Varieties of Causation and the Humean Problem

**Note** My view of Causation is very much influenced by the reading of Edward Feser's Scholastic Metaphysics and David Braine's The Human Person. Further reading can be found in these works. Also, this is not meant to be a Philosophical Paper and thus the rigor with which I approached it was lower. Feser has blogged about a similar subject here: http://edwardfeser.blogspot.com/2017/04/the-problem-of-humes-problem-of.html


David Hume is the most loved and yet hated philosopher. Some have called him a sophist, and yet others are compelled to call him one of the greatest philosophers of all time. While Hume is often considered to be critiquing the possibility of metaphysics, it seems more accurately to evaluate him on his more immediate opponents, the rationalists.

Introduction

On the face of it, Hume's objection contra causality seems almost like Zeno's objection against motion. Hume, like Zeno, takes a commonsensical intuition that we have and questions it, which is no different than most philosophers. However, what makes Hume's and Zeno's objection connected more intimately is not the commonsensical nature of causation but the implications of the position.

That is, Humean skepticism about causality fundamentally rejects experience as such. Hume asserts that we can never say that we experience causation; merely our sense experience assumes that things will behave the way they always have because of our observation of them. That is, our experience is so permeated with causal relation that, like motion, it becomes hard to construe our experiences in terms of anything else. In fact, if causality was determined to be some kind of false idol by which our experience is ruled then it becomes hard to see why anything except analytic truths ought to be trusted.

If the regularity of our sense data is not to be trusted then for what reason ought we trust any non-necessary truth? The Cartesian question becomes more pertinent when the foundation of epistemic justification cannot be based on regularity. A posteriori synthetic truths do not seem to have any particular appeal if one moment is not causally related at all to the rest.

However, I would say this poses a problem most intimately with analytic philosophy more than any other group. Analytic philosophy has come a long way since the 1950's and 1960's of logical positivism but there still exists a trend which scientific knowledge is ultimately seen as fundamentally more true than philosophical knowledge. I would like to do a post on that at some point.

Science is more at threat than any other discipline because there is no reason to conclude any empirical test can support anything other than what happens in one instance. Probability cannot even to be thought of as reliable in this context because if each test cannot be connected by temporal priority then it is hard to see how any test could measure anything(for the introduction of a variable is not temporally dependent on this view). Furthermore, inductive analysis is needed to conclude from any particular example to some general principle but induction cannot be appealed to if these two events cannot be explicitly connected without begging the question. The proverbial stone which falls in China might as well have caused my headache as much as any scientific explanation. 

This is not to say that modern philosophers(which is not surprising given how specialized the field can be) completely ignore the problem, in fact most focus on trying to establish the validity of induction by discrediting terminology and trying to rework induction in science to fit into some kind of epistemologically pleasing set up. However, very rarely do they directly confront the problem of attacking the idea that one moment isn't connected to another logically and that causes and effects are loose and separate.

So quite obviously I consider this a huge problem for Modern Philosophy. This doesn't mean that I consider it to be a correct characterization of causation, in fact I find it to be quite sophistic when considered very closely. When I say sophistic, I mean that it seems prima facie to be correct but is really quite incorrect. However, I should that accepting Humean assumptions, it is very coherent. Coherence, however, is not the sole marker of truth, correspondence must be considered. I would also note that many do (implicitly) accept event causation within their philosophy.

That is, it is based on a certain historical position critiquing that movement but if we consider the broader range of views regarding causation, it seems to fall flat. So while I find it to be parasitic on Analytic Philosophy per se, I do not find it to be particularly compelling given alternate views of causation.

An Exploration of Causation 

In accordance with the principle of charity, I want to be clear on what I am critiquing. I am critiquing:

"All events seem entirely loose and separate. One event follows another; but we never can observe any tie between them. They seem conjoined, but never connected. And as we can have no idea of any thing which never appeared to our outward sense or inward sentiment, the necessary conclusion seems to be that we have no idea of connexion or power at all, and that these words are absolutely, without any meaning, when employed either in philosophical reasonings or common life." Hume, Of the Idea of necessary Connexion, Part II, Paragraph I

The first idea that I would like to discuss is that causation can not be characterized properly as some kind of thing(most people do not think this of course but the point of this is to realize that Causation is not easily justified). Hume correctly characterizes causation as not a type of thing that we observe. That is, we never observe a thing called causation. 

Hume than talks about how what we really experience is events. However, Hume says that is not sufficient. This is related to a fallacy known as Post Hoc Ergo Propter Hoc(After this, therefore because of this), in that if events are simply sequential, then there is no reason to conclude they are intrinsically related. Thus Hume concludes, causality is a habit of the mind rather than real. This would be known as event causation, as opposed to agent causation. 

One thing that Kant Scholars like to point out is that Hume is really fighting the Rationalist notion that the mind mirrors the world. Causation is simply another casualty of the demolition of Rationalist Metaphysics. Unfortunately, even he had to deny he believed his critique(thus emphasizing it's primarily rhetorical effect).

"But allow me to tell you that I never asserted so absurd a Proposition as that that anything might arise without a cause: I only maintain'd, that our Certainty of the Falsehood of that Proposition proceeded neither from Intuition nor Demonstration; but from another Source."

Some people try to use this to disprove the problem, but this is of course as silly as walking to answer Zeno.

Given this event causation, it seems that no two events can ever be plausibly be connected. Thus Science cannot conduct any experiments since prerequisite conditions have no effect on outcomes.  Furthermore, most of our daily interactions seem strange, almost unintelligible.

Many would go onto address Hume(most of them unsuccessfully), most notably Immanuel Kant who derived causation from the power of Consciousness to impose certain characteristics upon the world. Thus starting the Copernican Revolution of Philosophy.

This may make causation seem hopeless to save while keeping the world ontologically/epistemologically separate from the mind. One of the strengths of the Humean critique is that it appeals to the difficulty with justifying causation. Most people simply carry with them certain assumptions and these assumptions make the argument hard to answer.

To answer this critique, we must realize that what we do experience is potentialities being activated. This requires a bit of explanation of causal powers and potentialities. For something to have the potential to do something else, it must be able to do that given some circumstances. This is contrast with actuality which is where something is in a certain way. For example, a block of marble has the potential to be a statue, when it becomes a statue, it is now actually a statue. This is what we call the reduction of potency to act. 

When I observe fire burning something, I derive that fire has some type of causal power when acting upon something else. Similarly, when we observe any type of causal relation, our minds perform an intellection of the participants and derive any formal conclusions.

Hume could critique that we merely observe these events and thus have no grounds for avoiding the objection. As Hume was responding to the Rationalists, who are chiefly influenced by Cartesian Bifurcationism(see Wolfgang Smith's The Quantum Enigma for a great analysis/refutation of Descartes), he seems to make quite a few assumptions about causation inherent to that worldview.

Suffice to say(without going into a lengthy critique of Descartes), Scholastic philosophers do not accept the fallibility of the senses to be sufficient grounds for doubting the intellect's ability to grasp concepts and forms. Further while sense experience is a part of the beginning of the representation, the chief part of intellection is the formal analysis rather than observation via sense experience.

Further, because these powers are metaphysically necessary being derived from potentialities and actualities, the Humean Critique is simply wrongheaded toward Scholastics(and to be fair, he was not chiefly critiquing them). The fundamental problem with the event causation view doesn't simply lie in the metaphysics but also lies in commonsense analysis.

Hume does have a sort-of response to the idea that causation is metaphysically necessary. He essentially appeals to the fact that we can conceive of effects without causes. I can imagine a watermelon appealing on my desk, therefore it must be possible.

There are many problems with this. One could object to the notion that we cannot imagine impossible things. While it is true that I cannot imagine Logically impossible things(like violation of the law of identity), that doesn't mean that cannot imagine metaphysically impossible things. We can know this from Kripke's analysis of Water as H2O(a posteriori analytic). While I can imagine not being H2O, that doesn't mean that Water could not be H2O. Given the metaphysical necessity of Water(by definition) being H2O, we can conclude that we can imagine metaphysically impossible things.

Furthermore, conceivability is not the same as imaginability. While I can imagine that a watermelon could appear for no reason, I cannot conceive that it could happen because if it did happen, I would never believe it. I would never conclude that a Watermelon simply appeared. Instead, I would investigate and try to figure out why or how it happened. G.E.M. Anscombe pointed this out. 

Of course Hume could deny that causation is in fact metaphysical, but this simply begs the question against the Aristotelian. Further, given all the benefits(that is, it has a great degree of explanatory power) and seems to more easily fit into how we experience the world, it seems that event causation would need more justification than agent causation. Nonetheless, ample justification is given for agent causation via Aristotle's writings. 

When someone says that something caused a certain other thing, they are implicitly appealing to agent causation. In fact, it seems rare for someone to say(if we are talking about simple causation) an event caused anything. Furthermore, it seems like we invoke events when explaining a complex cause(instead of, the fire caused high temperatures which caused the skin to heat which caused the burn which caused the blister, we say, being burned caused the blister). While this may be a useful conversation tool to avoid long strings of agent causation, it does not entail that events actually cause things.

Conclusion

Simply put, events don't really cause things, things that have powers do and thus we have a complete view of causation. The Aristotelian/Scholastic explanation of causation best explains and avoids the Humean critique of causation. The Aristotelian simply denies that the Humean assertion that anything could potentially do anything since for the Aristotelian, things are grounded in their own nature thus makes it metaphysically impossible for things to fail to act according to its nature. 

Furthermore, at least with respect to many reformed versions of Empiricism, it contains a considerably lesser degree of ad hocness since it predates the Critique. This all is without considering the other benefits of abandoning event causation with regards to Philosophy of Science. 

There is a reason that such a critique of causation doesn't exist in Scholastic texts. It simply doesn't have any merit on a Scholastic/Aristotelian worldview. Hopefully this was an informative introduction to the Scholastic answer to the Humean challenge.