Between Intellectual Optimism and Pessimism: B. Kant’s Solution (Column 495)

With God’s help

Disclaimer: This post was translated from Hebrew using AI (ChatGPT 5 Thinking), so there may be inaccuracies or nuances lost. If something seems unclear, please refer to the Hebrew original or contact us for clarification.

In the previous column I described what I call “the epistemic rupture.” In brief, I explained that we have no justification for our most fundamental modes of thought and cognition regarding the world. I noted that despite how fundamental this is, throughout the history of philosophy no serious answer has been offered to this problem. I will now discuss the solution proposed by Kant, who, as noted, was the only one in history who at least tried to propose a systematic solution that seemingly holds water.

Kant’s Solution

We saw in the previous column that Kant offered a renewed and general formulation of the problems Hume raised against empiricism, gathering them all under the problem of the synthetic a priori. Kant then proceeded to solve them. He did this by what he himself called the “second Copernican revolution.” Copernicus reversed our point of view about the universe and argued that it is not the earth that stands at the center with everything revolving around it, but the sun at the center with everything revolving around it,[1] and by inverting the point of view he solved many problems and complications in the existing cosmological description. Kant suggested that this is how we should proceed with respect to the epistemic rupture as well.

The prevailing picture for us is that facts exist in the objective world itself, and we, as scientists, stand before them and try to know them. In this pre-Kantian picture, there are two images that stand opposite one another: the world as it is in itself, which science deals with; and opposite it stands science, which consists of our claims about the world. Science is in our cognition, and it tries to hit upon what occurs in the world itself. Kant’s and Hume’s questions deal with the fit between these two planes: our knowledge and cognition (= science) versus the world as it is in itself (= the facts). But Kant claims that this picture is mistaken and that it is precisely what leads us to the epistemic rupture. His claim is that the facts and events with which science deals are not located in the world-in-itself but are found only within us too—this is his “second Copernican revolution.”

Science does not concern itself with the world as it is in itself, but with its images that reside in our cognition. Every phenomenon we try to understand is a phenomenon we have experienced in our own cognition, and as such it should not be seen as a bare fact from the objective world. These images in our cognition are constructed in a way that necessarily depends on our cognitive faculties. We shape and process them and insert them into our conceptual and sensory-cognitive frameworks, and therefore it is clear that they contain components that are not objective. Thus, the “world” we engage with is not the world of things-in-themselves (the noumenon) but the world of things as they appear to us (the phenomenon). Epistemology, then, does not concern itself with the fit between our thinking and the world-in-itself, but between an image that is in our cognition and our cognition and thinking themselves. This is no longer such a wondrous fit, since it is a fit between two things that are bound up with human cognition, and not between two independent images. The epistemic rupture stems from our assuming a fit between two independent images. But in light of his Copernican revolution, Kant claims, it is no wonder that we can determine things about the “world” by thought alone (i.e., that the laws of nature are a priori).

To complete the picture, I will add that Kant explains that the way we arrive from observations to synthetic-a-priori claims is by means of what he called “transcendental arguments.” A transcendental argument in effect derives, a priori, various constraints on the image of the world that will be experienced, and it does so from reflecting on the form of our thinking and our conceptual world. We saw that our conceptual world and the principles of our thinking dictate very basic elements within the image of the world we will see, and therefore it is no wonder that we can know them even before we have conducted observations. The details we of course learn from observation, but there is no need for observations of the world to know the constraints of our own perception. For example, a person with red cellophane over his eyes can know a priori that every object he sees will be red. He need not conduct observations to determine this, since it is the result of the influence of the cognitive system and not of facts in the objective world as it is in itself. This is the reason we can know certain things about the world a priori. Thus, for instance, space and time are categories that are found only within us, yet our entire scientific image of the world takes place within the conceptual framework of space and time. The facts that science explains are already tinted by the colors of space-time, and therefore it is no wonder that the laws of science speak about certain dependencies of physics on space and time, for that is how our cognition is built. So too with causality. Our cognition grasps things through the dependence of cause and effect, and it is no wonder that one can determine a priori that every event has a cause. This is not a claim about reality as it is in itself (the noumenon), but our cognition imposing that principle upon reality as it is apprehended by us (the phenomenon).

Problems with the Kantian Solution

I have written more than once that the Kantian solution is indeed brilliant, but on closer inspection it is clear that it does not satisfy. Let us take as an example Newton’s second law, which states a direct proportionality between force and acceleration (force equals the mass of a body times the acceleration it produces). According to Kant, this law is a result of the constraints of our cognition and not a property of the world as it is in itself. Let us suppose, for the sake of discussion, that in reality itself a body’s acceleration is not proportional to the force (as the second law states) but to the square of the force. Why, Kant claims, are we unable to notice this? Is there any reason our cognition cannot discern the motion of such a body? In other words, what actually causes us always, in every event we observe, to see a direct proportionality between force and acceleration? Let us be more concrete. Suppose there is a body whose mass is 2 kg, and a force of 6 newtons acts upon it. According to Newton’s second law, the acceleration of such a body would be 3 m/s² (force divided by mass). Are there in reality cases where the acceleration is different and we simply do not notice them? Why should we not notice them? If Kant wishes to claim that we only notice bodies that move in accordance with Newton’s second law, he must explain what prevents us from seeing that very acceleration of that very body in cases where a different force acts upon it. After all, we could notice another body moving with such acceleration, or even the same body upon which a different force acts that produces such acceleration. Does a force of a different magnitude simply block our eyes, and thus we miss all the events in which the second law does not obtain? This is a bizarre assumption. The picture Kant proposed has no better justification than the picture it came to replace. It is no less problematic than the ordinary picture (the one before the second Copernican revolution).

So far I have dealt with Newton’s second law. What about Newton’s law of gravitation? In the world itself (not in our cognition), are there massive bodies that stand suspended in the air and do not fall, or that fall to the ground with an acceleration different from the one familiar to us? If so, what prevents us from seeing this? So too with all the other laws of nature. The cognitive constraints of which Kant spoke surely exist (we cannot see wavelengths outside the visible range, or hear sounds beyond our hearing range), but I see no logic in the assumption that the constraints of our thinking are necessarily translated into constraints of our cognition (that what we cannot think we also cannot see). The fact that our thinking reaches certain conclusions about the laws of nature (using synthetic-a-priori principles like causality and induction) cannot be explained by cognitive constraints, since our cognition is not limited by the same limitations as our thinking. It is therefore very odd to accept the Kantian thesis as an explanation for the synthetic-a-priori problem. The idea is brilliant, but it seems it does not hold water.

Until now I have dealt with the laws of nature, but similar claims can be raised regarding meta-scientific principles. If the law of induction is not true, then in the world itself there are events that happen at certain times and will not recur at other times or places. So why do we not see them? If there is a law we have observed consistently and suddenly it ceases to hold, we look for explanations for this—and usually find them. It does not happen that a mode of behavior that held until now simply stops being so for no reason. The same holds for the principle of causality. Somehow we always find a cause for what happens. Is this merely coincidental—perhaps because anything that occurs without a cause simply does not reach our cognition? What prevents us from seeing it?

The truth must be told: specifically with respect to meta-scientific principles like causality and induction, one could perhaps answer that this is a constraint of thought and not of cognition. We assume that there is always a cause, and so we always search for and find something on which we can hang matters. The same holds for induction (when it does not hold we ascribe it to some cause, because we assume it must hold). I think if you reflect a bit you will see that even this is very implausible, but for our purposes here the difficulties arising from the laws of nature suffice. The difficulties regarding the meta-scientific principles are merely an additional, correct but not necessary, reinforcement of my argument.

It is important to note that even if Kant’s proposal does not truly provide a reasonable answer to the epistemic rupture, it is still clear that he is right in asserting that there is a gap between the world as it is in itself (the noumenon) and the world as it appears in our cognition (the phenomenon). It is entirely clear that science deals with the phenomenal and not with the noumenal, for the facts it explains are the facts in our cognition. There is no doubt that the facts in our cognition are colored by hues that cognition dictates to them. This is certainly true, yet we still must seek a reasonable solution to the epistemic rupture.

Lev Shestov’s Difficulties

Let us now return to our acquaintance Lev Shestov. On p. 176 of the book, Zeitlin reports that the various interpretations of Kant’s doctrine offered solutions to many of its difficulties, but in his view there are two difficulties for which you will not find a solution in any interpretation (he surveys the various interpretations and shows that they do not answer them).

The first difficulty concerns the relation between the noumenon and the phenomenon. Which begets which? How can we even speak about the noumenon if it is not accessible to our cognition? How can we even know that it exists? The common assumption among Kant’s interpreters is that the noumenon is the cause of the phenomenon, i.e., the thing-in-itself gives rise to its image that appears in our cognition. But to this Solomon Maimon’s objection presses: causality as such belongs entirely to the phenomenal world, and it cannot be applied to the very genesis of that world. It is like saying that the principle of causality exists because of the principle of causality. In other words, in the noumenal world there are no causal relations, for it does not obey the rules and laws that are within our cognition. Therefore one cannot say that the noumenon is a cause of the phenomenon. The objective world simply does not belong to the conceptual framework of phenomena and cannot be described in its terms.

The second difficulty is: whence were the fundamental principles of cognition and thought imprinted in us—the ones that generate the transcendental framework from which we derive synthetic-a-priori claims? If they were imprinted in us by the phenomenon, then it is hard to say that the phenomenon is constituted by them, for they are the ones that constitute it. Kant explained that we attend to the world and create the phenomenon because of our cognitive faculties. But it now turns out that they themselves are produced by the phenomenon—that is, they are its products.

Zeitlin then goes through Kant’s major interpreters and shows that none of them solved these difficulties; I will spare you that here. I will now argue that these two difficulties arise from a common misunderstanding regarding the Kantian distinction between phenomenon and noumenon, and that in fact they have a very simple solution.

Correcting a Mistake Regarding the Kantian Distinction[2]

It is commonly thought that Kant claims we are unable to grasp the world of things-in-themselves because we are trapped within our cognitive faculties. This claim assumes that the Kantian distinction reflects a human limitation. Because of our limitations we cannot grasp the world as it is in itself. God, for example, is exempt from this, for as an omnipotent being He has no limitations, and therefore for Him there should be no problem in grasping the thing-in-itself. But to my mind this is a very basic mistake in understanding the matter.

Our inability to say things about the noumenon is not the result of our limitation. More than that, it is not even correct to say that we cannot grasp the noumenon. We certainly can—and the result of that grasp is precisely the phenomenon. Think of a red table standing before you. The common interpretation is that the table as it is in itself does not necessarily have the color red. The red is in our cognition, while the table itself perhaps has a different color or no color at all. But that is not right. The table has properties (such as its crystalline structure) that produce in whoever observes it a red appearance. The red color is a description of the table’s crystalline structure (or the dispersion of light from its crystalline structure) in our cognitive-visual language. Someone else who observes the table, but whose eyes are wired to the auditory center of the brain, will hear the table rather than see it. Another might see it in green, or sense its crystalline structure as a taste or as a sense unfamiliar to us.

In other words, one could say that the table as such has no color at all. But that is not because we cannot grasp its color; rather, by definition color is the image that arises in our visual cognition when we attend to the table. Color does not exist in the world as such but only in our cognition. Consider the famous question: when a tree falls in a forest and there is no one there to listen, does it make a sound? The answer is, of course, negative. Such a tree makes no sound. It displaces air and creates in it an acoustic pressure wave. But “sound” arises only when such a pressure wave strikes the eardrum of some ear. In that case a sensation of sound is created in the cognition of the bearer of the ear. In the world as such there are no sounds—there are acoustic waves. Sound, like color, does not exist in the world itself but only in our cognition. The tree does not emit sound; it produces a physical phenomenon that our ear presents as sound (another creature might present it as music, or as some other sense).

Therefore there is no meaning to speaking of the color of the table-in-itself or of sound in the world-as-it-is-in-itself. The table-in-itself has properties such as a crystalline structure and the like, but in our cognition those properties are expressed within our system of concepts and forms of apprehension (color, sound, shapes, etc.). By definition, perception relates to the thing-in-itself. But to perceive a thing means to insert it into our conceptual and cognitive framework. That is the meaning of apprehending the noumenon. Therefore we certainly can and do apprehend the table-in-itself. The apprehension of the table means creating a representation of it in our cognition. There is no perception without a perceiver, and thus by definition the result of perception is an image in the perceiver’s cognition; that image is shaped by his system of concepts and forms of apprehension.

If so, talk of a limitation that prevents us from apprehending the thing-in-itself is a misunderstanding. There is no such limitation, and we certainly apprehend the thing-in-itself. The product of that apprehension is the phenomenon that arises in our cognition. This is not a limitation, for every apprehension is defined by the faculties of the perceiver and is conducted within their framework. Thus every creature apprehends things outside itself. In apprehension we translate the thing or event we observe into our internal language. Were I not timid, I would say that in principle this is true even with respect to God Himself.

In any case, we can now see that in light of the correction I have proposed to the meaning of Kant’s distinction, Shestov’s two difficulties vanish of themselves.

Solving Shestov’s First Difficulty

We have seen that the phenomenon is a property of the thing-in-itself. Think of a property of Reuven, for example that he is kind-hearted or tall. Is it correct to say that Reuven is the cause of his kindness or of his height? These are his properties, and the relation between a thing and its properties is not causal. A car is made of metal and appears elegant to me. Is the car the cause of its being elegant or the cause of its being metal? Certainly not. The thing does not cause its properties; rather, it is the bearer of those properties, which characterize it. Now consider, in light of this, whether it is correct to say that encountering Reuven’s kindness or his tallness still leaves a question mark regarding the existence of the object as it is in itself. Can one ask us, “Whence do you know that Reuven really exists—after all, you only encountered his properties?” That is absurd. If we encountered the properties, then evidently there is something or someone who bears those properties.

It is important to understand that this conclusion is not connected in any way to the principle of causality. I know that Reuven exists not because he is the cause of his properties (he is not), but simply because I saw him. What I actually saw were his height, width, skin color—or, more abstractly, I encountered his kindness. From all this I can say without any problem that I encountered Reuven himself; that is, infer that there exists someone whose properties these are. The principle of causality is not involved here in any way. Seeing Reuven created in my cognition an image that represents him, but that image is the apprehension of Reuven-as-he-is-in-himself. Thus a host of problems included in Shestov’s first difficulty quietly dissolves.

A Different Formulation

In light of what I have said here, one can add and argue that even if I adopt the assumption that the noumenon is the cause of the phenomenon, I still see no difficulty in applying the principle of causality to the noumenon. The fact that I use the principle of causality is because of my cognitive faculties—but once these are my faculties, why should I not use them also with respect to the noumenon?! For example, my eyesight was produced by evolution. Does that mean I cannot use my eyes when I study evolution? Not to mention using my very thinking faculties to study the modes of my own thinking. That is exactly what many researchers in different fields do (logicians, psychologists, biologists, etc.), and I see no problem in it. The circularity here is only apparent. Not every circularity yields a paradox—see on this in columns 157 – 158 and in more detail in the lecture series on self-reference.

In columns 99, 164, and 465 I offered a similar example regarding the use of the time axis with respect to the past. There I cited the claim of Rabbi Shem-Tov Gefen, who explains that if we adopt the Kantian view that time is a subjective category of the human being (and does not exist in the world itself), then the problem of the age of the world (which is not ~6,000 years, as tradition says, but about 14 billion years) is solved at once. His claim was that time is subjective and therefore it was born with the human being, whence it is clear that the age of the world is that of the human being (the homo temporalicus). I explained there that R. Shem-Tov Gefen errs here, for even if time is a subjective category that serves us, there is no impediment to using it with respect to the past—both for times that preceded my own existence (my grandfather’s birth) and even for times that preceded all of humanity (the creation of the world). Not every application of a tool to stages logically prior to it is a vicious logical loop.

Accordingly, to the same extent I can use the principle of causality and all my rational assumptions to analyze and make claims about the noumenon. It is a subject like any other, and I treat it with my existing tools of reasoning and analysis, which, by definition, belong to the phenomenal realm.

Solving Shestov’s Second Difficulty

Shestov’s second difficulty concerned the question of who created within me the templates that shape the phenomenal image in my cognition. Here too he errs in the same way. These templates are only a language by which I express reality as it is in itself. They do not create a separate world that stands opposite and contrary to the world-as-it-is-in-itself; rather, they mirror and translate the world-as-it-is-in-itself into my personal language (or, in fact, the inter-personal language of human beings). Therefore it does not matter at all how this language was created—just as there is little point in discussing who created the Dutch language or the Zimbabwean language. So long as it is a language that can be spoken, there is no impediment to using it to describe anything, including the mechanisms that created it itself.

In other words, it is not correct to say that the phenomenon imprinted cognitive and conceptual categories within us. There is no such thing as “the phenomenon” at all, and therefore it cannot imprint anything in us or do anything else to us. The phenomenon is a reflection of the noumenon in our cognition, but what actually exists are only the things-in-themselves, and thus only they can act and bring about consequences. My cognitive faculties were produced by evolutionary mechanisms, or by the Creator, or by a demon in the shape of a monkey. None of this matters. Once they were produced, this is the language I use, and there is no impediment to using it to describe the noumenon and thereby create the phenomenal image. In short, this set of tools is what creates the phenomenon, not the other way around—and thus Shestov’s second difficulty is resolved.

Conclusions and Interim Summary

Thus Shestov’s difficulties ultimately fall as well. The Kantian distinction between phenomenon and noumenon is entirely reasonable, and likely correct. There is no inherent difficulty in it, so long as one understands it correctly. However, as I noted above, even if it is correct, it does not solve the epistemic rupture. We saw that it does not offer an answer to Hume’s difficulties (i.e., to the problem of the synthetic a priori, in Kant’s terminology). Thus we are still left facing a broken trough. Philosophy, with all the variety of its schools, and all the proposals raised over history, fails to provide a sufficient justification for our use of the tools of thought and cognition when we seek to know the world (for example, when we do science)—and this includes Kant’s heroic attempt. It turns out that we have information about the world that is not the product of observation, and the empiricist alternative collapses. On the other hand, the critique of rationalism seems very reasonable. It is implausible to build knowledge about the world solely from my intuitions. My thinking consists of structures embedded within me, and there is no reason to assume they reflect what happens in the world as it is in itself.

These are the epistemic ruins that accompany Lev Shestov. The difficulties he raised do not destroy the rational world, but he is right that this world indeed does not hold water. Among the graves of these philosophical and scientific ideas wander Zeitlin and Shestov, seeking (and perhaps also finding) God. I will address this point in the next column.

A Closing Note on the Nature of Philosophy

To conclude, I will note a common view about philosophy. People feel there is no point in engaging in philosophy because it contains so many schools and opinions, as numerous as the sand on the seashore, and we have no way to decide between them (for observation cannot decide a philosophical question—otherwise it would be science, not philosophy). From this, many people—especially scientists—conclude that philosophy cannot advance us anywhere.

The epistemic rupture is a good counterexample to this view. Here is a question to which many answers were proposed in the history of philosophy—yet it turns out that none of them hold water. They are not empirically refuted, since this is philosophy and not science, but the very thinking that produced them is what overturns them. They are a priori (i.e., philosophical) refutations. Thus, anyone who proposes a conclusion that rises from these ruins can indeed claim that philosophy obligates us to adopt that conclusion. In other words, one can certainly learn from philosophy, and a multiplicity of opinions does not always indicate a multiplicity of truths. Sometimes it indicates a confusion rich in errors, and sometimes it consists of different formulations of the same answer. There are questions in philosophy that can certainly lead us to concrete conclusions. In my series of columns on philosophy (155 – 160) I discussed this point at length (indeed, the whole series is devoted to it).

In the next column I will walk from the ruins of the epistemic rupture toward God, and we will return to the difference between Zeitlin’s intellectual pessimism and the intellectual optimism that I propose.

[1] This is a common but incorrect description. Copernicus did not discover any new scientific fact about the world, and the views that preceded him are not wrong. Copernicus merely proposed a coordinate system more efficient and convenient than its predecessor. I discussed this in my article here, in column 481, in the Q&A here, and elsewhere.

[2] I should note that my goal here is not to reconcile Kant’s doctrine. Even if the interpreters are right in the claim I will bring here, they are not right in terms of the truth. Therefore, the following should be read not as an interpretation of Kant but as my philosophical claim regarding the Kantian distinction between phenomenon and noumenon.