Immanuel Kant (1724-1804)


Modern Rationalists ended up with the skepticism of Descartes.  Modern Empiricists ended up with the skepticism of Hume.  Something is terribly wrong with this picture.  Recall that Plato anticipated that experience alone could not account for knowledge, but his solution was rejected by subsequent philosophers as too “spooky.” (Speculative)  But as Raphael Demos point out:


“To some, the conception of a previous life with its opportunity for a glimpse of the eternal essences may appear fantastic. Yet to any one who believes that the soul survives the body the view that the soul antecedes the body should not seem unreasonable. In any case, the transcendental theory is only an interpretation of the immediate fact that experience fails to account for all of knowledge. The doctrine of the limitation of empiricism remains, whatever one's view about the origin of abstract ideas may be. We cannot derive our categories -- thinghood, quality, relation, causality, -- from experience, because we use them in understanding experience; we cannot derive our laws of thought -- such as the law of contradiction -- from experience, because they are presupposed in any actual process of thinking; we cannot derive universal principles from experience, because experience is limited to particular cases; finally, we cannot derive any concepts (such as white-square) from experience, because they constitute standards by which the data of experience are measured. The kernel of the Platonic theory is rationalism, namely that there is a non-empirical element in knowledge.”[1]  (emphasis added)


As we have seen, the empiricism of Aristotle avoids these limitations of Locke and co. only by appeal to Nous, arguably no less spooky that Plato.  So is the only alternative to spooky metaphysics an epistemological skepticism?


Kant does not think so.  However, he credits Hume with posing the problem so clearly.


I openly confess my recollection of David Hume was the very thing which many years ago interrupted my dogmatic slumber and gave my investigations in the field of speculative philosophy a quite new direction.[2]


One of the errors which lead Empiricism to its skeptical conclusions according to Kant, was their passive view of mind and experience.  Think the passive metaphors employed by Locke and Hume for instance.  (“Tabula Rasa,” “Impressions”) Such views suggest that experience of the world is passive affair where the mind merely “received” what it “given” in perception.


For Kant the problem was we were not paying close enough attention to the active role that mind plays in the constitution of our experiences of the world.  The rationalists had a point; we bring something to the party, something without which intelligible experience would not be possible.  And the empiricist were partially right; all knowledge originates in experience.  But the rationalists were wrong to believe that what we bring are IDEAS.  And the empiricists were wrong to think that all knowledge is GIVEN in experience.


Key Kantian Insight: Mind is NOT passive in experience, but rather active.  Mind constructs experience out of the raw sense data that the world provides.  (Sometimes referred to as Kant’s Copernican Revolution in Epistemology.)  Rather than asking “How does knowledge impress itself onto mind?” (passive metaphor) Kant asks “How does mind construct knowledge?”


We do not have access to the world as it exists in-itself (what Kant refers to as Noumenon), since all human experience is mediated by the active application of concepts (categorizing) of mind.


We can and do have precise knowledge of the world as organized and interpreted by human cognition (what Kant refers to as Phenomena).


According to Kant, the noumenal world may exist, but it is completely unknowable to humans.


To understand why he says this, it is necessary to see human experiences as having different content, but a consistent form.  If we were to abstract all content from human experience we would arrive at the pure form of experience. 


Think of this pure form of human cognition as a blank template into which mind pours all sensory information and thus arrives at a coherent experience.


Alternatively think of my (very old, MS DOS based) Maillist program that can organize records according to one and only one pattern.  Thus I have knowledge of how my 100th record and any other record will look (in broad outline) that is a priori (that is may knowledge is not grounded in the particular experience of my 100th record).


Though I don’t know what the CONTENT of the record is, I know the form because when I am referring to this program’s records, I am referring to products of its organizing function which does not/ cannot change.


Kant is very specific about what these forms and categories of experience are, but I’ll only refer to a few for illustration purposes.


Space and Time are the two pure forms of experience according to Kant.


All human experience will/ must conform to 3 dimensional Euclidian Space.

All human experience will/ must conform to uni-directional time.  (Past to present to future).


Kant gives us a way of resolving the age-old disputes of metaphysics-questions concerning reality as such. Since the claims of the metaphysicians are all synthetic a priori, Kant provides us with the following policy:


  1. Those that are not rules by which we must interpret our experience are either analytic, contingently true, or contingently false. (Analytic A Priori, Synthetic A posteriori) –roughly what Hume acknowledged.
  2. Those claims that are rules by which we must interpret our experience are true necessarily true. (Synthetic A Priori)
  3. Those claims that contradict rules by which we must interpret our experience are false necessarily false. (Synthetic A Priori)
  4. Finally, those claims that cannot be decided by appeal to the rules of our experiences and make no difference to our experience one way or the other are to be rejected as possible topics of knowledge.- commit them to the flames…


If a truth is not true because of our experiences, nor is it true because of the grammar or meanings of the sentences of our language, how else could it be defended? Kant gives a way in his synthetic a priori knowledge, knowledge that is of our own rules with which we (necessarily) constitute reality. Hume's fork had only two tines and consequently left unjustified many of our most important beliefs.  Kant provides a third tine to the fork.   A belief can be true, necessarily true, if it is one of those rules that we impose to constitute our experience. While the principle of universal causation is neither a matter of fact (generalization from experience) nor a relation of idea (an analytic truth) we can nevertheless know that it is necessarily true (of necessity the world will appear to us In the form of effect with causes.  So too with the principle of induction.  So too for our belief in the "external" or material world we shall always interpret our experience of objects in space as external to us and as material or substantial. But notice, our metaphysical notion of substance is no longer that which is, by definition, outside of our experience. It is now part of the rules by which we set up our experience.


Given’s Kant's revolution truth is no longer correspondence between our ideas and reality.  Rather truth arises from our imposition of our own system of rules (concepts or categories) upon the sense data given to us, and by which we constitute our reality (experience). Knowledge then is no longer to be thought of as gaining an understanding of a reality beyond our experience (i.e. things in themselves), but rather an understanding of how we constitute experience for ourselves.  This does not mean knowledge of experience is distinct from knowledge of objects, for the objects of our experience are all the objects that there are for (our) reality.


And elements of this knowledge can be know with certainty, for, he argued, we can be certain of the rules of our own experience. Kant defended the necessity of the truths of arithmetic and geometry as those rules that have to do with the a priori forms of our intuitions of space and time. According to Kant's philosophy in general, reality is the world of our experience, as we constitute it through the concepts of our understanding. Therefore, we can know it with certainty, for truth, in general, is our own construction.


Some might object here.  The world as I constitutes it in NOT the “real” world.  The real world exists with or without me, independently of how I constitute it, and it is THAT real world that I want to come to know.  Anything less is not genuine knowledge.  But note two things about this objection.  First, I’ll never see (or taste, or touch, or hear, of feel or smell) that “real world.”  The world I live in, the world that makes a common sense difference to me is not the ideal world of speculations dreamed up by philosophers.  There are the root s pragmatism here, I think.  Second, were I asking to understand the world UN-constituted by mind, what sort of request would that be?  Am I asking to conceive of the world without concepts?  To understand the world free of human understanding?  On the face of it, that appears to be a logical and well as an epistemological impossibility.


Let us return to Hume’s critique of induction and causality.


Hume asks: If a truth is not be justified on the basis of our experiences, nor by appeal to grammar or meanings of the sentences of our language, how else could it be justified?  


It Kan’t (LOL)


"Truths of reason" nor "matters of fact.” were thought to exhaust the possible types of justification. This was Hume's dilemma, and with this two-test system of justification, he argued that many of our most important beliefs are both “unjustified” and "unjustifiable." 


Relations of Idea: Analytic A Priori

Matter of Fact: Synthetic A Posteriori


But Kant gives us a general way of giving an account of all those truths that metaphysicians have always argued about. Using Kant's terminology, we can say they are forms of synthetic a priori knowledge. Such knowledge is in fact knowledge of our own rules with which we (necessarily) constitute reality.  This then, is a third way: A belief can be justified as necessarily true if we can demonstrate that it is one of those rules that we impose to constitute our experience.   Incidentally, Kant defended the truths of arithmetic and geometry claiming that they are synthetic a priori claims.  He attempted to show that they were the (a priori) forms of intuition, the ways in which we must experience our world.


Kant grants that the principle of universal causation is neither a generalization from experience nor an analytic truth.  What it is, however, is a rule for "setting up" our world.  Think about my Maillist program again.  If I ask, how can I be certain that the “Name” field will always come first?  Because that’s the rule my program imposes in its activity constituting the record.  Alternativelly, like a rule in chess, this is not a move within the game but one of those rules that defines the game.


Likewise with the principle of induction; it is neither based upon experience nor a trivial truth, but a rule with which we govern all of our experience. So too for our belief in the "external" or material world, which Berkeley and Hume found so problematic.  We shall always interpret our experience of objects in space as external to us and as material or substantial, but not because we have empirical proof, but because we cannot experience the world any other way.  But notice, the pervious “metaphysical” notion of substance is no longer that which is, by definition, outside of our experience. Now it only refers to the rules by which we set up our experience and make experience intelligible to ourselves.


Initial Objections:


Objection: Einstein talks about warped, non-Euclidian space where the shortest distance between two points is NOT a straight line.


Response:  But even Einstein cautions us: don’t try to picture (image) this.


Objection: We can imagine (maybe even achieve) time travel.


Response: But our experiences will still be “forward.” (First I did, then I did, then ....)


Objection: Mystics talk about experience where “space and time drop away and all is one and time is unreal”


Response: Yes, well, even they claim that such experiences are “ineffable.”  They may be simply unintelligible as well, just as Kant suggests.  It is a controversial matter what, if any, knowledge one can get out of such experiences.


Positive and negative noumena


It is important not to think of noumena and phenomena as two different set of object.  Rather the first is reality uncognized by human mind than the latter of the very same reality cognized (filtered and interpreted ) by human mind.


Kant distinguishes between positive and negative noumena:


"If by 'noumenon' we mean a thing so far as it is not an object of our sensible intuition, and so abstract from our mode of intuiting it, this is a noumenon in the negative sense of the term".[3]


"But if we understand by it an object of a non-sensible intuition, we thereby presuppose a special mode of intuition, namely, the intellectual, which is not that which we possess, and of which we cannot comprehend even the possibility. This would be 'noumenon' in the positive sense of the term."[4]


These positive noumena, if they existed, would be non-empirical but nevertheless intelligible realities which we apprehended by some special non-sensory faculty.  (Perhaps something like Plato’s Forms mystically imparted to us, or Aristotle’s Essences which we apprehend through nous, again a mysterious "intellectual intuition." )


But Kant;s does not believe in these hypothesized mode of apprehension.  Thus, despite wishing to know the noumena positively, this is beyond the abilities of humans. 


Use of the categories of understanding therefore can never extend to anything other than to the objects of our experience (phenomena). There “object(s)” that correspond to the objects of our experience (source the sense data) but our concepts of understanding, being mere forms of thought for our sensible intuition, could not in the least apply to them.


“Doubtless, indeed, there are intelligible entities corresponding to the sensible entities; there may also be intelligible entities to which our sensible faculty of intuition has no relation whatsoever; but our concepts of understanding, being mere forms of thought for our sensible intuition, could not in the least apply to them. That, therefore, which we entitle 'noumenon' must be understood as being such only in a negative sense.[5]


The noumenon as a limiting concept


The noumena act as a limiting concept.  They demonstrate that our knowledge of phenomena is knowledge is a limited and qualified kind of knowledge. 


"Further, the concept of a noumenon is necessary, to prevent sensible intuition from being extended to things in themselves, and thus to limit the objective validity of sensible knowledge".[6]


"What our understanding acquires through this concept of a noumenon, is a negative extension; that is to say, understanding is not limited through sensibility; on the contrary, it itself limits sensibility by applying the term noumena to things in themselves (things not regarded as appearances). But in so doing it at the same time sets limits to itself, recognising that it cannot know these noumena through any of the categories, and that it must therefore think them only under the title of an unknown something".[7]


For Kant, noumenal/ phenomenal distinction is key to limiting reason to what he perceives to be its proper bounds (human experience).  This makes traditional metaphysics and its questions (such as the existence of God, the soul, and free will) beyond the scope of theoretical reason.  These questions may ultimately be the proper objects of faith, but not of reason.




·         There is epistemological justification for Causality (Contra David Hume)


Human cognition always organizes human experience of the world according to the concept of causality.  Therefore we can be certain a priori that all human experience will have/must have the same basic character since human cognition can only organize it one way.  In particular, we can be certain the every effect will have a cause since this is the way our minds always puts it together for us.


·         (Traditional) Metaphysics is impossible.


To conceive of reality (much less talk or speculate about) reality outside of space and time or "transcendent reality" is impossible, because, necessarily, any such conception would use human concepts and thus be mediated by mind.


These mediating concepts are perfectly serviceable for the constitution and organization of human experience, but inapplicable for gaining immediate knowledge of things-in-themselves.  Hence we cannot have theoretical knowledge of the way things "really exist" apart from human experience or consciousness of them.


·         Freewill




There is a curious (seeming) inconsistency between theoretical reason and practical reason. Theoretical reason (science) sees reality as a seamless series of causes and effects (determinism), moral reason does not. Any judgements of praiseworthiness or blameworthiness require the concepts of free agency and moral responsibility for personal choices. In judging the actions of a moral agent right or wrong, one necessarily presupposes that the action was uncompelled by prior events (free). 


In short, making sense of Moral Experience (and corresponding moral judgements) requires precisely the sort of personal free agency that casual determinism denies.


Kant’s Resolution:


Unconditioned causes, necessary for moral judgement, never occur nor can they occur in the world as we experience it  (i.e. the phenomenal reality: reality as cognized by human minds).   However we have no theoretical evidence (nor could we) for or against the claim that causal determinism is true of reality independent of human cognition (things-in-themselves, Noumenal Reality).


For all we know, causal determinism is not true of things-in-themselves.  Furthermore, given the “freewill” is a necessary presupposition for rational moral experience (the only alternative to absurdity) we have moral reason (though no evidence) to believe in (have faith in?) freewill.


Since moral experience only makes sense on the presumption of freewill, we therefore have moral reason to believe a metaphysical, a claim about things-in-themselves (i.e. that we have free will).  We can be a (morally) certain that humans have freewill as we are (morally) certain that Hitler ought not to have done the evil things he did,.  And, since there is no theoretical evidence against free will and it is the only rational alternative to absurd moral judgements, we ought to believe in freewill where the “ought” is both rational and moral.


But for Kant moral/practical reason is the only vehicle we have to speculate and draw conclusions about transcendent reality (things-in-themselves).  He believed that the existence of things like God, freedom, and the soul which could neither be proved nor disproved by theoretical (pure) reason, were necessary postulates of practical reason (systematic moral experience).  From a practical (moral law) point of view, it makes much more sense to accent to the existence of God, freedom and immortality then to deny them or to remain agnostic.


Opens the door to Radical Relativism:


Kant believed that our (human) empirical knowledge was subjective but also universal (NOT RELATIVE) because the pure forms of experience and the categories of thought were universal for all humans. Since you are “running the same organizing program” that I am running, you are putting the world together in pretty much the same way that I am porting the world together.  Thus my subjective truth is the same as yours.  Therefore, he could believe that what is true for one human is true for all humans. 


(Now God or aliens from another planet may have very different forms of experience and thus different knowledge and truths, but the human task of inquiry doesn’t involve them- yet at least.  These are merely speculative concerns, not practical ones about which scientists need to worry.) might object to Kant’s view.   For instance, what if we do NOT all put the world  together in basically the same way (e.g. woman according to a female template, men according to a male template)?  If “Men are from Mars and women are from Venus” then we are not experiencing the same worlds because we’re building our worlds with the same input but according to different templates.  We are, in a very real sense, living in different worlds, and truth must be relativized to groups of cognizers.  Rather than univalent, truth becomes bivalent or, perhaps, multivalent.  It is potentially as multifaceted as there are minds, and no basis would exist for claiming that any worldview was privileged among the plurality.  If this were the case, it is unclear what could recommend one worldview over another except prejudice or a political agenda.

[1] [1] Introduction to Plato Selections, ed. Raphael Demos (1927)

[2] Immanuel Kant, Prolegomena to Any Future Metaphysics, trans. Lewis White Beck. Copyright © 1959. Reprinted by permission of Pearson Education, Inc., Upper Saddle River, NJ.

[3]Critique of Pure Reason A250/B307,P267(NKS)

[4] Critique of Pure Reason  A250/B30,P2677(NKS)

[5] Critique of Pure Reason B309,P270(NKS)

[6] Critique of Pure Reason A253/B310

[7] Critique of Pure Reason A256/B312,P273