|
Word Gems
self-knowledge, authentic living, full humanity, continual awakening
|
Great Books
Summary and Review
|
Immanuel Kant
Critique of Pure Reason
Study III: Section 2: We Are in Possession of Certain A Priori Cognitions
return to 'Kant' main-page
Commentary by ChatGPT
Study III: Section 2: We Are in Possession of Certain A Priori Cognitions
This section is short, but it is foundational. Kant now moves from the possibility of a priori knowledge to the stronger claim:
We actually possess it.
That distinction matters. He is no longer merely asking, "Could there be knowledge independent of experience?" He is saying, in effect:
Look at mathematics and the fundamental principles of natural science. They demonstrate that such knowledge exists.
1. Where Are We Now?
In Section I, Kant established the distinction:
Empirical knowledge depends upon experience.
A priori knowledge does not depend upon particular experience.
He also gave us the two principal marks of a priori knowledge:
necessity
strict universality
Now he takes the next step:
Certain human cognitions actually possess these characteristics.
So the argument progresses:
Experience is indispensable
↓
But not everything in knowledge is grounded in experience
↓
A priori knowledge is possible
↓
And we can identify actual examples of it
This is the foundation for everything that follows.
2. Kant's Basic Argument
Kant asks us to consider mathematics.
Take:
7 + 5 = 12
You don't establish this by going out into the world and examining seven apples plus five apples.
You don't need to count every possible group of seven and five things in the universe.
The proposition possesses a kind of necessity that no collection of experiences could provide.
Likewise, consider geometry:
A straight line is the shortest distance between two points.
Again, Kant's saying, it's not that we have measured every possible pair of points.
The proposition claims something that goes beyond the accumulated evidence of particular observations.
This is exactly what Kant means by a priori cognition.
3. But Kant Makes an Even More Important Observation
He says that mathematical knowledge is not merely a collection of definitions.
Mathematics gives us new knowledge.
This becomes enormously important later because Kant will distinguish:
analytic judgments
from
synthetic judgments.
We haven't reached that distinction yet, but Kant is preparing the ground for it.
A mathematical proposition doesn't merely unpack something already contained in a definition.
It extends our knowledge.
Yet it does so with:
necessity and universality.
And that combination is precisely what fascinates Kant.
4. The Example of Arithmetic
Take:
7 + 5 = 12
We can easily imagine someone saying:
"Of course that's based upon experience. We've learned to count by encountering objects."
But Kant wants us to distinguish learning arithmetic through experience from the basis upon which the mathematical proposition is true.
A child may learn:
7 + 5 = 12
through physical objects.
Seven blocks plus five blocks make twelve blocks.
But the physical blocks don't establish the necessity of the mathematical proposition.
They merely provide a concrete illustration.
You don't need twelve blocks in front of you to know that:
7 + 5 must equal 12.
That is the distinction Kant wants us to notice.
5. Mathematics Is Therefore a Major Exhibit
Kant's reasoning is roughly:
Mathematics possesses necessity and strict universality.
↓
Therefore mathematics contains a priori cognition.
And this gives us something extremely important:
A priori knowledge isn't merely a philosophical fantasy.
Human beings actually possess it.
The question now becomes:
How is it possible?
That question will eventually become one of the central questions of the Critique.
6. Kant Then Turns to Natural Science
He doesn't stop with mathematics.
He points toward what he regards as the principles underlying successful natural science.
The example he famously gives is:
"Every alteration must have a cause."
Again, notice the difference between:
"I have observed many alterations that had causes."
and:
"Every alteration must have a cause."
The first is empirical.
The second claims necessity.
And Kant wants to know:
Where does that necessity come from?
Experience can show us that something has happened repeatedly.
But it cannot, merely by repeated observation, establish that something must happen in every possible case.
7. This Is Where Hume Is Waiting in the Background
This point will eventually bring us directly to David Hume.
Hume had famously challenged the rational justification of necessary causal connection.
We observe:
A happens.
Then:
B happens.
Then:
A happens.
Then:
B happens.
We become accustomed to expecting B after A.
But where, Hume asks, do we actually see necessity?
We see constant conjunction.
We don't see some mysterious metaphysical force called causation.
Kant took Hume's challenge with enormous seriousness.
His response will eventually be:
Perhaps causality isn't something we learn by observing the world. Perhaps causality is one of the structures through which the understanding organizes experience in the first place.
We are not yet at that argument.
But this section is pointing toward it.
8. The Extraordinary Claim Being Prepared
Notice what is happening.
Kant began with:
"All our knowledge begins with experience."
We then discovered:
Knowledge need not be entirely grounded in experience.
Now we discover:
Some of our knowledge actually possesses necessity and universality.
Therefore:
There is genuine a priori cognition.
And now the great question becomes:
How is a priori cognition possible?
That is the question that will drive the Critique.
9. The Glass Analogy Again
Our glass-and-water analogy is especially useful here.
Experience provides:
the water.
But Kant has now shown us something remarkable:
The water we receive is organized in ways that cannot simply be explained by the water itself.
There are structures that seem to belong to the glass.
Mathematics provides one of Kant's first great pieces of evidence.
The necessity of mathematical knowledge seems to outrun particular experience.
So Kant asks:
What is the structure of the glass that makes this possible?
That is where the Critique is headed.
10. "Possession" Is Important
The title of this section says:
"We are in possession of certain a priori cognitions."
Notice the modesty of the claim.
Kant doesn't say:
"All knowledge is a priori."
Nor:
"Human reason knows everything independently of experience."
He says:
certain a priori cognitions.
This is a carefully limited claim.
Human knowledge is a mixture:
experience + a priori contribution.
The great task is to determine:
Which is which?
and ultimately:
How do they work together?
11. A Very Important Distinction for Our Glossary
We should keep three things separate:
A priori
Knowledge whose justification does not depend upon particular experience.
Pure
A priori knowledge with no empirical element mixed into it.
A priori cognition
An actual instance of knowledge possessing this character.
This distinction matters because Kant's project isn't simply:
"Let's think without experience."
He's investigating the a priori components that make experience and knowledge possible.
12. What Kant Has Established So Far
We can now put the opening sections together:
Section I
Knowledge begins with experience, but not everything in knowledge is grounded in experience.
Section II
We actually possess knowledge that is not grounded in particular experience.
Therefore:
A priori cognition is a genuine feature of human knowledge.
And now the Critique can legitimately ask:
How is such knowledge possible?
13. Questions to Carry Forward
I would preserve these questions as our intellectual bookmarks:
1. Where does necessity come from?
How can a finite being know:
"It must be so"?
2. Where does universality come from?
How can we legitimately know:
"There can be no exception"?
3. What does the mind contribute?
If experience cannot supply necessity, what does?
4. How does mathematics achieve certainty?
What makes mathematical knowledge possible?
5. How is necessary natural science possible?
How can statements about the natural world possess necessity rather than merely empirical probability?
And ultimately:
What is the architecture of the "glass"?
That is the question toward which Kant is now steering us.
14. Plain-English Version
Kant's argument in this section can be reduced to:
We learn about the world through experience. But we also possess knowledge that seems to be necessary and universally true—not merely true because we have observed it repeatedly. Mathematics is the clearest example. Therefore, some knowledge must have a source other than particular experience. The great question is now: where does this a priori knowledge come from, and how is it possible?
Mental Anchor
A priori knowledge isn't merely possible—we actually have it.
Or, even more pointedly:
Experience gives us facts; mathematics gives us necessity. Where does the necessity come from?
That question is going to become increasingly difficult—and increasingly interesting—as Kant moves from a priori knowledge to a priori forms, and eventually to the categories of the understanding.
Ed: The example of 7 + 5 = 12 isn’t clicking with me as a priori knowledge.
The child learns 7 + 5 = 12 with blocks – and so later, with other examples, and with more complex math, isn’t the child merely extending the “blocks” experience? I don’t see the “necessity” in this. Why isn’t this simply an extension of “block experience” writ large?
Chat: Your objection is exactly the right objection to raise, because it forces us to distinguish two things that are easily blurred:
How a person learns mathematics
from
What makes a mathematical proposition necessarily true.
Kant's claim concerns the second, not the first.
And I think my previous use of the blocks example made the distinction less clear than it should have been.
Let's stay with your child
The child has seven blocks:
[1 2 3 4 5 6 7]
and then five more:
[1 2 3 4 5]
The child puts them together and counts:
1, 2, 3 ... 12.
You say:
Why isn't the child simply extending this block experience?
At the level of learning, it is.
The child may very well learn arithmetic through concrete experiences with blocks, fingers, toys, apples, etc.
Kant is not claiming that the child mysteriously possesses the proposition 7 + 5 = 12 before ever encountering objects.
The important question comes afterward:
Does the physical experience with the blocks constitute the foundation of the truth that 7 + 5 = 12?
Kant says no.
Here's the distinction that I think will make it click
Imagine that the child has counted:
7 blocks + 5 blocks = 12 blocks.
Now remove the blocks.
The child can subsequently reason:
7 + 5 = 12.
Now remove every physical object in the universe.
The proposition remains true.
And here's the really important part:
Suppose tomorrow someone announces:
"We've discovered a new kind of physical universe in which seven objects plus five objects make thirteen."
That would not be a scientific discovery.
It would be a contradiction of the mathematical operation itself.
The physical objects illustrate the proposition; they don't establish its necessity.
But your objection goes even deeper
You could say:
"Fine. Perhaps the child generalizes from countless experiences. Seven plus five has always produced twelve. Eventually the child develops an expectation of twelve."
This is a very good empiricist response.
But notice what you've actually obtained:
"Whenever I have performed this operation, I have gotten 12."
That's an empirical generalization.
You have not yet obtained:
"It is necessarily the case that 7 + 5 = 12."
And this is Kant's point.
No matter how many times you repeat the experiment, you accumulate:
observed regularity
not logical necessity.
You could perform the operation:
1 million times
1 billion times
1 trillion times
and the evidence would still have the logical form:
"It has always happened this way."
It would never, merely by accumulation, become:
"It must be this way."
Here's an even better test
Forget the child entirely.
Suppose nobody had ever had physical blocks.
Suppose there were no apples, no fingers, no physical objects whatsoever.
Would:
7 + 5 = 12
become false?
Clearly not.
That is what Kant wants us to notice.
The physical world can give us instances of arithmetic.
But the truth of the mathematical relationship isn't established by those instances.
Now we come to the genuinely difficult part
You may reasonably respond:
"But then where DOES the necessity come from?"
Exactly.
That is Kant's question.
And this is where we must resist moving too quickly.
Kant is not merely saying:
"The child learned it from the blocks, but somehow the answer is a priori."
He's saying:
The experience of the blocks cannot explain the necessity contained in the mathematical proposition.
So he asks:
What is the source of that necessity?
His eventual answer involves the a priori structures of human cognition, particularly our forms of intuition—space and time—and his distinction between analytic and synthetic judgments.
And that is why we should not try to settle 7 + 5 = 12 right here.
The example is actually a doorway into a much larger argument that we haven't yet reached.
There is another reason your resistance is valuable
Modern readers often find Kant's treatment of arithmetic in this passage difficult because Kant's particular account of why:
7 + 5 = 12
is synthetic a priori is not obvious to us.
And I don't want us to simply say:
"Kant says it's a priori, therefore it is."
That would violate the entire spirit of our study.
We should make him earn the conclusion.
When we reach the relevant discussion, we should stop and ask:
Why isn't arithmetic merely an elaborate abstraction from physical counting?
Then we can examine Kant's actual argument.
And we can bring in other perspectives—mathematical logic, geometry, and later developments in the philosophy of mathematics—without prematurely forcing them into Kant's framework.
For now, I would revise our statement
Instead of saying:
"The child learns 7 + 5 = 12 through blocks, but the proposition is nevertheless a priori."
I would say:
"The child may learn arithmetic through experience, but Kant asks whether the necessity of arithmetic itself can be grounded in experience."
That's much clearer.
And it preserves your objection.
Your "blocks" insight actually gives us a very useful distinction
Psychological origin
How did I come to learn this?
Perhaps:
blocks → counting → practice → abstraction → arithmetic.
Epistemological foundation
What makes this knowledge necessarily true?
That's Kant's question.
Ultimate philosophical question
What must be true about the human mind for such necessary knowledge to be possible at all?
That's where the Critique is taking us.
And I think that third question is the one we really want to follow.
So I wouldn't try to make 7 + 5 = 12 "click" yet. Your resistance is telling us that we haven't supplied enough of Kant's machinery to make the example intelligible.
We should keep the question alive:
Can necessity ever be extracted from accumulated experience?
That is the real issue.
And when Kant gets to the analytic/synthetic distinction, we will return to the blocks and see whether his argument actually succeeds.
Ed: Maybe we should wait then; however, my burning question right now is, why can’t accumulated experience – a “trillion times” you said – legitimately establish necessity? This is done all the time in science with their so-called “laws” even though a proposition has not been tested everywhere in the universe. But then, this “discounting the possibility of the black swan” has been pointed out by others – Goldman, Feynman – as an error in logic.
Chat: I think your question exposes a very important distinction that we should make before proceeding further into Kant.
The short answer is:
Accumulated experience can give us extraordinarily strong grounds for expecting something to be true, but it cannot, by accumulation alone, logically transform "has always been observed" into "must always be true."
That is the classic problem of induction.
Consider your trillion observations
Suppose we have observed a particular physical law operating:
1 time
10 times
1 million times
1 trillion times
The evidence becomes overwhelmingly persuasive.
But logically, there remains a distinction between:
It has happened this way in every case observed.
and:
It necessarily must happen this way in every possible case.
The first is an empirical proposition.
The second is a necessity claim.
And no increase in the number of observations, by itself, changes the logical category.
Your scientific "laws" example is exactly right
Science routinely says things such as:
"Objects near Earth's surface accelerate downward at approximately 9.8 m/s²."
But science doesn't mean:
"We have personally tested every object that has ever existed."
Rather, scientific laws are generalizations supported by evidence, theory, prediction, replication, and successful application.
They are extraordinarily powerful.
But science itself, properly crafted, distinguishes:
empirical confirmation
from
logical necessity.
A scientific law can be extremely well established and nevertheless be revised when a broader theory explains phenomena that the earlier law could not.
Newtonian mechanics is the obvious example.
For centuries it worked extraordinarily well.
Then relativity showed that Newton's equations were not the final description of mechanics under all conditions.
So:
"It has worked in every relevant case we've tested"
is not identical to:
"It could not possibly work otherwise."
And here is where your black-swan point becomes important
David Hume had already identified the basic problem centuries before the modern scientific discussion.
You observe:
white swan
white swan
white swan
white swan
...
Eventually you say:
"All swans are white."
But the next swan might be black.
And here's the logical sting:
The observation of a black swan doesn't merely add one more piece of information. It reveals that the universal proposition was never logically entailed by the previous observations.
The million white swans were all perfectly compatible with:
"All swans are white."
but they were also compatible with:
"Nearly all swans are white, but somewhere there exists a black one."
That's why accumulation doesn't produce necessity.
But I want to make an important correction to our earlier discussion
I don't want us to accidentally make Kant's argument appear stronger than it is.
You said:
"Science does this all the time with its so-called laws."
Absolutely.
And therefore we should distinguish three levels of confidence:
1. Observation
"I have observed A followed by B."
2. Inductive generalization
"A is consistently followed by B."
3. Necessity
"A must be followed by B."
The move from 1 → 2 can be enormously productive.
The controversial step philosophically is 2 → 3.
And that is where Kant enters the picture.
Now something fascinating happens with Hume and Kant
Hume essentially says:
Experience gives us regular succession, but not necessary connection.
You never actually observe the "must."
You observe:
A → B
A → B
A → B
and your mind develops an expectation.
Kant is deeply dissatisfied with this.
He asks:
What if necessity isn't something we discover in the objects through repeated observation?
What if:
necessity is supplied by the structure of the understanding itself?
Now we are getting much closer to Kant's real argument.
And this returns us to your original objection about 7 + 5 = 12
You were right to resist my earlier presentation.
The blocks demonstrate:
7 physical objects + 5 physical objects = 12 physical objects.
But that demonstration doesn't explain the necessity of the mathematical proposition.
And now we can see why Kant thought the example mattered.
He wasn't primarily interested in how the child learns arithmetic.
He was asking:
What makes the proposition necessarily true?
That's an entirely different question.
And I think we should put that question aside temporarily, exactly as you suggested, until Kant gives us the conceptual machinery—especially analytic/synthetic judgments and his treatment of space and time—with which he thinks he can answer it.
One final point: I think your "trillion times" formulation actually helps us sharpen Kant
We should not say:
"Experience can never establish a scientific law."
That would be wrong.
Experience can establish an enormously well-supported scientific generalization.
Nor should we say:
"Scientific laws are merely guesses."
Obviously not.
Rather:
Experience can establish extremely powerful, reliable, predictive knowledge without establishing metaphysical necessity.
And that distinction is going to be one of the central fault lines of our Kant study.
It leaves us with a superb question to carry forward:
Where does the "must" come from?
Does it come:
from the world?
or:
from the mind?
or:
from some relationship between mind and world that neither the empiricist nor the rationalist has adequately understood?
That, I think, is the question Kant is really going to make us wrestle with.
Ed: Your last comment raises a very interesting item: Hume’s famous point about the failings of induction suddenly becomes clear: there is no “must” in his billiard balls example. There is no necessary causality – because we could theoretically create a situation in which the billiard balls do not bounce off each other. I hadn’t seen that Hume was the forerunner of Kant’s “necessity” embedded in a priori knowledge.
Chat: You have just put your finger on one of the most important pieces of the Kant story. And I would make one small refinement to your formulation.
Hume did not intend his analysis as a forerunner of Kant's doctrine of a priori necessity; rather, Hume's attack on our ability to derive necessity from experience helped create the problem that Kant then tried to solve. Kant himself famously said that Hume's work awakened him from his "dogmatic slumber."
And your billiard-ball example makes the issue unusually clear.
Hume's billiard balls
We watch one billiard ball strike another:
Ball A moves → strikes Ball B → Ball B moves.
We watch this happen thousands of times.
What do we actually perceive?
We perceive:
proximity
contact
sequence
motion.
But Hume says we don't perceive some additional thing called:
NECESSITY
There isn't a little metaphysical "must" traveling from Ball A into Ball B.
And this is the crucial insight you've just articulated:
We can imagine the next collision going differently.
We can imagine:
Ball A strikes Ball B → Ball B does not move.
That may be physically unlikely under ordinary circumstances, but there is no logical contradiction in imagining it.
And that tells Hume something important.
If the opposite is conceivable without contradiction, then the proposition:
"A must cause B"
cannot be established merely by observing A and B.
Hume's devastating question
Hume effectively asks:
Where did you get the "must"?
You saw:
A, then B.
You saw it again:
A, then B.
And again:
A, then B.
But nowhere in the sequence did you actually observe:
A must produce B.
The necessity seems to be added by the mind.
Hume explains this in terms of custom or habit:
repeated conjunction → expectation → belief in causation.
So the mind moves from:
"A has always been followed by B."
to:
"A will be followed by B."
But the second proposition isn't logically contained in the first.
And now Kant enters
This is where your observation becomes so important.
Kant looks at Hume's analysis and essentially says:
If experience cannot supply necessity, perhaps necessity comes from somewhere else.
And that "somewhere else" cannot simply be another experience.
Because then we would have exactly the same problem again.
So Kant asks:
Could necessity be supplied by the structure of cognition itself?
That is the beginning of a radically different approach.
Hume says approximately:
We don't discover necessary causality in experience; our minds develop the expectation through habit.
Kant asks:
What if causality is not merely a habit but an a priori condition under which we experience events as an ordered world at all?
Now we are approaching Kant's category of causality.
This gives us a beautiful three-step progression
Empiricism
I observe A followed by B.
Hume
I observe succession, but not necessity. The "must" cannot be extracted from experience.
Kant
Then perhaps the "must" originates in the mind's a priori structure for organizing experience.
That is an enormous philosophical transition.
And it connects directly to the question you raised earlier:
Does the mind contribute something to what we call reality?
Kant's answer increasingly becomes:
Yes—not by inventing the objects, but by supplying conditions under which objects can become intelligible objects of experience.
And notice the subtlety
Kant isn't simply saying:
"Hume was wrong."
In one respect, Kant agrees with Hume completely:
You cannot observe necessity by looking at repeated events.
Where they diverge is over what follows from that discovery.
Hume:
Since necessity isn't given in experience, we cannot rationally establish necessary causal connection; we have habit and expectation.
Kant:
Since necessity isn't given in experience, we must investigate whether necessity belongs to the cognitive structure through which experience itself becomes possible.
That is the great move.
And this makes your earlier "glass" analogy even more powerful
Perhaps the billiard balls provide the water:
Ball A
Ball B
contact
motion
succession.
But where does the mind's organization of these into:
cause → effect
come from?
Hume says:
habit.
Kant will ask:
Is it something more fundamental—a necessary form of the understanding?
And that is exactly the territory we will eventually enter.
Mental Anchor
Hume discovered that experience gives us succession, but not the "must." Kant's great question is whether the "must" comes from the mind.
And I think this is precisely why your realization is so valuable at this early stage. We're beginning to see that Kant's Critique isn't a collection of arbitrary technical distinctions. It is an attempt to answer a problem that Hume had driven into the heart of philosophy.
Ed: And so, 7 + 5 = 12 is not primarily a creature of early learning but, more fundamentally, an expression of “must”.
Chat: Yes -- I think you have now put your finger on what Kant is after.
7 + 5 = 12 is not primarily interesting to Kant as something a child learns; it is interesting as an instance of necessary knowledge.
The child may learn it through blocks. But the blocks don't create the necessity of the proposition.
The distinction
The child learns:
"When I put seven things together with five things, I count twelve."
That is the psychological history of learning.
Kant asks a different question:
"Why is it necessarily the case that 7 + 5 = 12?"
And that is an epistemological question.
The physical blocks can demonstrate the result:
7 objects + 5 objects → 12 objects.
But they cannot establish:
7 + 5 could not possibly equal anything other than 12.
That's the "must" we have been looking for.
And now I see why your earlier objection was useful
You were saying, in effect:
"But if the child learned it from blocks, why call it a priori?"
The answer is:
The origin of someone's learning is not the same thing as the ground of the proposition's necessity.
A child can discover something through experience whose truth does not ultimately depend upon that experience.
That's a subtle distinction, and we shouldn't hurry past it.
But there's a fascinating problem waiting for us
We should not yet conclude that Kant has solved the problem.
We've merely identified his question.
He still has to explain:
Where does mathematical necessity come from?
And his answer is going to be surprising.
He will argue that mathematical judgments are synthetic a priori, meaning—roughly—that they add genuine knowledge while nevertheless possessing a priori necessity.
And this is precisely where I think we should slow down considerably.
Because your instinctive objection—
"Why isn't mathematics simply enormously sophisticated abstraction from experience?"
—is a serious philosophical position.
Kant has an answer.
But we need to let Kant build the machinery before we judge whether his answer works.
So I would actually put your formulation into our working notes:
Mental Anchor
7 + 5 = 12 is not fundamentally interesting because a child learned it with blocks; it is interesting because it expresses a "must." The blocks may teach us the proposition, but they do not create its necessity.
And then, underneath it:
The question we must now pursue: Where does the "must" come from?
That question may prove to be one of the keys that unlocks the entire Critique.
|