exist,” when presented as a conclusion of speculative reason about a supersensible reality, can be just as metaphysically overconfident as “God necessarily exists,” if neither position has legitimate theoretical access to the object. The critical position is more restrained: speculative reason cannot establish either proposition as theoretical knowledge. The practical question is different.

This also explains why Kant thinks criticism can be beneficial to religion without being a conventional apologetic for religion. He is not going to rescue theology by producing a better proof of God. He is going to dismantle the demand that God must be theoretically demonstrable before rational commitment to God can have legitimacy. This is a very different project from traditional natural theology.

The second paragraph shifts from the philosophical argument to Kant’s vision of what the Critique will leave to posterity. He says the future metaphysical system will replace “random groping” with scientific method. This phrase connects directly to the opening of the Preface, where metaphysics was described as an arena of endless conflict. The problem was not that human beings lacked intelligence. The problem was that they lacked a secure method. They were attempting to answer questions before determining whether their conceptual instruments were capable of answering them.

This is why Kant’s concern with method is so profound. He is not merely proposing another metaphysical doctrine to compete with Leibnizianism, Wolffianism, empiricism, skepticism, or rational theology. He is attempting to change the rules under which these disputes are conducted. Before asking whether the soul is immortal, ask what would count as cognition of the soul. Before asking whether the universe has a first cause, ask whether “the universe as a totality” can be given as an object. Before asking whether God exists as a necessary being, ask whether the concept of necessity can legitimately establish the existence of a supersensible being.

That is what Kant means by replacing groping with method.

Then he makes an extraordinary claim about education: the Critique will benefit “the inquiring mind of youth” by preventing students from wasting their intellectual powers on “the idle attempt to invent new ideas and opinions.” This is almost autobiographical. Kant has spent years inside the metaphysical tradition. He knows its temptations from the inside. His target is not philosophical creativity itself but the endless production of metaphysical systems whose fundamental legitimacy has never been examined.

And then comes the claim that the Critique will benefit morality and religion “above all.” Here Kant introduces what he calls the “Socratic method,” meaning the method of exposing the ignorance underlying an opponent’s claim. The reference to Socrates is important. Socrates often defeats his interlocutors not by replacing their doctrine with another doctrine, but by revealing that they do not actually know what they thought they knew. Kant sees the Critique as performing something similar on metaphysics. He does not necessarily prove the religious object; he shows that the supposedly decisive objections and counterproofs often exceed the legitimate powers of theoretical reason.

The Greek philosophical background here is worth keeping in mind. Socrates’ “I know that I do not know” becomes, in Kant, something much more technically developed: “I know the conditions and limits under which I can know.” Kant transforms Socratic ignorance into transcendental critique. It is not merely personal humility. It is a systematic investigation into the structure of cognition.

Then Kant says something very interesting: “the world has never been, and, no doubt, never will be without a system of metaphysics of one kind or another.” This returns us to the very first paragraph of the Preface. Human reason cannot stop asking these questions. Metaphysics is not simply an academic hobby. It arises from reason’s own nature. This is why Kant cannot simply recommend that humanity abandon metaphysics. The questions will return. The only question is whether they will return in a disciplined or undisciplined form.

That is what Kant means when he says philosophy must “render it powerless for harm, by closing up the sources of error.” This is a remarkable formulation. Philosophy cannot prevent metaphysics from existing; it can prevent metaphysics from becoming dogmatic. Kant’s project is therefore prophylactic. He wants to identify the mechanisms by which reason generates illusion. The Transcendental Dialectic will later be precisely this: an anatomy of reason’s tendency to treat its own necessary ideas as though they were objects of knowledge.

And now Kant turns to a potentially sensitive point: the loss of the “fancied possessions” of speculative reason does not harm humanity. The “schools” will lose something, but humanity will not. “The schools” here means the professional philosophical establishment—the academic traditions that had claimed specialized access to metaphysical truths. Kant is challenging their monopoly.

This is one of the most democratic passages in the Preface. Kant says that ordinary human beings already possess the fundamental rational convictions that matter: hope for a future life, consciousness of freedom, and belief in a wise author of nature. What philosophy has failed to do is establish those convictions through abstruse metaphysical proofs. Kant thinks this failure is not catastrophic because the convictions do not originate primarily in those proofs.

Look at how he explains them. “The hope of a future life arises from the feeling … that the temporal is inadequate to meet and satisfy the demands of his nature.” This is not yet Kant’s formal argument for immortality. It is a phenomenology of the human moral condition. Human beings experience a tension between moral vocation and finite temporal existence. We demand a completion that empirical life does not seem to provide. That gives rise to hope for a future life.

Likewise, “the clear exhibition of duties in opposition to all the claims of inclination, gives rise to the consciousness of freedom.” This is incredibly important. Kant is suggesting that freedom is encountered practically rather than theoretically. When I experience myself as obligated to do something despite my desires, I become conscious of myself as capable of standing in relation to a law that is not simply identical with inclination. This will become central to the Second Critique. Freedom is not first discovered by looking inward for some metaphysical object called “the free will.” It is disclosed through practical reason’s demand upon me.

And then God: “the glorious order, beauty, and providential care, everywhere displayed in nature, give rise to the belief in a wise and great Author of the Universe.” Kant is describing the traditional teleological intuition. Nature’s order naturally leads human reason toward the idea of an intelligent author. But he does not yet endorse the inference as a theoretical proof. Indeed, the First Critique will later show why the teleological argument cannot establish God’s existence with the certainty demanded by theoretical cognition.

What Kant preserves is the rational significance of the conviction without preserving the dogmatic proof. That is the pattern throughout the passage: preserve the phenomenon, reject the illegitimate metaphysical explanation, and then ask whether another form of rational justification is possible.

This is where Kant’s famous distinction between “the schools” and the “great mass of men” becomes especially interesting. He is not simply saying ordinary people are wiser than philosophers. Rather, he is saying that the professional philosopher has no right to claim a special theoretical insight into supersensible matters that ordinary rational beings could never possess. If the supposed proof of immortality requires an elaborate argument about the simplicity of substance that ordinary reason cannot even comprehend, Kant asks what exactly has been accomplished.

The “schools” have confused sophistication with knowledge.

This is why the closing quotation from Persius is so fitting: “Quod mecum nescit, solus vult scire videri.” Roughly: “What he does not know with me, he alone wishes to seem to know.” The line is a satire of pretended knowledge. Kant is saying that the metaphysician often wants to appear to possess a secret key to truths that everyone else can only receive secondhand. Critique breaks this monopoly by asking the metaphysician to submit his claims to the same tribunal of reason.

There is a deep connection here with Kant’s earlier metaphor of reason as a tribunal. The “schools” claim to possess private jurisdiction over the supersensible. Kant says: no. The tribunal belongs to reason itself, and its laws are public. If a claim cannot survive rational examination, no academic pedigree can save it. Conversely, if a moral conviction is rationally grounded, it does not become less legitimate merely because it cannot be expressed as a scholastic proof.

And now we can see the full trajectory of the Preface. Kant begins with metaphysics in crisis. Dogmatism claims too much; skepticism attacks dogmatism; neither possesses a secure method. Kant turns reason upon itself. He investigates the conditions of a priori cognition. He discovers that experience itself has a priori structure. He then discovers that the same structures cannot legitimately be extended beyond experience. This resolves the metaphysical contradictions by distinguishing phenomena from things in themselves. Then he realizes that this limitation is not merely restrictive: it protects practical reason from speculative reason’s attempt to reduce everything to natural mechanism. Freedom becomes thinkable. God and immortality remain unavailable as theoretical knowledge but become candidates for practical postulation.

So “I must abolish knowledge to make room for belief” is really the culmination of a much longer argument. Kant is not making an emotional turn toward faith after giving up on philosophy. The exact opposite is happening. His philosophy has become sufficiently rigorous to distinguish several things that previous metaphysics had collapsed: knowing, thinking, believing, experiencing, acting, and morally obligating oneself. Once those distinctions are made, religion and morality no longer need to defend themselves by pretending to possess a kind of theoretical knowledge they do not possess.

And this is precisely where your initial decision to start with the First Critique becomes especially fruitful. When you showed me the Preface to the Second Critique, Kant’s statement that freedom is the “keystone” of the entire system could initially look like a sudden elevation of morality over metaphysics. Now we can see that the First Critique has been preparing the ground for it from the beginning. The First Critique says: speculative reason cannot prove freedom, but it also cannot legitimately eliminate it. It clears the space. The Second Critique will then say: practical reason encounters the moral law, and through the moral law we become conscious of freedom. The relation between the two works is therefore not accidental. The second grows directly out of the boundary established by the first.

There is also something beautifully paradoxical about Kant’s conception of philosophy here. Philosophy becomes more powerful by admitting that it cannot know everything. It serves religion by refusing to turn religion into bad metaphysics. It serves morality by refusing to reduce human action to a purely theoretical object. It serves science by clarifying the conditions of legitimate cognition. And it serves ordinary reason by preventing professional metaphysicians from claiming an exclusive monopoly over questions that belong to humanity as such.

So if we were to condense this entire passage into one sentence, I would put it this way: Kant’s Critique does not save God, freedom, and immortality by proving them as objects of theoretical knowledge; it saves the rational possibility of relating to them by preventing theoretical reason from falsely proving or disproving them, thereby protecting the practical domain in which morality, freedom, and rational faith can legitimately emerge.

And with that, the Preface is essentially finished. What Kant has given us is not yet the machine, but the blueprint for the machine. The next thing we should do is resist the temptation to jump straight to the famous “Transcendental Aesthetic” without understanding the first distinction Kant needs to establish: the distinction between pure and empirical cognition. That distinction is going to make the words “a priori,” “a posteriori,” “analytic,” “synthetic,” “intuition,” “concept,” “sensibility,” and eventually “transcendental” much more precise. Once those are clear, we can enter Kant’s actual argument rather than merely hearing him announce it.

This is a very important closing passage of the Preface because Kant now makes an institutional distinction that is easy to miss: he is not saying that the Critique itself should become “popular philosophy.” In fact, he explicitly says that it cannot and should not. He is assigning philosophy, particularly academic philosophy, a peculiar public function: the philosopher does work that the public does not need to understand directly, but from which the public nevertheless benefits. The analogy is almost like a constitutional court or a scientific laboratory. Most people do not need to understand the technical arguments of the Critique, but they benefit if the intellectual conflicts that eventually spill into religion, politics, morality, and culture have been subjected to rigorous examination. This is why Kant says the speculative philosopher remains “the sole depositor” of a science that benefits the public “without its knowledge.” The phrase “sole depositor” is interesting because Kant is not celebrating philosophical elitism in the ordinary sense. He is saying that there is a specialized labor of reason which cannot simply be translated into popular slogans without losing its force. The public may rightly care about God, freedom, immortality, morality, materialism, atheism, and so forth, but the technical question of what reason can legitimately establish about them requires a level of abstraction that most people have no reason to master. Kant therefore preserves a division between philosophy as a rigorous science and philosophy as popular instruction.

But there is an important tension here. Kant says the Critique “can never become popular,” and yet he insists that its consequences are public. This anticipates a problem that will recur throughout modern philosophy: the difference between the accessibility of a conclusion and the accessibility of the argument that establishes it. A person might reasonably believe that they are free without being able to explain Kant’s transcendental argument for freedom. Likewise, someone might possess a moral conviction without understanding the structure of the categorical imperative. Kant’s point is not that ordinary people are incapable of reason. Quite the contrary. His earlier remarks emphasized that ordinary human beings already possess rational moral and religious orientations. What they lack is necessarily the specialized theoretical apparatus required to adjudicate the metaphysical conflicts surrounding those orientations. Philosophy therefore has a kind of custodial function. It does not manufacture the fundamental concerns of humanity; it examines the conceptual structures through which those concerns are articulated.

This is why Kant says that “finespun arguments in favour of useful truths” make just as little impression on the public as the equally subtle objections brought against them. That sentence is almost an attack on apologetics and anti-apologetics simultaneously. Suppose someone produces an extraordinarily sophisticated metaphysical proof of God. Kant says that this is not what actually grounds ordinary religious conviction. But suppose another philosopher produces an equally sophisticated refutation. That also does not necessarily destroy ordinary religious conviction. Both are operating at a level that is disconnected from the actual sources of popular belief. Kant therefore wants to move the philosophical question somewhere deeper: not “Which side has the cleverer argument?” but “What are the conditions under which either side could legitimately claim knowledge?” That is the critical question.

And this explains why metaphysical controversy eventually becomes dangerous even to the public. The public may not understand the arguments, but it can inherit their consequences. A philosophical dispute about whether freedom is possible can become a theological dispute about sin and responsibility; a dispute about whether God can be demonstrated can become a dispute about atheism; a dispute about causality can become a dispute about determinism; a dispute about the soul can become a dispute about immortality. The technical metaphysical quarrel may therefore have social and moral consequences even when almost nobody has read the arguments. Kant thinks the university has a responsibility to settle the philosophical jurisdiction before the conflict becomes a cultural one.

The language becomes especially striking when Kant says that criticism can save “metaphysicians (and, as such, theologians too)” from controversies and “the consequent perversion of their doctrines.” Notice that Kant is not simply anti-theological. He thinks theology itself becomes vulnerable when it depends upon uncriticized metaphysics. If a theologian bases religion on a bad metaphysical argument, then an attack on that argument threatens the religion that has been attached to it. Kant’s strategy is therefore protective but destructive at the same time: destroy the bad metaphysical foundations so that theology does not depend upon claims that speculative reason cannot support. This is one reason the Critique was so consequential for later theology. Kant does not simply say, “religion is true.” He fundamentally changes the kind of question that theology can legitimately ask.

Then comes the astonishing list: criticism can strike “at the root of materialism, fatalism, atheism, free-thinking, fanaticism, and superstition,” as well as “idealism and scepticism.” The list is deliberately symmetrical. Kant is attacking both extremes. Materialism and fatalism threaten to reduce human beings to natural necessity; fanaticism and superstition pretend to possess supernatural knowledge without legitimate grounds. Atheism, in the particular sense Kant is targeting, can become dogmatic if it claims theoretical knowledge of the nonexistence of a supersensible God; superstition can become dogmatic if it claims theoretical knowledge of supernatural entities. Idealism and skepticism are likewise dangerous when they undermine the possibility of knowledge altogether. Kant’s critical position is therefore supposed to occupy a narrow middle path: neither dogmatic affirmation nor dogmatic negation.

We should pause over “free-thinking.” Kant’s German term is “Freigeisterei.” It does not simply mean the modern liberal virtue of independent thinking. In Kant’s eighteenth-century context, it can refer to free-thinking in the sense of a rejection of religious authority or established belief. Kant’s criticism is not that one should never question religion. In fact, the entire critical project insists that everything must submit to examination. What he rejects is the idea that freedom from dogma means freedom from rational standards. Critical philosophy is neither obedient faith nor rebellious skepticism. It is disciplined autonomy.

His remark about governments is equally interesting. If governments want to interfere with scholarship, Kant says, they would be wiser to encourage criticism than to protect the “despotism of the schools.” Here he is making a political point about intellectual freedom. Governments sometimes intervene because they fear philosophical controversy will disturb society. Kant’s answer is essentially: let philosophy settle its disputes openly and rigorously rather than allowing institutions to protect established doctrines. The phrase “destruction of cobwebs” is particularly vivid. The schools make a tremendous noise when their inherited metaphysical constructions are attacked, but the public scarcely notices that those constructions have been destroyed. Kant is mocking the idea that the collapse of a scholastic metaphysical system is itself a social catastrophe.

Now we reach perhaps the most important conceptual clarification in this passage: “This critical science is not opposed to the dogmatic procedure of reason in pure cognition; … but to dogmatism.” This distinction is absolutely essential for reading Kant correctly. Kant is not anti-dogmatic in the sense that he thinks philosophy should never demonstrate anything. Quite the opposite. He says that “pure cognition must always be dogmatic,” meaning that when reason has established principles a priori, the subsequent procedure must proceed by strict demonstration from those principles. “Dogmatic” here means systematic, demonstrative, proceeding from established principles. The enemy is “dogmatism,” which is the presumption that we can simply begin using those principles without first asking how reason acquired them and whether it has the right to use them.

This distinction is one we should hold onto throughout the entire Critique: dogmatic method is legitimate; dogmatism is illegitimate. Kant does not want to destroy systematic philosophy. He wants to make systematic philosophy possible. He is not saying, “Stop making metaphysical arguments.” He is saying, “Before making metaphysical arguments, determine the legitimate scope and source of the principles with which you are reasoning.”

The German is useful here because Kant’s distinction is even sharper. He writes: “Diese Kritik steht nicht dem dogmatischen Verfahren der Vernunft in ihrem reinen Erkenntnisse entgegen, als vielmehr dem Dogmatismus…” The critical project opposes not the “dogmatic procedure” of reason in pure cognition but “Dogmatism.” The distinction is therefore deliberate. A scientific metaphysics will still have to be rigorous, systematic, demonstrative, and a priori. What it cannot do is simply inherit principles from the philosophical tradition and assume their legitimacy.

And this brings us back to Christian Wolff, whom Kant names as “the greatest of all dogmatic philosophers.” This is an important historical moment because Kant himself came out of the Wolffian tradition. Wolff was not a sloppy thinker. In fact, Kant praises him precisely for qualities Kant himself values enormously: fixed principles, clear definitions, rigorous demonstrations, and systematic organization. Wolff’s failure was not that he was insufficiently rigorous within his system. His failure was that he never critically examined the “organon” through which pure reason was operating. He possessed a sophisticated instrument but had not first investigated the instrument itself.

This is perhaps the best way to understand the entire First Critique: Kant wants to turn philosophy’s attention from the products of reason toward the conditions of reason’s production. Wolff asks: Given the principles of rational metaphysics, what follows? Kant asks: How did we acquire these principles? What makes them legitimate? What objects can they legitimately apply to? What happens when we extend them beyond those objects? The difference is enormous.

And the phrase “criticism of the organum” is particularly revealing. “Organon” means instrument or tool. Kant is essentially saying that Wolff needed a critique of the instrument of metaphysics itself. What is reason? What are its powers? What are its limits? What happens when reason operates without empirical material? What principles does it possess a priori? Which of those principles legitimately determine experience, and which merely tempt us into transcendental illusion? These questions are what Kant means by the “Critique of Pure Reason.”

There is also a subtle historical irony here. Aristotle’s Organon was the classical name for the logical works that provide the instruments of thought. Kant’s language therefore places him within a very old philosophical tradition: before we use reason to construct a science, we need to understand the instrument with which the science is constructed. But Kant’s move is different from Aristotle’s. Aristotle develops formal logic and analyzes valid inference; Kant asks what makes a priori cognition of objects possible at all. This is one of the places where Aristotle becomes an important background for understanding Kant, but not because Kant simply follows Aristotle. Kant is transforming the question.

The same is true of Leibniz and Wolff. Leibniz had attempted to construct a rational metaphysics in which the structure of reality could be derived through principles of reason. Wolff systematized much of this into an enormous philosophical architecture. Kant inherits the aspiration toward systematicity but questions whether the architecture has legitimate foundations. This is one reason the Critique can feel simultaneously like an attack on rationalism and its ultimate fulfillment. Kant does not abandon the rationalist demand for necessity; he radicalizes it. He asks whether reason can itself justify the conditions under which it claims necessity.

And then Kant ends with a wonderfully severe phrase: “to change labour into sport, certainty into opinion, and philosophy into philodoxy.” “Philodoxy” literally means something like love of opinion or love of mere opinion. This is Kant’s final warning against what we might today call intellectual improvisation. Once you reject rigorous demonstration, philosophy becomes a competition in who can produce the most interesting position. You can endlessly generate “views,” but you have abandoned the demand for knowledge. Kant would rather have a smaller philosophy with genuine necessity than an enormous marketplace of ingenious opinions.

This also clarifies something important about the tone of the Critique. Kant can sound almost authoritarian about rigor: no hypotheses, no opinions, no loose analogies, no premature conclusions. But the reason is not simply stylistic pedantry. He thinks the object itself requires this severity. If we are investigating the conditions under which reason can claim necessity, then we cannot casually introduce assumptions and call them conclusions. The method has to embody the very certainty it is investigating.

So, at the end of the Preface, we can now see Kant’s project with much greater precision. He is not proposing skepticism. He is not proposing irrational faith. He is not rejecting metaphysics. He is not rejecting systematic philosophy. He is not rejecting “dogmatic procedure” in the legitimate sense. He is doing something much more surgical: he wants to distinguish legitimate a priori cognition from illegitimate metaphysical overreach, and he believes this requires reason to examine its own instrument before attempting to use it.

That is why I would mark this sentence as the hinge of the whole First Critique:

“Dogmatism is thus the dogmatic procedure of pure reason without previous criticism of its own powers.”

Everything that follows can be read as an attempt to perform the missing criticism.

And this gives us a very good place to stop treating the Preface as merely an introduction. Kant has now told us exactly what problem he thinks philosophy has failed to solve. The next question is: what does he mean by “pure reason” in the first place? Before we enter the Transcendental Aesthetic, we should therefore take the opening of the Introduction very slowly. The first words—“There can be no doubt that all our cognition begins with experience”—look almost empiricist. Kant then immediately says that it does not follow that all cognition “arises out of experience.” That tiny distinction is going to open the entire Kantian system.

This is Kant closing the Second Edition Preface by doing something very unusual: he is simultaneously defending the architecture of the First Critique and admitting that the architecture can be presented badly. That distinction is important. Kant says, essentially, “I have not changed the philosophy, but I have changed the exposition.” The 1787 edition is therefore not a new philosophy in his mind. He believes that the underlying system discovered in 1781 remains intact, but that certain passages were obscure enough to invite serious misunderstanding. This is one reason the difference between the A-edition of 1781 and the B-edition of 1787 matters so much for us. We should not simply read the 1787 text as though it were the only Kant. Kant himself regarded the system as stable while believing that its presentation could be improved. And some of those improvements are philosophically substantial enough that scholars still debate whether the B-edition merely clarifies the A-edition or subtly changes the emphasis of the argument.

The first striking image is Kant’s description of pure speculative reason as an “organic structure”: “pure speculative reason is an organic structure in which there is nothing isolated or independent, but every single part is essential to all the rest.” The German is “ein organisiertes Ganzes,” an organized whole. This is not just a metaphor for a well-written book. Kant thinks reason itself has systematic organization. This is why he is so concerned with completeness. If one principle is wrong, it should eventually create consequences elsewhere. His method is therefore both constructive and diagnostic: begin with the simplest elements and build upward toward the whole; then start from the whole and work backward toward the individual elements. If the two directions converge, the system has what we might call structural coherence.

This is also where Kant’s conception of philosophy begins to anticipate something that will become enormous in Hegel. Kant does not yet have Hegel’s dialectical system, and we should not collapse the two, but the idea that philosophical concepts cannot be understood as isolated propositions becomes increasingly important after Kant. Hegel will eventually object that Kant’s categories remain insufficiently dynamic and that Kant does not allow the determinations of thought to generate one another dialectically. But the demand that philosophy be an articulated whole is already here. Likewise, much later, Heidegger will repeatedly insist that individual Kantian concepts cannot be understood outside the question of the structure of transcendental cognition. Husserl will be interested in the systematic investigation of the conditions under which objects can be given to consciousness. Derrida will, in a very different way, become intensely interested in the places where supposedly complete philosophical systems depend upon distinctions they cannot entirely stabilize. So we can already see why this paragraph is architecturally important, but for now we should resist following those descendants too far. Kant is still the center of gravity.

There is another beautiful point in the passage: Kant says that his confidence in the system is not “vanity” but comes from the fact that the result remains the same whether one proceeds “from the simplest elements up to the complete whole” or “backwards from the whole to each part.” This resembles a kind of philosophical triangulation. If you build the system upward and then reconstruct it downward, you should arrive at the same structure. The system therefore tests itself internally. This is also why Kant thinks the Critique cannot simply be modified piecemeal. If you alter a foundational distinction, the alteration propagates throughout the system. Change his conception of space and time, and you affect mathematics, sensibility, phenomena, and the limits of knowledge. Change the deduction of the categories, and you affect the objective validity of the understanding. Change the relation between internal and external sense, and you affect his account of experience itself.

This brings us to the areas he says he has revised: the “æsthetical part,” time, the Deduction of the categories, the principles of pure understanding, and the Paralogisms. These are not random editorial choices. They are exactly the places where the machinery of the Critique is most delicate. The “Transcendental Aesthetic” is where Kant establishes space and time as forms of sensible intuition. The “Transcendental Deduction” is where he tries to establish the objective validity of the categories. The “Analytic of Principles” shows how those categories actually govern possible experience. The Paralogisms concern the mistaken attempt to derive metaphysical conclusions about the soul from the mere fact of self-consciousness. In other words, Kant is revising the joints where sensibility, understanding, and self-consciousness meet.

Then comes the footnote, which is philosophically enormous. Kant says that the one genuinely new addition is “a new refutation of psychological Idealism” and “a strict demonstration … of the objective reality of external intuition.” This is his response to a problem that had become increasingly important after the first edition: how can we know that an external world exists? Kant is dissatisfied with the idea that we must simply believe in an external world while having immediate access only to our own representations. He thinks that would leave philosophy in an embarrassing position. If all I immediately know is what appears “in me,” then how do I distinguish an actual external world from a dream, hallucination, or mere imagination?

This is why the famous phrase “I think” is not enough for Kant. Descartes had begun from certainty of the thinking self: “Cogito, ergo sum.” Kant accepts a crucial part of this but radically changes what follows from it. I can certainly be conscious of myself as thinking, but this intellectual consciousness does not yet tell me what I am as an object, nor does it give me a complete knowledge of my existence. In fact, Kant’s argument will turn the relationship around: my consciousness of myself as existing in time requires a relation to something permanent outside me.

This is the core of the “Refutation of Idealism.” Kant is targeting what he calls “problematic idealism,” especially the Cartesian position that the existence of external objects remains doubtful because only one’s own existence is immediately certain. Kant’s argument is surprisingly subtle: I cannot even determine my own existence in time without reference to something permanent that is not merely one of my changing representations. Internal experience alone gives me a succession of representations, but succession requires a temporal framework, and Kant argues that determining my own existence in time requires something permanent against which change can be determined. That permanent cannot simply be another representation because representations themselves are variable.

The crucial sentence is: “I am conscious, through internal experience, of my existence in time.” This seems innocent, but it is doing enormous work. Kant is saying that self-consciousness is not a timeless, self-contained point. The “I” that is conscious of itself is encountered as existing through temporal determination. My representations change. My thoughts change. My sensations change. My internal states change. To determine that change as change, however, we need permanence. And Kant argues that this permanence cannot be supplied merely by another changing representation within inner sense.

Here we need to be very careful with the word “permanent.” Kant is not saying that there is some unchanging external object sitting somewhere that we can simply point to. He is making a transcendental argument about the conditions of experience. The question is not, “What particular external object proves that the external world exists?” The question is, “What must be presupposed for my experience of myself as temporally determined to be possible at all?” His answer is: there must be an external relation to something permanent.

This is why Kant says that consciousness of one’s existence in time is “identical with the consciousness of a relation to something external to me.” That is the shocking reversal. The external world is not inferred from an already complete internal self-consciousness. Rather, the very possibility of determining myself internally depends upon a relation to something external.

This is profoundly different from Descartes. Descartes begins with the certainty of the thinking self and then faces the problem of getting from the inner certainty of thought to the outer certainty of the world. Kant argues that the opposition between “inner” and “outer” has been misconceived. Inner experience itself is temporally structured, and the determination of that temporal existence requires the external. So instead of saying:

“I know myself first, and then I infer the external world,”

Kant says something much closer to:

“I can determine myself as existing in time only through a relation to what is external.”

The relation is therefore reciprocal at the transcendental level. The external world is necessary for the determination of inner experience, while inner experience is itself one mode through which experience becomes conscious.

This also helps explain why Kant is so careful about the distinction between “representation” and “object.” He says that a representation of something permanent is not the same as a “permanent representation.” That distinction is extraordinarily important. Suppose I look at a tree. My representation of the tree changes: I blink, move, change perspective, remember it, imagine it, see it in different lighting. The representation is not permanent. Yet the changing representations can refer to something that is relatively permanent. Therefore, we must not confuse the stability of the object with the stability of the representation.

Kant is thus resisting a very tempting form of idealism: “Since everything I experience is a representation, perhaps the external world is nothing more than my representations.” No. The representation can change while referring to something whose existence is not reducible to the representation. The representation is an appearance; the object of the appearance is not thereby identical with the representation itself.

But—and this is absolutely crucial—we must not read Kant as suddenly becoming a naïve realist. He does not say that we now possess access to things in themselves. His entire critical distinction remains intact. We experience external objects as phenomena. Their existence is objectively real within experience, but this does not mean we have theoretical cognition of what those objects are independently of the conditions of possible experience.

This is why the final sentence is so strange: “To the question How? we are no more able to reply, than we are, in general, to think the stationary in time, the coexistence of which with the variable, produces the conception of change.” Kant is acknowledging a limit in his own transcendental argument. He can establish that permanence is necessary for temporal determination, but he cannot give a positive metaphysical picture of what this permanence is “in itself.” He cannot explain the noumenal mechanism by which external permanence produces internal temporal consciousness. That would again be an attempt to transcend the limits of cognition.

And notice the extraordinary consistency here. Kant has just spent the Preface telling us that the Critique will establish boundaries. Now, even when he presents his new argument for the external world, he refuses to cross those boundaries. He can establish a necessary condition for experience without thereby claiming knowledge of the ultimate metaphysical nature of the thing that satisfies that condition.

This gives us a very useful preliminary map of Kant’s three levels:

First, there is representation: what is given in consciousness.

Second, there is the object as phenomenon: the object insofar as it is encountered within possible experience.

Third, there is the thing in itself: whatever the object may be independently of the conditions through which it appears to us.

The Refutation of Idealism is principally concerned with securing the second level. Kant wants to say that empirical external objects are not merely fictional constructions inside my mind. External experience is genuinely objective. But this does not give me theoretical access to the third level.

This is also why Kant’s treatment of “I am” is so different from a simple Cartesian cogito. Kant says that if I had an intellectual intuition of myself, then I could determine my existence intellectually. But I do not. My self-consciousness, as it becomes an experience of myself, is mediated through inner sense and therefore through time. The “I think” accompanies my judgments and representations, but it is not itself a sensory portrait of the self. I know that “I think,” but I do not thereby know what the thinking “I” is as a metaphysical substance.

That distinction will become absolutely devastating for the traditional rational psychology of the eighteenth century. The rational psychologist takes the “I think” and tries to infer that the soul must be a simple substance, that it is immaterial, that it is immortal, and so forth. Kant will show that these conclusions do not follow. The unity of self-consciousness is not the same thing as knowledge of a simple metaphysical substance. This is why the Paralogisms are one of the major revisions between the first and second editions.

So the Second Edition Preface ends by preparing us for one of Kant’s deepest methodological principles: we must distinguish the conditions under which an object can be experienced from claims about what that object might be independently of experience. Kant’s transcendental argument operates in the first register. It asks, “What must be the case for experience to be possible?” It does not pretend to answer, “What is reality absolutely like behind experience?”

And that is probably the most useful way for us to approach the Critique from here onward. We should treat Kant almost like a mathematician whose definitions have to be nailed down before we allow ourselves to follow the proof. Whenever he says “intuition,” “representation,” “concept,” “object,” “experience,” “a priori,” “transcendental,” “phenomenon,” “thing in itself,” or “understanding,” we should stop and ask exactly what role the term is playing. Kant’s arguments become much less mysterious once those terms stop sliding around.

One final historical point is worth preserving for later. This passage is where the pressure exerted by Descartes and Leibniz becomes particularly visible. Descartes gives Kant the problem of the isolated thinking subject and the external world. Leibniz gives him the problem of how rational principles can reach beyond experience. Wolff systematizes Leibnizian metaphysics into the very “dogmatic” system Kant wants to reform. Hume then attacks causality and wakes Kant from his “dogmatic slumber,” forcing him to ask how necessary connections can be known at all. Kant’s response to all four is neither simple acceptance nor simple rejection. He takes the problems they generate and asks a new question: what must the subject contribute to experience for objective cognition to be possible in the first place?

That question is the doorway through which we are about to enter the Critique proper.

This final paragraph of the 1787 Preface is easy to overlook because Kant seems to be doing something merely editorial: explaining why he revised some passages, omitted others, and expects readers to compare the second edition with the first. But philosophically, it is extremely revealing. Kant is telling us how he thinks a philosophical system ought to be read. More importantly, he is telling us something about what he thinks a philosophical system is. The central word here is “whole.” He is not asking the reader to evaluate isolated propositions one at a time, as though the Critique were a collection of independent arguments. He wants the reader to understand the internal architecture of the system first and then judge individual propositions in light of that architecture. This is already an important methodological warning for us as we begin reading Kant: if we encounter a sentence that seems strange, we should not immediately decide that Kant has contradicted himself. We should first ask what function that sentence has within the entire structure of the Critique.

The first paragraph is straightforward on the surface. Kant says that in making the second edition clearer, he has omitted or shortened passages from the first edition that were not necessary to the completeness of the argument. But notice the distinction between completeness and usefulness. A passage can be useful without being structurally necessary. This is a very Kantian distinction. He is concerned with the “system” rather than with maximizing exposition. He admits that some readers will miss the omitted material, and therefore directs them back to the first edition. This is one reason the difference between the A-edition of 1781 and B-edition of 1787 matters so much. Kant did not simply revise his prose. He substantially rewrote parts of the Transcendental Deduction, changed his treatment of the paralogisms, added the famous Refutation of Idealism, and altered the presentation of the Transcendental Aesthetic. Consequently, when we eventually reach those sections, it will be worth reading A and B side by side. In some cases, the differences illuminate what Kant himself came to regard as vulnerable or obscure in his first formulation.

The next paragraph introduces one of Kant’s characteristic tensions: he wants clarity, but he does not want “popularity.” He praises those German thinkers capable of combining “profundity of view” with “lucid exposition,” but immediately says that he himself does not possess this talent. There is a kind of deliberate modesty here, but it is also slightly ironic. Kant knows that the Critique is difficult. He is not apologizing for difficulty as though difficulty were inherently virtuous; rather, he thinks the difficulty is partly generated by the nature of the subject. He distinguishes what he calls the “scholastic” character of the science from its popular presentation. “Scholastic” here does not mean “badly academic” in the modern sense. It means systematic, rigorous, technical, disciplined. Kant thinks a science of pure reason cannot be made genuinely rigorous by simply making it conversational or accessible. There is something about the subject matter that requires conceptual precision.

This connects directly to the distinction he made earlier between “discursive or logical clearness” and “intuitive or aesthetic clearness.” Logical clarity means that the concepts and relations between them are precisely established. Intuitive clarity means that examples, illustrations, analogies, and concrete cases make the argument easier to grasp. Kant says that he has prioritized the first. This is crucial for how we should read him. If we continually demand that Kant give us an intuitive picture of every abstract distinction, we will sometimes misunderstand what he is doing. He is building a conceptual machine. We can certainly use examples to understand the machine, and I will give you those examples, but we should always return to the conceptual structure itself. Otherwise, as Kant says earlier, the examples can obscure the “articulation” of the whole.

Then comes a remarkable sentence: “the danger is not that of being refuted, but of being misunderstood.” That tells us how Kant regarded the situation in 1787. He does not think the principal problem is that his fundamental system has been successfully disproved. His concern is that readers will seize upon individual formulations and misunderstand their relationship to the whole. This is exactly what happened historically. The first Critique was famously difficult to understand, and Kant’s own contemporaries interpreted it in radically different ways. Some regarded him as an idealist who denied the external world; others thought he had merely reproduced traditional metaphysics in a new vocabulary; still others interpreted the thing-in-itself as an incoherent remnant of dogmatism. Kant is therefore defending not merely particular propositions but the systematic coherence of his project.

The word “Propädeutik” is worth stopping over. Kant says he is elaborating the system of this “Propädeutik.” A propaedeutic is an introductory or preparatory discipline: something that prepares the way for a more complete science. This tells us something fascinating about the status of the Critique. Kant does not consider the Critique itself to be the completed metaphysical system. It is the examination that establishes the possibility, limits, and principles upon which such a system can be constructed. Think of the Critique as determining the foundation and architectural constraints before the building of metaphysics itself. The “Metaphysics of Nature” and “Metaphysics of Morals” would then be constructed on that foundation. This is exactly why the title “Critique of Pure Reason” matters: it is not “The System of Reality.” It is an investigation into the capacity of pure reason to establish such a system.

This also helps us understand why Kant repeatedly insists that the Critique is not itself metaphysics in the ordinary sense. Metaphysics traditionally asks questions about God, soul, world, substance, causality, freedom, and ultimate reality. Kant’s question is more prior: under what conditions can reason legitimately ask and answer questions of this sort? This is a second-order inquiry. Instead of immediately asking, “Does the soul exist as an immaterial substance?” Kant asks, “What would have to be true of cognition for such a proposition to constitute knowledge?” Instead of immediately asking, “Is the world finite?” he asks, “What happens when reason applies the concept of totality to the world as a whole?” The Critique therefore operates at a peculiar level: it investigates the possibility of metaphysics before constructing metaphysics.

His mention of his age is also more significant than it initially appears. “As I have advanced pretty far in years” — he had reached sixty-four — Kant says that he needs to economize his time if he is going to complete the Metaphysics of Nature and Metaphysics of Morals. There is almost a sense of intellectual urgency. The Critique is supposed to establish the principles; now he wants to demonstrate what can be built from them. In reality, his subsequent work would take a somewhat different shape than this plan suggests. The Critique of Practical Reason appeared in 1788, the Metaphysical Foundations of Natural Science in 1786 had already appeared before the second edition, and the Metaphysics of Morals would not appear until 1797. But the architectural ambition is already present here: Critique first, systematic metaphysics afterward.

Then Kant gives us one of the most important metaphors in the entire Preface: “A philosophical system cannot come forward armed at all points like a mathematical treatise.” He recognizes that philosophy cannot attain the same kind of immediate formal transparency that mathematics can. This does not mean philosophy is merely subjective. Rather, the objects of philosophical investigation are so interconnected that local difficulties may be resolved only by understanding the system as a whole. This is why he describes pure reason as an “organic structure.” The German expression is “ein organisiertes Ganze.” An organized whole is different from an aggregate. In an aggregate, you can remove one piece without necessarily changing the others. In an organism, the parts exist through their relations to the whole. Change one organ and the function of the others may be altered. Kant thinks pure reason is like this.

This organic metaphor is extremely important for the later history of philosophy. Hegel will take this idea of philosophical systematicity much further. Hegel will insist that philosophical truth cannot be adequately expressed as a collection of isolated propositions because the truth of a determination depends upon its place within the movement of the whole. That is not yet Hegel, of course, but the soil is already here. Kant is already saying: don’t tear individual determinations out of their systematic relations and then complain that they contradict one another. The whole governs the interpretation of the parts.

The next sentence makes this explicit: “By confining the view to particular passages, taking these out of their connection and comparing them with one another, it is easy to pick out apparent contradictions.” Notice the word “apparent.” Kant is not claiming that every contradiction someone identifies is automatically meaningless. He is saying that apparent contradiction can arise because the reader has failed to understand the function and level of a particular proposition. This will become especially important in the Critique because Kant frequently makes distinctions that sound contradictory until we understand that he is speaking from two different standpoints. The most famous example is freedom and causality. Human action can be causally determined as an appearance while the rational agent can nevertheless be thought as free in another sense. Without the architecture of the phenomenon/noumenon distinction, Kant’s statements can look contradictory. Within the system, he thinks they are reconciled.

There is a very subtle methodological principle here: philosophical interpretation requires what we might call a “view of the whole.” Kant uses the German expression “die Idee des Ganzen,” the idea of the whole. The individual propositions are not self-interpreting. Their meaning depends partly upon their systematic location. This is why our method with Kant should be cumulative. When we encounter a difficult proposition later, we should not treat it as though we are encountering Kant for the first time. We will ask: what distinctions have already been established? What problem is this proposition responding to? What role does it play in the architecture? What would break if we removed it?

His final remarks about “action and reaction” are also philosophically suggestive. A theory with internal stability can survive criticism. Indeed, criticism can improve it by exposing superficial roughnesses. Kant imagines philosophical development almost organically again: objections, responses, revisions, and reinterpretations gradually smooth the system’s surface. But the underlying structure remains. This is precisely what Kant hopes the second edition accomplishes. He is not retracting the fundamental argument of the first edition; he is trying to make its articulation more robust.

There is, however, an interesting historical irony here. Kant says that he has “found nothing to alter” in the propositions and demonstrations themselves, yet the B-edition changes some of the most important parts of the work. This is particularly significant because the A/B distinction has generated enormous scholarly debate. Why did Kant radically rewrite the Transcendental Deduction if the propositions themselves remained unchanged? One possibility is that Kant genuinely believed he was preserving the same argument while making its exposition more precise. Another is that the criticisms of the first edition forced him to reconsider how his argument should be framed. We should not resolve that question prematurely. When we eventually read the A and B versions of the Deduction, we can watch Kant’s thought under pressure. That will be far more illuminating than simply being told that “Kant revised the Deduction.”

There is another historical layer worth noting. Kant’s reference to the “licence in thinking” is partly a response to the intellectual climate of late eighteenth-century Germany. He is defending disciplined philosophical reasoning against what he sees as fashionable claims to genius that dispense with rigorous argument. This is one reason Kant’s admiration for Christian Wolff is important. Wolff’s philosophy was systematic to an almost obsessive degree: define concepts, establish principles, prove consequences. Kant eventually rejects the dogmatic metaphysical content of the Wolffian system, but he retains something of its methodological ideal. His quarrel with Wolff is not simply that Wolff loved systematic philosophy. Kant wants systematic philosophy too. His objection is that Wolff constructed the system before critically examining the instrument by which the system was constructed: pure reason itself.

This is one of the deepest points in the Preface. Kant is not opposed to “dogmatic procedure” as such. He explicitly says that “pure cognition must always be dogmatic” in the legitimate sense: it must proceed through strict demonstration from secure a priori principles. What he opposes is “dogmatism”: using reason’s principles without first examining the right reason has to use them. This distinction will become essential later. Kant is therefore not the philosopher who says, “Never be certain.” He is almost the opposite. He wants genuine certainty. But he thinks certainty requires first determining the legitimate conditions under which certainty is possible. Skepticism says, “We cannot know.” Dogmatism says, “We know.” Critique asks, “What does it mean to know, and under what conditions is such a claim legitimate?”

That distinction also explains why Kant’s system is not simply a compromise between rationalism and empiricism. He is trying to transform the question itself. Rationalists had asked: what can pure reason know independently of experience? Empiricists had asked: what can experience give us? Kant asks: what are the a priori conditions that make experience itself possible, and what is the legitimate scope of cognition independent of experience? This is the beginning of transcendental philosophy.

And now, because this is our first real reading of Kant, I would make one rule for our work together: whenever Kant makes an apparently paradoxical statement, we should not immediately resolve it with a simplified formula. We should reconstruct the distinction that makes the statement possible. Kant is constantly working through distinctions: a priori/a posteriori, analytic/synthetic, intuition/concept, sensibility/understanding, understanding/reason, phenomenon/thing-in-itself, constitutive/regulative, possibility/reality, thinking/cognizing, speculative/practical. His philosophy lives in the precision of these distinctions. If we collapse them too quickly, Kant becomes either impossibly obscure or deceptively simple.

So the final page of the Preface leaves us with a methodological instruction that is almost as important as anything in the Critique itself: learn the idea of the whole. Kant wants us to see the Critique as an organized structure in which the parts mutually determine one another. And that is exactly how I think we should proceed. We should not rush through the text merely to “get the gist.” We should take Kant slowly, identify the German term when the English translation conceals something, locate the philosophical problem being answered, identify the historical interlocutors behind it, and then keep track of where the argument is taking us. Aristotle and Leibniz will appear when Kant is silently transforming their questions; Locke and Hume will appear when Kant responds to empiricism; Newton will appear when Kant explains scientific necessity; and later, once the architecture is genuinely ours, we can watch Hegel, Husserl, Heidegger, and Derrida begin their own arguments with Kant.

The Preface has now essentially handed us the map. The next thing is not to interpret Kant through those later philosophers. It is to enter Kant’s own house. The doorway is the Transcendental Aesthetic, and its first question is deceptively simple: what is sensibility, and what does it mean for something to be “given” to us at all?

“KÖNIGSBERG, April 1787” is more philosophically significant than it first appears. It is the small temporal and geographical signature at the end of an enormous attempt to determine the limits of human reason. Kant has spent the Preface insisting that philosophy must become systematic, that reason must examine its own powers before making claims about what lies beyond experience, and that the parts of philosophy cannot be understood adequately when detached from the whole. Then, almost abruptly, he places the work back into the world: Königsberg, April 1787. The signature reminds us that this supposedly “pure” investigation of reason has nevertheless been undertaken by a finite human being, at a particular place and moment, within a particular historical situation.

There is also an important tension here. Kant’s project seeks the a priori: what is necessary independently of particular experience. Yet the Critique itself has a history. There is a first edition in 1781, a second edition in 1787, revisions, controversies, misunderstandings, and Kant’s own aging body and finite amount of time. The universal system emerges through a temporal process. “Königsberg, April 1787” therefore quietly embodies a problem that Kant’s philosophy repeatedly confronts: how can something necessary and universal be discovered by beings who are finite, situated, historical, and temporal? The transcendental conditions of cognition may not themselves be historical in the ordinary sense, but our discovery and articulation of them certainly are.

And this is why the final location matters. Kant is not claiming to speak from nowhere. He is not a disembodied Reason dictating metaphysics from outside history. He is saying, in effect: this investigation has been conducted here, by me, at this point in time, and yet what it seeks to uncover is not merely “my” philosophy. The signature marks the boundary between the empirical author and the transcendental claim. Kant the man is in Königsberg in April 1787; the reason he is investigating claims a validity that cannot be reduced to Königsberg, April 1787. The philosophical drama of the Critique is contained in that little gap between the situated thinker and the universal conditions he discovers.

SANTA ANA, August 2026