I agree that he doesn't give a reasoned justification. But he does give a causal explanation which is presented as being true. People believe x because y. I'm not trying to do a cute reversal for the sake of cute reversal, but the Humean (anti)project really is seriously compromised by this kinda self-contradictory maneuver. Conjunction has this effect on the person who witnesses it...

It's not bullshit though, for Hume didn't say the sun wouldn't rise. He said it might not. Such a possibility is perfectly coherent with a world in which the sunrises every single day!

Hume never denies there can be a world which always follows the same rule. If that's what the world does, then that's what happens. Just because anything and everything else is possible, it doesn't mean it happens.

Under Hume, the definition of what the world does moves from the idea (e.g. the rule which constrains possible to form what happens), to the world itself. The sun can always rise tomorrow, but it takes the existence of that state every morning, not a rule. Rules become expression of existing states (i.e. the behaviour of states), rather than constraints which limit what states could be-- even as the sun rises every morning, any other state was possible. Actuality doesn't undo the truth of possibility.

Conjunction doesn't need to be constant at all. In terms of a cause, it only needs to happen once. Let's I find this object. It's the only one of its kind. If I pick it up it will be destroyed and l result in the death of all life on Earth. Is there no causal relationship here? Well, no. There is one. If I refuse to pick up this object, the death of all life on Earth will be avoided.

"Rules" tend a little be different. In the above instance, for example, the rule of killing all life on Earth is only relevant for one brief moment. Unless, that relationship repeats in future times, there is no danger to picking up the object. While there is a "rule" to any instance of a cause, we don't talk about them in this sense because then we wouldn't have any use for them going forward.

So for our usual notion of causal "rules," we need repetition-- e.g. I know I will be alive tomorrow, and the sun will rise, so I can do X,Y,Z, plan my day, respond to how the world repeatedly affect me, etc.

We tend to confuse this repetition of conjunction with a constant. Instead of realising that the rules are only how the world is behaving at the moment, that it's a repetition which might never have been and could end in the next moment, we think they must necessitate what happens.

It's ourselves which are the primary concern here-- if a conjunction is constant, then we can't be wrong, the world will always turn out how we expect and, perhaps most importantly, we won't be dead. I mean just imagine what it would mean if we could just "pop" out of existence tomorrow: I might be dead and would have no way of preventing it!!

The desperation to deny radical contingency, to claim it doesn't make sense with a meaningful existence, is our pretence that we are not the sort of thing which could just cease tomorrow, which might be wiped out on whim. In saying, "But the sun MUST rise tomorrow," we really are telling ourselves the lie that we cannot die, that we are beyond the possibility of death or non-existence. We are simply too afraid to accept we might not be (which is quite silly when you think about it, for that one might not exist, does not mean that one will).

"What are the chances?" is an entirely irreverent notion in this context. Since radical contingency deals in logical possibility, rather than probability, there is no defined chance to anything. Possibility isn't causal. There is no means or standard to define what's going to happen or what's more likely to happen.

In fact, considering the world itself, we might say there is no chance at all. For given the casual states themselves, the is only one outcome: what exists. If we ask, for example, "Why does the sun rise rather than not. How come it the rules of the world didn't change to day?," there is no answer. That's just what happened. There is no chance this sun would do anything else, even though it could have.

Yeah, Hume tends to assume a kind of uncritical naturalism when it suits him, which is in tension with his anti-naturalistic epistemological views. But I wonder if this sort of thing can be seen from within – that is, from one's own case, one can come to know that we just tend to follow habits and impulses around without justifying them. Not a very satisfying way of thinking to someone with a Kantian soul, but if we're concerned with whether it's true or not, we need to be more careful.

I like to think I have a non-kantian enough soul that I don't instinctively recoil from that kind of thing. But what's Hume's soul like? I think he's dead-earnest about his explanation, and also wants it to be received in earnest.

So: the least Kantian of souls would recognize that one thing always leads to another, that whats going on now is tied to what happened before, and is leading to what will happen next. That's just how it goes, because that's how it goes.

A slightly more compromised soul, one exposed to ideas about "ideas", about decontextualized objects which press against the mind, might start to see experience as a series of these isolated impressions. It might seem like experience is built up of instances -perceptual freeze-frames of various objects and ideas -& that awareness is always awareness at an instant, utterly disconnected (or at least disconnectable) from other instances. So then: how do we connect them??

All of which is to say Hume is as steeped in artifice and inherited ideas as Kant, imho, but it doesn't seem like it because his prose is fun and avuncular.

If Hume isn't important for what he says he's doing but because he helps us see our own case from within, then, I mean, Kant just as easily can be someone who helps us see, from within, that we're always caught up in a causal series etc etc

Conjunction doesn't need to be constant at all. In terms of a cause, it only needs to happen once. — TheWillowOfDarkness

I stopped reading here, because here it became very clear you either haven't read Hume or you've utterly forgotten what you've read.

I didn't mention this earlier because I wanted to see where it was going, but Kant rejected a lot of Newtonian principles, and most of the principles for relativity are present in Kant, as well as Einstein being a fan, and co-opting his quote about concepts and precepts into one about science and religion.

I don't know where the idea that he was defending Newtonian physics comes from. Like most everything, he thought it was half-right, at best.

Can you expand on that? I haven't encountered that argument before and it feels a little confusing prima facie

Firstly he reminds us that thoughts of empirical concepts take place in a three dimensional void, which is logical, and then we construct physical objects in the imagination like this. He rejected the law of inertia mainly because of the wording of the principle, feeling that it would be taken to be empirical and not logical, leaving out the reactive living aspect of matter, and only talking about the physical quantifiable aspects of motion in a logical void.

Kant's very thing is ultimately objecting to an objectivity or primacy of any perspective, but only of mathematical determination with respect to natural principles. Rather than imagining an absolute space, we rather need a universal constant, it would seem to me.

As for the Einstein link you can see that Einstein was a fan and read the CPR as a teen, there's also that quote stealing thing, but mostly I just think that a lot of it can just be seen there.

I'm running on memory here, but my understanding was that Newton's principia, ala Kant, was partially a priori synthetic and partially a posteriori. So it was both empirical and logical -- not transcendental, but rather a sort of fact which couldn't be true in light of Hume's critique of causation.

Or, maybe a better way to say it: a fact which is true, while Hume's critique of causation also seems true, and these two things cannot both be true.

So I don't know if I'd say he was out to defend Newton, or provide a foundation for him either, as I said before. Rather, Newton is given as a kind of evidence for our having knowledge of causation in spite of Hume's critique.

The death of all natural philosophy” (4:544). In a later remark in the Mechanics, Kant explicitly objects that “the terminology of inertial force (vis inertiae) must be entirely banished from natural science, not only because it carries with it a contradiction in terms, nor even because the law of inertia (lifelessness) might thereby be easily confused with the law of reaction in every communicated motion, but primarily because the mistaken idea of those who are not properly acquainted with the mechanical laws is thereby maintained and even strengthened” https://plato.stanford.edu/entries/kant-science/#PhyPreCriPer

Here's the rejection in his words, and he actually thought that the law contradicted experience.

So: the least Kantian of souls would recognize that one thing always leads to another — csalisbury

The idea that things cause each other is grounded in everyday experience. The idea that they do infallibly according to necessary principles isn't. Anyone can tell you life has regularities and inconsistencies. The Enlightenment philosopher just sort of denies the latter as a matter of principle, not based on any evidence (and where the evidence seems to be against it, future theory will resolve the apparent inconsistency).

All of which is to say Hume is as steeped in artifice and inherited ideas as Kant, imho, but it doesn't seem like it because his prose is fun and avuncular. — csalisbury

I agree with this, and I'm not really sympathetic to Hume more than Kant. The point is just that when you have a problem, you can't bring in fantasies to solve it. Maybe we don't have Humean problems because Hume made some more basic mistake. For example, only if we have an impoverished view of perception are we tempted to think that we can't see one thing causing another.

For example, only if we have an impoverished view of perception are we tempted to think that we can't see one thing causing another — The Great Whatever

Why did Hume think we couldn't perceive causation? Because we only see the constant conjunction and not the underlying cause? Hume assumed that if there is such thing as causation, it had to be something unperceived.

Was that because we're sometimes wrong in assuming that A necessitates B, when it was really C that necessitates B?

Seems like since Hume most people have taken it for granted that we don't perceive causation. Either it's a habit of mind (which is contradictory since habits are causal as pointed out by Csalisbury), or it's something real, but unperceivable, like universals or laws of nature, which aren't empirical.

But what if one took a different tract and argued that we do perceive causation?

Isn't causation a conceptual relationship--a mental construct, an explanatory model, a way of thinking about the world that has no actual independent existence outside the mind doing the thinking, the explaining, constructing the model?

Mental construction is causative. It's the same as saying that causation is a habit of thought.

I don't understand what you mean here. Can't tell if you're agreeing or challenging.

My point is that relationships (of any kind) cannot meaningfully be said to exist in any mind-independent sense. Thus, causation is entirely mental--in this case a built-in way our minds apprehend the world--as Kant says.

Or maybe, you know, I'm drawing out the notion of radical contingency Hume uses against the notion of a continuing causal force.

No doubt Hume begins with the notion of causal forces as constant and continuing, but he quickly rejects it in radical contingency. Since anything may happen at anytime, the idea there are causal forces which must continue falls. Causation is just a partcular correlation of states.

This is where radical contingency makes everyone so anxious. It turns all causal relationships into a coincidence. Yes, the sun was caused to rise this morning, but that was nothing more than the whim of correlated states.

Hume's main point is against "continuation" is causal forces, at least in the sense of an outside force beyond the flux of finite states. He rejects the idea of acting causal force which moves the pieces of the world. For Hume, any causal relationship is just a concurrence of particular states themsleves.

I'd argue he didn't. We can't perceive the "thing" of causation because it is not a seperate state or actor in the world.

If I drop a rock, for example, I see an opening hand and a falling rock, not a "causation (whatever that might supposed to be)." Hume is more or less pointing out there is not some seperate object of a "causal force." There's nothing more to a causal relationship than certian states which occur together. When we observe causality, it's these correlated states we are aware of.

Hume is not really putting causation beyond observation, but rather pointing out it's only ever states of the world which do it.

To observe a cause and effect is nothing more than being aware of states which bring each other about-- we observe the cause of an opening and and the effect of a falling rock. There is no seperate object, force or rule acting upon the hand and rock to make this causal event occur.

(so Hume is really in agreement with TGW: we see things causing other things all the time. It why we don't see "causation." Things themselves do it all on their own, so we see them as causes and effects, rather than observing "causation." )

Why did Hume think we couldn't perceive causation? Because we only see the constant conjunction and not the underlying cause? Hume assumed that if there is such thing as causation, it had to be something unperceived. — Marchesk

One thing not being mentioned is that causation - out there in the world - is heavily contextual. Things happen in predictable fashion because the world is organised in some way that constrains what is possible. And that history accumulates over multiple spatiotemporal scales.

So the car crash couldn't have happened at that junction unless 100 years ago the road hadn't been built. Or if two seconds earlier, the driver hadn't been distracted by the phone ringing.

But physics of course is a reductionist modelling of causality that plays the useful trick of imagining timeless laws animated by instantaneous measurements. So when it comes to conceiving and perceiving the causes of events through this lens, it leads to the Humean situation where the perceived event seems to take up no time and thus have no causal history, nor future. We imagine the event to be punctate and contain no information apart from some number that gives it an instantaneous value - like a momentum or inertial velocity.

So everyone was reacting to Newtonian mechanics - a new metaphysics that broke the world apart in this particular fashion. And if you took it literally, perception became identified with acts of measurement. It was imagined that events had punctate value that could be abstracted away from all the surrounding context. Causality became bound up in a property like momentum that a mass possessed. These values could be plugged into rules - the equations - that were like Platonic ideas.

Thus causality was pushed out of sight. It either became hidden in timeless laws. Or it was concealed within the value assigned as the identity of some timeless event. Causation was reduced to correlation as an act of the abstracting scientific imagination. Real things got replaced by the numbers that stood for them within a new system of sign.

If that's the way we find best to model causation, then it makes it quite legitimate just to count events and treat a regularity of conjunction - a matching of theory and prediction - as "seeing causality at work". The damn thing - Newtonian mechanics - works. The philosophical error is then to pretend to be confused - to start claiming an epistemic crisis like Hume, and even Kant.

Logic itself is the same trick - the abstraction to the timelessness of a syntax of rules and variables. A system of pure sign that leaves its semantics outside of it as something to be determined in some other "informal" fashion. Someone has to decide the meaning of the words in a proposition, just as they have to decide what counts as properly measuring some event in the world with sufficient care.

So it should be clear to us - as the inventors - that we have developed a powerful modelling trick (one that takes modelling itself to its formal extremes). And the world "in itself" is exactly what had to be left out so that we could choose precisely what then to include back in as the abstracted elements of a formalised and timeless approach.

Hence events became perceived as contextless, memoryless and historyless as the way to assign them some punctate value (like some weight of motion in a direction). And from there it became difficult to see why one thing leads to another except that we have constructed some laws as an act of conception. The events themselves - due to the way we measure them - can no longer give us a necessary connection to some actual lived past that is the world "in itself". We no longer seem to see (at a scientifically modelled level) what we in fact do feel we see (at a regular biological Bayesian brain level) with our own eyes.

The animal brain is evolved to reason inductively. It works by taking a guess and predicting its future states, and that creates a context in which the suprising can stand out. The unexpected - the breaks with expectable causality - is what is being looked for. The lack of Humean continguity is the feature, not the bug, as it is the failures of causal reasoning which are the teachable moments for the critter.

But philosophy turns nature on its head with this new language-based trick of deductive thought. It flips us into the timeless view of the world where causes are eternal ideals - like laws - or essentialist properties, like the numbers assigned as the values of instantaneously measured events.

And now there is no connection that can be seen between one instant and the next. But that is just the way our formalisms operate - the timeless view we have imposed so as to make time itself an abstraction within the modelling.

I decided to track down the quotes I was thinking of in making my assertions. I found that I've made a mistake in saying Newton is the counter-example to Humean causation. What I was thinking was Newton, but that's not really stated in the quotes I was thinking of.

Something very much like the argument I outlined is there -- but not Newton. So, my bad there. But, to go over the quotes I was thinking. . .

In the preface to the 2nd edition, at the end of Bx:

Two [sciences involving] theoretical cognitions by reason are to determine their objects a priori: they are mathematics and physics. In mathematics this determination is to be entirely pure; in physics it is to be at least partly pure, but to some extent also in accordance with sources of cognition other than reason

That's definitely the quote I was thinking of in saying Newton, though this in particular doesn't link physics to Newton (as K. was definitely interested in physics at large, and not just Newton), or how that might serve as a counter-example to Humean criticisms of causation.

Later, on B21 there is a footnote in the introduction to the second edition, 2 paragraphs after introducing the central question of the critique, to these lines:

How is pure mathematics possible?
How is pure natural science possible?

Since these sciences are actually given [as existent], it is surely proper for us to ask how they are possible; for that they must be possible is proved by their being actual.

And the footnote reads:

This actuality may still be doubted by some in the case of pure natural science. Yet we need only examine the propositions that are to be found at the beginning of physics proper (empirical physics), such as those about the permanence of the quantity of matter, about inertia, about the equality of action and reaction, etc., in order to soon be convinced that these propositions themselves amount to a physica pura (or physica rationalis). Such a physics, as a science in its own right, surely deserves to be put forth separately and in its whole range, whether this range be narrow or broad

To this footnote the translator adds a footnote of his own, appended to the last sentence:

This Kant did in his Metaphysical Foundations of Natural Science (1786), Ak. IV, 465-565

This is where I got the notion of him deriving Newtonian physics, but surely Newton is not mentioned here either. Nor is the notion of Newtonian physics serving as the counter-example against Humean skepticism.

Earlier in the introduction, under II. "We are in Possession of Certain A Priori Cognitions, And Even Common Understanding is Never without Them" at B5 Kant stated:

...Now, it is easy to show that in human cognition there actually are such judgments, judgments that are necessary and in the strictest sense universal, and hence are pure a priori judgments. If we want an example from the sciences, we need only look to all the propositions of mathematics; if we want one from the most ordinary use of understanding, then we can use the proposition that all change must have a cause.

This is getting closer to how Kant is in disagreement with Hume, and highlighting a sort of principle which the common understanding uses (though, perhaps, this principle isn't something that Newton uses -- again, no support for that particular claim of mine).

Later we get closer to the language I used, albeit admittedly not with Newton referenced. I'll just note here I'm now uncertain why I thought Newton in particular to Hume. Kant certainly references Newtonian physics throughout the CPR, but I overstepped in stating that it was Newton who served as the counter to Humean skepticism, I believe, unless there's some reference I missed. However, even in that case I overstepped, because after reading this highlighted portion I'm pretty sure this is where I was getting everything I stated before. So even if the reference is there, I was in error because these were the sections I was thinking of anyways.

Mea culpa.

At B127, or in the section titled "Transition to the Transcendental Deduction of the Categories":

The illustrious Locke, not having engaged in this contemplation, and encountering pure concepts of understanding in experience, also derived them from experience. Yet he proceeded so inconsistently that he dared to try using these concepts for cognitions that go far beyond any boundary of experience. David Hume recognized that in order for us to be able to do this, the origin of these concepts must be a priori. But he was quite unable to explain how it is possible that concepts not in themselves combined in the understanding should nonetheless have to be thought by it as necessarily combined in the object. Nor did it occur to him that perhaps the understanding itself might, through these concepts, be the author of the experience wherein we encounter the understanding's objects. Thus, in his plight, he derived these concepts from experience (viz. from habit, a subjective necessity that arises in experience through repeated association and that ultimately is falsely regarded as objective). But he proceeded quite consistently after that, for he declared that we cannot use these concepts and the principles that they occasion in order to go beyond the boundary of experience. Yet the empirical derivation of these concepts which occurred to both cannot be reconciled with the scientific a priori cognitions that we actually have, viz., our a priori cognitions of pure mathematics and universal natural science, and hence this empirical derivation is refuted by that fact.

That last sentence, in particular, is pretty much what I was thinking of. I believe I must have basically interpreted "universal natural science" as equivalent to Newtonian physics, though by no means is that asserted here.

The facts, though, which are meant to stand as counter-examples to the Humean account of causation are the sciences of pure mathematics, and universal natural science.

Understand that both are actually mathematics, but mathematics is pure, and natural science requires empiricism and thus is impure, and the "pure part" in it that he mentions is itself mathematics. He is without a doubt influenced by Newton, and that article I linked says that people used to think he meant Newton, but now they think more so from Locke, but I think that it's pretty equally true to say that they both influenced him, and equally wrong to think that his notion of universal natural science just was one of their's.

I'm reading this article, though, and I don't think he's opposed to the law of inertia. He's opposed to the term "inertial force" -- because a force is what acts on an object or is the cause for a bit of matter's changing course.

EDIT: That was a really cool article. Just finished it now. Thanks for sharing it.

I agree with you on the latter point, now. I just had to read the text again to see why I thought that, and came up short. Thanks for the clarification.

In a later remark in the Mechanics, Kant explicitly objects that “the terminology of inertial force (vis inertiae) must be entirely banished from natural science — Wosret

This is a pretty Newtonian declaration actually, insofar as Newton himself famously refrained from 'feigning any hypothesis ('hypothesis non fingo') regarding what force is. It could be said that Kant was just trying to carry though this declaration to it's end.

--

Otherwise, the early modern debates over causality are actually really interesting. Christian Kerslake has a great book which covers alot of these 'contextual' issues, and one of his most important points I think is that the model of causality which we reflexively think of today (efficient causality) was, as he says, the least popular of all the available models of causality. Some excerpts:

"Both Hume and Leibniz are sensitive to the problems of justifying the concept of causality. This is in part due to the conjunction of available theories of causality in the eighteenth century: the notion that now strikes us as the most sensible approach to causality, that finite substances are responsible for the changes they cause in other substances (then called the theory of physical influx), was at the time the least popular. This was because the only way available to conceive the idea that a substance with a set of properties caused a change in another substance was through the explanation that there was a transmission of properties from the first to the second, which was held to be inconceivable. Therefore, the notions of occasionalism and pre-established harmony became popular among philosophers as elaborate avoidances of physical influx.

On Hume: "...Hume’s philosophy can also be seen to arise from the failure of the physical influx theory: he can find no evidence from the senses of any ‘transmission’ of properties, given that all the senses provide us with are distinct impressions. Given a lack of objective ground for the order found in the world, Hume turns to custom, and, ultimately ... to the notion of a pre-established harmony."

On Kant, "One of Kant’s most celebrated moves in the Critique of Pure Reason amounts to the construction of an abstract formalisation of the problem facing notions such as causality in the eighteenth century... . The concept of a causal relation must be synthetic: Leibniz, Kant and Hume all agree on this, if not in terminology. Furthermore, they agree in principle that the problem about causality concerns connections that should be, if they are to exist at all, a priori. Kant’s notion of the synthetic a priori simply names a problem faced by eighteenth-century philosophy – that of how to account for any possible nonlogical a priori connections. How is one to synthesise a priori two or more elements, whether they be Humean sensations, or Leibnizian perceptions?

...Kant will often address the situation functionally by simply saying that synthesis requires a 'third'. As Kant says in the Critique, ‘where is the third thing that is always requisite for a synthetic proposition in order to connect with each other concepts that have no logical (analytical) affinity?’ (CPR A259). Kant’s answer as to what this tertium quid is will vary enormously, but the ‘triangular’ structure of a priori cognition will remain constant. As we will see, in the early writings Kant seeks the third thing between God and world (cf. LM 15, Ak. 28:52), whereas later time (A155/B194) and experience in general (A157/B196) are said to be third things".

On Newton and Kant: "To explain the interaction of substances, Kant appeals to universal gravitation, and this will remain as the extralogical formal principle for the reciprocal action (succession and coexistence) of his system right up to the ‘Inaugural Dissertation’. Universal gravitation, as the sphere of nature, is the ‘phenomenal eternity of the general cause’ (TP 405; Ak. 2:410). Any determinate relation between substances thus depends on the status of the ‘world-whole’ ... The principle of real, as opposed to logical, determination has its final ground in the whole. ... Against Leibniz, Kant wants both to affirm physical interaction, and also, with Newton, to shift the ground for the determination of forces to the whole field of forces. As we will see shortly, this provides the rudiments for a scientific theory that resolves the physical influx controversies".

I wish I could quote the whole chapter, but the whole history is just so long and fascinating full of twists and turns - especially because, as Kerlsake points out, Kant actually changed his mind multiple times in the lead up to the CPR regarding the status of time, causality and force - all of which complicates his relation to Newton (quotes from Kerslake's Immanence and the Vertigo of Philosophy).

Kant, like Leibniz, ruled out absolute space and time. This is sort of anti-Newtonian. Space is absolute if it can exist independently of objects and so can be said to contain them.

On space for certain I agree with you, though I'd express uncertainty on my part about saying Kant was in line with Leibniz -- though maybe you're just saying it's similar, not the same. I think he takes a position in-between the two -- which is kind of his "move", if you think about it.

What I had more in mind were the laws of motion and our ability to predict the motion of matter with them, which we know and carries a kind of mathematical necessity with it which goes against the assertion that we can never know some effect through a cause but are merely habituated by repetition. Given such and such conditions, I can tell you where some bit of stuff will be in so much time, and I know this will be so -- I am not merely habituated to it.

But I think that was a bit of my own reading into the opening, there, to make sense of it. Natural science certainly doesn't have to be Newtonian -- it just seemed to make sense given its mathematical certainty and its relation to cause-and-effect. (and, of course, the frequent references to the three laws of motion, or at least formulations really similar to them)

On space for certain I agree with you, though I'd express uncertainty on my part about saying Kant was in line with Leibniz — Moliere

I didn't say their arguments were identical. Leibniz's approach strikes me as closer to Einstein's. Einstein occasionally relies on the reasoning in Leibniz's Law.

Fair. I wasn't sure which way you were saying.

Back to work, so not a lot of time for this, will be my excuse, but I'll attempt to explain my understanding of his rejection.

Firstly, "inertia" is taken from "inert", and matter is manifestly not inert. Indeed, he thought some kind of interaction, communication, or transference of information is at play, as well as not just external inert forces are at work, but dynamic living ones, which this doesn't give an account of.

His view of something more relational sounds like a famous thought experiment about two people moving towards each other in a void to me... "if all changes of motion are reciprocal and equal (since one body cannot move closer to/farther away from another body without the second body moving closer to/farther away from the first body and by exactly the same amount)".

The most glaring anachronism to me, is that he clearly doesn't think of matter and energy as the same things in different forms, but distinct things.

Yes, I meant Leibniz, not Locke. They begin with the same letter!