THE LIMITS OF SENSE AND REASON: A Line-By-Line Critical Commentary on Kant’s “Critique of Pure Reason,” #15–Prelude to Kant’s Philosophy of Natural Science.

By Robert Hanna

[I] was then making plans for a work that might perhaps have the title, “The Limits of Sense and Reason.” I planned to have it consist of two parts, a theoretical and a practical. The first part would have two sections, (1) general phenomenology and (2) metaphysics, but this only with regard to its method. (Letter to Marcus Herz, 21 February 1772 [C 10: 129])



Previous Installments:

#1: Introduction to The Limits of Sense and Reason

#2: Bii/GW91 The Motto

#3: Aiii/Biii/GW93–97 The Dedication

#4: Avii-ix/GW99 Preface to the First (A) Edition.

#5: Axi note/GW100–101 Preface to the First (A) Edition

#6: Axi note/GW100–101 Preface to the First (A) Edition

#7: Axii-xiv/GW101–102 Preface to the First (A) Edition

#8: Axv-xvi/GW102–103 Preface to the First (A) Edition

#9: Axvi-xvii/GW103 Preface to the First (A) Edition

#10: Axvii-xx/GW103–104 Preface to the First (A) Edition

#11: Axxi-xxii/GW104–105 Preface to the First (A) Edition

#12: Bviii-ix/GW106–107 Preface to the Second (B) Edition

#13: Bix-x/GW107 Preface to the Second (B) Edition

#14: Bx-xii/GW107–108 Preface to the Second (B) Edition

A Note on References to Kant’s Works


The first five installments in this series followed the 2019–2020 version of THE LIMITS OF SENSE AND REASON, aka LSR.

Starting with installment #6, subsequent installments follow the revised and updated 2021 version of LSR.

In any case, you can read or download a .pdf of the complete text of the 2021 version of LSR HERE.

Because LSR is an ongoing and indeed infinite task, revised and updated .pdfs of the complete text will be uploaded to that URL on a regular basis.


CPR TEXT Bxii-xiv/GW108–109 Preface to the Second (B) Edition

It took natural science much longer to find the highway of science; for it is only about one and a half centuries since the suggestion of the ingenious Francis Bacon partly occasioned this discovery and partly further stimulated it, since one was already on its tracks — which discovery, therefore, can just as much be explained by a sudden revolution in the way of thinking. Here I will consider natural science only insofar as it is grounded on empirical principles.b

When Galileo rolled balls of a weight chosen by himself down an inclined plane, or when Torricelli made the air bear a weight that he had previously thought to be equal to that of a known column of water, or when in a later time Stahl changed metals into calxc and then Bxiii changed the latter back into metal by first removing something and

a Kant’s text reads “gleichseitig” (equilateral); but on the basis of his correction in a letter to Schütz of 25 June 1787 (10: 466), he appears to have meant gleichshenklig (isosceles).


c Kalk. Kemp Smith translates this as “oxides,” but that is anachronistic; prior to the chemical revolution of Priestley and Lavoisier, the calx was conceived to be what was left of a metal after its phlogiston had been driven off; only later was it discovered that this process was actually one of oxidation.

then putting it back again,* a light dawned on all those who study nature. They comprehended that reason has insight only

* Here I am following exactly the thread of the history of the experimental method, whose first beginnings are not precisely known.

into what it itself produces according to its own design; that it must take the lead with principlesa for its judgments according to constant laws and compel nature to answer its questions, rather than letting nature guide its movements by keeping reason, as it were, in leading-strings; for otherwise accidental observations, made according to no previously designed plan, can never connect up into a necessary law, which is yet reason what reason seeks and requires. Reason, in order to be taught nature, must approach nature with its principlesb in one hand, according to which alone alone the agreement among appearances can count as laws, and, in the other hand, the experiments thought out in accordance these principlesc — yet in order to be instructed by nature not like a pupil, who has recited to him whatever the teacher wants to say, but like an appointed judge who compels witnesses to answer the questions he puts to them.

Thus even physics owes the advantageous revolution in its way of thinking to the inspiration that what reason would not be able to of itself Bxiv and has to learn from nature, it has to seek in the latter (though not merely ascribe to it) in accordance with what reason itself puts into nature. This is how natural science was first brought to the secure course of a science after groping about for so many centuries.



Kant now moves on to an extremely important, although tantalizingly brief, discussion of how natural science or physics became an authentic a priori objectual science.

One crucial feature of Kant’s discussion here is very clear and distinct: physics, like mathematics, entered the “highway of science” (Heeresweg der Wissenschaft) by virtue of a “sudden revolution in the way of thought ”(CPR Bxii).

This thought-revolution, again like mathematics, consisted in shifting from the idea that our rational human a priori knowledge of necessary or essential properties of objects is derived by induction from individual or collective samples, to the idea that a priori knowledge is generated by self-knowledge of the spontaneous cognitive activity of human theoretical reason in non-empirically introducing formal features into its mental representations of objects.

This is an absolutely fundamental epistemic point that needs to be re-emphasized, because it attaches to all kinds of a priori knowledge, whether in logic, mathematics, natural science or physics, or metaphysics:

[R]eason has insight (Einsicht) only into what it itself produces (hervorbringt)according to its own design (Entwurfe). (CPR Bxiii)

Crucially, rational insight has its own characteristic cognitive phenomenology,[i] that is, rational insight has its own consciously-experienced specific character.

So a priori knowledge for Kant is how human reason consciously experiences and self-consciously recognizes the results of its own cognitive activity in the structured non-empirical products of that very activity, in such a way that the apparent, phenomenal, or manifestly real natural world also necessarily conforms to those very non-empirical structures.

Insofar as this relationship obtains, then not only are the necessity and non-empirical character of human rationality exported from the cognizing subject to the form of the manifestly real natural world, but also the purposiveness and categorical normativity of human rationality are correspondingly exported from the cognizing subject to the form of the manifestly real natural world.

Via this subject-to-world exporting, the manifestly real natural world is thereby also constituted as a meaningful world in all the relevant senses of that phrase.

By that I mean that the manifestly real natural world, for Kant, is an “enchanted” world shot through with logical form, mathematical structure, informative modes-of-presentation, truth-value, logical consequence, synthetic a priori consequence, counterfactuals and other subjunctive conditionals, especially those relating to:

(i) choice and free will (for example, “If Lincoln had not finally resolved to end American slavery by means of The Emancipation Proclamation, then he would not have been assassinated by Wilkes Booth”),

(ii) law-governed natural causal necessitation,

(iii) organismic life,

(iv) purposiveness of all kinds,

(v) aesthetic value, and

(v) moral value.

Hence the manifestly real natural world, for Kant, is a fully human-mind-apt and human-rationality-apt world, for better or worse, and not a “fundamentally physical” and naturally mechanistic world that inherently excludes life, consciousness, beauty/sublimity, freedom, and non-instrumental value.[ii]

Of course the necessary-conformity component is what mediates this exporting, and is also what smoothly carries over our cognitive phenomenology and self-knowledge into our non-empirical knowledge of the manifestly real natural world.

As we have seen, the built-in limitation of human cognition to manifestly real natural objects alone, together with the necessary conformity component, jointly constitute transcendental idealism.

So this Kantian theory of a priori knowledge works if and only if transcendental idealism is true.

But a second crucial feature of Kant’s discussion of natural science or physics is not so very clear and distinct, and needs to be teased out.

Kant says explicitly that “here I will consider natural science only insofar as it is grounded on empirical principles” (CPR Bxii).

This is prima facie puzzling, for two reasons.

First, by his own explicit admission, Kant is attempting to show how logic, mathematics, and natural science or physics are all authentic a priorisciences, in order to compare and contrast them with metaphysics.

But how could this be consistent with focusing on how natural science or physics is grounded on empirical — and thereby a posteriori — factors?

Second, the very notion of a “principle” (Grundsatz, Prinzip) for Kant carries the prima facie sense of apriority (hence also the prima facie senses of necessity, non-sensory character, purposiveness, and categorical normativity).

But how could there then be anything like an empirical principle?

The very phrase “empirical principles” (empirische Prinzipien) seems to be an oxymoron.

Sorting out and then explaining these two prima facie puzzling features will take us to the very heart of Kant’s philosophy of natural science.

Much of this sorting-out and explaining can wait until we come to my commentaries on section V of the B Introduction, the Transcendental Deduction of the Pure Concepts of the Understanding, and the System of Principles of Pure Understanding.

Nevertheless, for the present purposes, we need to understand the texts surrounding the crucial “reason has insight…” text at Bxiii.

Here’s Kant’s basic line of argument, as I understand it.

Manifestly real material or physical nature is rationally comprehensible via natural scientific investigation, and thereby knowable a posteriori, only to the extent that it is governed according to principles or laws that have the epistemic, modal, non-sensory, purposive, and normative properties of necessary a priori truths, but are nevertheless also empirical,in that these principles and laws bind together apparent, phenomenal, or manifest material or physical objects and states-of-affairs that are themselves actual-world bound, and contingent.

Indeed, it’s precisely the principle-governedness or causal-law-governedness of manifestly real actual-world bound, contingent material or physical nature that makes it objective, and therefore a proper subject for the authentic objectual a priori science of physics.

So, odd as it might at first seem, even an empirical science like physics is an authentic science only and precisely to the extent that it has a non-empirical foundation that of course includes both logic and mathematics, but also extends beyond the purely logico-mathematical part of its foundation into the necessary and objectual a priori law-governed causal connections between actual-world bound, contingent manifestly real material things and states-of-affairs.

Otherwise put, according to Kant, natural science/physics is on the secure path of science only and precisely to the extent that it tracks the penetration of the necessary, non-sensory, purposive, and categorically normative structures of human rationality into the actual-world bound, contingent manifestly real material or physical world, all the way down.

Kantian naturalism is therefore a liberal naturalism in these three senses:

(i) nothing really exists outside of space and time,

(ii) everything real has causal powers in spacetime, and

(iii) mental properties are as essential to the basic architecture of nature as physical properties are.

Now the causal natures of these manifestly real material or physical objects and states-of-affairs are knowable a posteriori in all their specificity by means of experimental investigations that involve not only Baconian (that is, simple colligative, descriptive, and generalizing) induction, but also another method only partially anticipated by Bacon, namely, what I call Kantian abduction, aka Kantian inference-to-the-best-explanation.

As I noted earlier in LSR, what Charles Sanders Peirce calls abduction and what Kant calls reflecting judgment in the third Critique, is an inference from given particulars to a newly-created general conception or theory that adequately comprehends those particulars.

The theory of Kantian abduction was worked out both the Critical period and also the post-Critical period, via his scattered and all-too-brief remarks on “the method of those who study nature” (CPR Bxviii-Bxix n.), “the empirical affinity of the manifold” (CPR A113–114), “empirical laws” or “particular laws” (CPR A127–128 and B163–165) (Prol 4: 318–322), and also under the rubrics of what he calls the “regulative use of the ideas of pure reason,” the “hypothetical use of reason” and above all, “reflecting judgment” (CPR A642–668/B670–696) (CPJ 20: 211–217, 5: 179–181).

Kantian abduction, as exemplified by Galileo, Torricelli, and Stahl — Kant’s own cases-in-point — is not itself merely an empirical or a posteriori method, but in fact systematically closes the epistemic and semantic gap between empirical/a posteriori generalizations and non-empirical/a priori principles and causal natural laws.

It doesn’t do so, however, by what classical Logical Empiricist philosophy of science calls the hypothetico-deductive method, according to which general propositions about the material or physical world, originally derived by induction, are laid down, more or less arbitrarily, like extra axioms added to first-order classical logic, and then particular propositions about empirical consequences deduced from these axioms, which in turn are tested by observations.

For such a procedure would be unable to distinguish between, on the one hand, inductive hypotheses that are noumenal, and therefore humanly unknowable and anthropocentrically meaningless, and, on the other hand, humanly knowable empirically meaningful hypotheses that are specifically grounded on the objectively valid and objectively real transcendental idealism-based real metaphysics of rational human experience.

By sharp contrast to the hypothetico-deductive method, Kantian abduction doesn’t operate by induction + analytic stipulation + deduction + observation, but instead, as I pointed out earlier, it operates by synthetic a priori “counterfactual” or subjunctive conditional reasoning.

More precisely, according to the Kantian modal semantics of subjunctive conditionals that I am using, a Kantian abduction is a conditional proposition of this form,

Γ(X1, X2, X3, … Xn)[]→ Y

which in English says:

Necessarily, if Γ, a set of propositions X1, X2, X3, … Xn, jointly constituting a general conception or theory, were to be true, then Y, another proposition that describes an actual fact, would also be true.

This subjunctive conditional, in turn, is true if and only if

given the smallest restricted class of logically possible worlds, each member of which has the same basic transcendental structure as the manifestly real actual natural world, namely, the class of experienceable worlds, and is also consistent with the truth of all the propositions in Γ, then, in every member of this class of worlds, the truth of Γ synthetically necessitates the truth of Y.

Granting this truth-definition, then according to Kantian abduction, natural science advances inferentially


(i) the complete set of schematized synthetic a priori Principles of the Pure Understanding, which, in turn, collectively specify the basic transcendental structure of the apparent, phenomenal, or manifest actual natural world, and thereby determine the smallest restricted class of logically possible worlds — namely, the class of experienceable worlds,each member of which has the same basic transcendental structure as the manifest actual natural world, and is also consistent with the truth of some general empirical natural causal law proposition NCL,[iii]


(ii) NCL, which is partially derived by induction, but which also specifically reflects the creative and imaginative capacity for insight (aka the capacity for “genius”) of the individual natural scientist who formulates NCL, in whom “[genius] gives the rule to nature” (CPJ 5: 308), and which is postulated as the hypothetical antecedent of a subjunctive conditional of the form

If, given the schematized Principles of Pure Understanding, NCL were true in the experienceable worlds W1, W2, W3, … Wn,


(iii) a synthetically a priori entailed factual proposition Y in all those experienceable worlds, as the consequent of that same subjunctive conditional, of the form

… then Y would be true in the experienceable worlds W1, W2, W3, … Wn,


(iv) compares and contrasts that synthetic a priori subjunctive conditional implication Y with what is supplied by direct observational evidence in the actual world,


(v) also compares and contrasts the physical explanation provided by the subjunctive conditional proposition Γ[X1, X2, X3, … NCL, … Xn][]→ Y with all the other relevant possible sufficiently good physical explanations of the same actual apparent, phenomenal, or manifest natural facts, thus ruling out the worry that the explanation provided by Γ[X1, X2, X3,… NCL, … Xn][]→ Y is only “the best of a bad lot,” and not the best overall explanation,


(vi) asserts the general conception or theory Γ, constituted by the propositions X1, X2, X3, … Xn, including the general empirical natural causal law proposition NCL, as the true synthetic a priori representation of how a given natural causal law governs dynamic interactions and processes in the manifestly real actual natural world,

by Kantian inference-to-the-best-explanation.[iv]


[i] See, e.g., R. Hanna, Kant, Science, and Human Nature (Oxford: Clarendon/Oxford Univ. Press, 2006), ch. 7.

[ii] See, e.g., R. Hanna, THE END OF MECHANISM: An Apocalyptic Philosophy of Science (Unpublished MS, 2021), available online HERE .

[iii] See, e.g., R. Hanna, “Kant’s Neo-Aristotelian Natural Power Grid: On Kant and the Laws of Nature,” Critique 2018, also available online in preview HERE.

[iv] See also, e.g., I. Douven, “Abduction,” in E.N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Summer 2017 Edition), available online at URL = <>; and P. Lipton, Inference to the Best Explanation (London: Routledge, 1991).


Against Professional Philosophy is a sub-project of the online mega-project Philosophy Without Borders, which is home-based on Patreon here.

Please consider becoming a patron!



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store