Epitaph for a Historian of Philosophy
The frightening thing about the study of the history of philosophy is that, even on the day the universe perishes, it will still be impossible to fully study Kant, because there is no 'Kant-universe' out there — there is only the universe out there.
Some time ago I happened to open a paper I wrote during my doctoral studies (as usual, I submitted it once and never resubmitted). In that paper I discussed the deep structure of a rule. I found that when people live within rules, they grasp a predicative relation (p is q), whereas when they live above rules they grasp an entailment relation (if p, then q). The difference between the two is not whether people actually act according to rules, but whether they act according to rules with an explicit understanding.
I tried to use this distinction to pose a challenge to an important insight of H.L.A. Hart.
(1) To explain that people have a social rule rather than merely a social custom, Hart proposed the concept of the internal point of view: a person who has a social rule not only displays behavioral conformity, but accepts the pattern exhibited by the behavioral conformity as an explicit and public standard, as a guide to their own actions and a basis for criticizing others.
(2) The essence of the internal point of view is the practitioner’s reflective-critical attitude toward their practice. Only through reflective practice can people do more than merely live in a custom; they can possess a rule.
(3) The internal point of view concept has long been used by some Hartian followers to support Hart himself and certain legal positivists in addressing normative questions.
(4) Hart’s contrast between custom and rule feels commonsensical: we sense that the practitioner’s reflective-critical attitude toward their practice is crucial to obtaining a normative stance toward practice.
(5) However, my distinction led me to notice that Hart’s insight may be too easily understood as: the more reflection, the more normative the attitude. Hart did not carefully distinguish degrees of reflection and the two opposite kinds of experience that different degrees may lead to: under insufficient reflection, people may acquire devoted, affirmative, unconditional, direct experiences; after sufficiently reflective critique, they may instead acquire detached, hesitant, conditional, indirect experiences.
The degree of reflective critique of practice and the commitment to practice need not be linearly related; certain amounts of reflective critique may in fact undermine commitment to practice. The next question is, of course, what is the relation between a normative attitude toward practice and a committed attitude toward practice. That is a relatively independent question; I currently lean toward thinking that, at least in Hart’s context, the reflective-critical attitude used to explain normativity is essentially a matter of commitment to practice.
In my paper I gave an example Hart himself used. When unaware of the Paris standard metre, I might hold the belief “this thing is 2 meters long” — a rich, thick, unconditional belief about the object’s quality. But once I learn about the Paris standard metre, I realize “if the Paris standard metre is 1 meter, then this thing is 2 meters,” and I lose that earlier quality of belief about the object’s quality. In the latter case I will still, and more clearly, say “this thing is 2 meters,” just as before, but my belief state has changed.
There are similar situations in moral life. Living within rules means acting directly and devotedly, thinking that this is what I ought to do, that it is the right thing, or quite simply doing it. But after fully reflective critique, when rules are made explicit and publicly discussed and enter my practical consciousness, I lose the initial devoted, committed, unconditional belief that “that is what I ought to do, that is the right thing.” In such cases I will still say “that is what I ought to do, that is the right thing,” but the belief state has changed.
This resembles Bernard Williams’ criticism of Kant via thick and thin concepts. Having a thick-concept state means living within rules, grasping a predicative relation, direct and sincere; having a thin-concept state means that after full reflective critique one acts according to explicit, public rules, becoming indirect, conditional, and insincere. I see a person falling into water who is my wife. If I consider “according to a universalizable rule I ought to save my wife in this situation” as my reason to rescue her, then my morality might be called into question. Here, the moral action would rather be to rescue without deliberation, directly, not to rescue according to a rule. Understanding how to act according to rules may make my action appear morally problematic.
There are examples contrasting people living within rules and living above rules. If a close relative dies, people will typically cry. Observers can capture a regularity: “If a close relative dies, then one cries,” and this can form a social habit. The society may reflect and formulate a rule: “If a close relative dies, then one ought to cry.” A person’s practical reasoning following that rule will be: “If a close relative dies, then one ought to cry; my close relative died, therefore I ought to cry.” Yet some sincere people will be unwilling to cry as soon as they think this way. Others in the society, seeing that one does not cry at a parent’s grave, will remind or require them to cry, even criticize them for not crying. A sincere person will resist strongly. In this practical scenario, explicitly acting according to a moral rule makes the action appear immoral.
More interestingly, once the regularity “if a close relative dies, then one cries” is captured and reflected into a rule “if a close relative dies, then one ought to cry,” we may no longer tell whether a person sobbing before the corpse is truly crying or merely performing. In some rural Chinese funerals, visitors follow a family sequence; when a new member approaches the bier and kneels, the daughter-in-law will wail for a while and then stop. I once witnessed them laughing and talking during a pause, eyes swollen, switching fluidly between grief and conversation. I heard people talk about which daughter-in-law cried hardest, even “fainted from crying,” to judge whose husband is the most filial. It goes without saying hiring people to wail at graves to display filial piety.
Consider why a person might be shy. In my framework, a shy person is overly sincere and resists reflection, resists revealing motives, conditions, reasons, or purposes of their actions. They are used to doing the right thing directly, “acting from feeling and stopping at ritual” (meaning action stops at following the ritual). When they feel gratitude they may even be embarrassed to say “thank you,” because thanking is an obviously purposive act. When thanking requires giving a gift, this psychological resistance is even clearer: they are uncomfortable with acting according to a rule, because it makes their action conditional rather than arising naturally from within. A person who is no longer shy finally accepts that an action is rarely wholly from within but driven by reasons, motives, and purposes external to the self. Most actions are not purely self-originated but conform to rules or standards independent of the agent. Understanding how to act according to rules can even make actions feel less like one’s own.
Looking back, what might Kant’s problem be? He may think moral life depends on explicit, universal, and public reflection on rules, and that the clearer, more universal, and more public the rule the more moral the associated thought or action. Thus, after dispensing with God, he speaks of the rational subject’s self-legislation. He grounds morality in law, yet he uses legal language — “legislation,” “court,” “judgment,” “command,” — when discussing morality. His so-called ultimate autonomy of reason is merely obedience to a moral miracle of legislating for oneself. Why does Kant resort to law to describe morality? Or why must the highest legitimacy or normativity be handled with legal concepts? Perhaps the legal concept suits the practical notion he admires: acting on the basis of an explicit, public, and clear grasp of rules.
These questions may pose a challenge to some of Kant’s demands in moral philosophy. I took my question to a friend in the philosophy department who specializes in Kant (we had in fact not had an academic discussion for about two years). I told him my concerns, Williams’ criticism, and my understanding of some Kantian positions. His reaction was characteristically calm—he is someone who thinks more than he talks.
At the end of the discussion he casually said, “Think about it some more.” I was a bit agitated and retorted, “Don’t you think about it yourself?” That was a little offensive, I admit, because it was not his problem. But if I were him, I would probably feel the idea opened a new perspective on Kant and could not be unrelated to my existing interest in Kant; if it were sound it would be unlikely that nobody had discussed it before (for example, Williams). Those related discussions might already have been on his radar. That is what really upset me.
Afterwards I continued thinking and discussed with Claude; I summarized four possible Kantian defenses:
In the Paris standard-metre example, I did not say the later case made my belief worse; though changed, it is a change from a correct belief to a more correct belief, because I am clearer about why the object is 2 meters long.
After sufficient reflection and acquiring rules, insincerity need not follow; I can be clearer than before about why I act, so there need not be a dilemma between being clear and being sincere versus being sincere but blind.
Even if there were a dilemma that clarity undermines sincerity, from a practical standpoint there is no loss: clarity replaces certain sincerity and clarity has higher value than sincerity; conditional belief can be superior to unconditional belief.
Related to (1): upon learning about the Paris standard metre, my belief did not lose sincerity but transformed from one sincere belief to another sincere belief; if the first was a type of authenticity, the second is a different, perhaps higher-quality, authenticity.
I asked Claude whether these Kantian defenses would hold up, and we discussed further. Finally I asked whether these discussions constituted a genuine, research-worthy problem. Why was my friend so indifferent? Claude understood my feelings: the issue was not that he disagreed with my points but that he dismissed my concerns. I shared the conversation with him and received no further reply.
This made me reflect on scholarly styles. I have known him for some time. I once told him that our academic styles are opposite: he is cautious and meticulous, I am imaginative and free-ranging. Over the years my thinking has travelled widely while his has advanced very slowly.
Because we both use Gemini I had the chance to see his conversations with Gemini (he also has my account). I casually looked and found he recently asked Gemini to help revise a paper he wrote about five years ago — a paper I had seen before. He also asked Gemini about how Kant distinguishes the beautiful from the sublime, and so on. His queries were few, unlike mine which are step-by-step, heuristic conversations with AI used as an inspiration partner (who else do I have for conversation?). He is mostly devoted to understanding particular aspects of Kant.
I reflected carefully on our differences:
(1) Regarding subject matter, I study problems — I want to clarify whether A is B; he studies beliefs — he wants to clarify whether x believes that A is B.
(2) Regarding scholarly aim, my study may involve Kant, e.g. whether Kant thinks C is D. I might study that but ultimately return to my own questions. My interest is not merely to understand what someone believed but to see what I truly think. By contrast, when he hears about a problem I raise, he immediately seeks to position it within Kant scholarship: if it has a place he may be interested and pursue it as a Kantian question, and he will not return to the original problem; if it has no place, he judges it not a proper/specialized Kantian problem.
(3) Regarding scholarly fate, I imagine the epitaph on his tombstone might read:
XXX — died in year aaaa — at 70 finally figured out how Kant distinguishes the sublime from the beautiful.
My epitaph might read:
YYYY — died in year bbbb — at 70 finally figured out the thirteen ways to write the character “回”.
Maybe my questions are not as deep or serious as his, but I face problems, not other people’s beliefs about problems. I am responsible only for my own beliefs, not for others'.
Kant has a conceptual system that resembles a holographic projection of a universe. It contains many correct projections but also misplacements, absences, and overlaps. To study a pixel in Kant’s projection you must study neighboring pixels. Kant’s system sits among several tightly interrelated philosophical strata — secondary sources, tertiary, quaternary texts, and philosophers before and after him: Fichte, Schelling, Schopenhauer, Hegel, etc. Studying Kant thus becomes an endless task.
The endlessness of research on Kant differs from the endlessness of cosmic research. If we study the universe and fail after a million years, we can try for a trillion years and, in principle, the universe could eventually be fully understood unless it ends first. The most frightening thing about the history of philosophy is that, even on the day the universe perishes, it will still be impossible to ever fully comprehend Kant. There is no “Kant-universe” sitting there. Only one universe exists. Moreover, interpretation of Kant is a creative act; any interpretation will always face objection. Kant is dead and cannot tell us which interpretation is correct. (What if Kant were alive? I recall the anecdote of Ramsey and Wittgenstein, where Ramsey spent months with Wittgenstein listening to him explain the Tractatus, yet sometimes Wittgenstein himself was unclear about what he meant.)
Each of our thoughts is a projection of parts or strata of the universe. From universe to mind is a series of projections. Studying a person’s thought is a projection of a projection. The latter is doomed never to fully succeed compared to the former. Perhaps we can ask: even if we clarify a projection, what is the point? To accept the world’s projection through this projection? Is understanding how Kant distinguishes the beautiful and the sublime more meaningful than figuring out how many ways to write the character “回”?
The question in the history of philosophy is not whether one should study predecessors but whether one should devote oneself to such study (some philosophers, like Quine among twentieth-century analytic philosophers, refuse to study the history of philosophy). Once one devotes oneself, philosophy proper is no longer possible in a fundamental sense. Human life and energy are limited; perhaps years spent studying one Kantian concept may end up revealing that it is merely a misplacement or overlap in a vast projection.
Maybe this is a matter of differing aims and choices. To me, his epitaph is tragic, a complete tragedy. Of course I also know he would call my epitaph a complete tragedy.
For him Kant is an unclimbable summit; his life’s chief aim is to scale it, though he may never reach the top. Even so, that pursuit is more worthwhile than walking among other hills and plains. I do not regard climbing any summit as my ambition; I may gaze from afar or climb a little, but ultimately build my own dunes or simply enjoy the flowers on the plain.