Hume's Critique of Causality

Hello! Welcome to our next lesson.

In the previous lesson, we compared Aristotle's argument against an infinite regress of concurrent causes with Leibniz's argument from the Principle of Sufficient Reason. We saw that while both aim to establish a first principle, Leibniz's approach is rooted in the demands of rational explanation, a distinctly modern turn.

Today, we confront a philosopher who takes an even more radical step, launching a critique that threatens to undermine not just the Aristotelian and Leibnizian arguments, but any metaphysical claim to know the ultimate causes of things. Our focus is on the Scottish empiricist David Hume and his assault on the very concept of causation.

Your background in econometrics and statistics has trained you to be highly critical of causal claims, to distinguish correlation from causation, and to model relationships based on observed regularities. Hume's philosophy provides the deep epistemological foundation for this modern scientific and statistical mindset.

Our learning outcome is to analyze David Hume's critique of the empirical basis for the causal principle. This is a pivotal moment in our course, as Hume's arguments directly challenge the metaphysical notion of causal power that is the engine of Aristotle's proof for an Unmoved Mover.

1. Hume's Fork: The Limits of Human Reason

To understand Hume's critique of causation, we must first grasp the core principles of his philosophical project. Hume (1711-1776) was a radical empiricist. For him, all the contents of our mind derive from experience.

Hume's project begins with a fundamental distinction among the contents of the mind and the types of propositions we can formulate.

To summarize and formalize these crucial concepts:

• Impressions and Ideas:

  • Impressions are our direct, vivid sensory and emotional experiences (seeing red, feeling warmth, experiencing anger).
  • Ideas are the fainter copies of impressions we use in thinking and reasoning (remembering the color red, thinking about warmth).
  • The Copy Principle: Every simple idea is a copy of a corresponding simple impression. For an idea to be meaningful, we must be able to trace it back to an original impression. This is Hume's primary tool for interrogating philosophical concepts.

• Hume's Fork: All objects of human reason fall into two categories:

  1. Relations of Ideas: These are truths of logic and mathematics, like "2+2=4" or "all bachelors are unmarried." They are known a priori (by reason alone) and their denial leads to a contradiction. They tell us about the relationship between our concepts, but nothing about the world itself.
  2. Matters of Fact: These are propositions about the world, like "the sun will rise tomorrow" or "aspirin relieves headaches." They are known only a posteriori (through experience), and their denial is never a contradiction. We can always imagine the contrary of any matter of fact.

With this framework, Hume is ready to put the idea of causation to the test. Is our knowledge of cause and effect a "relation of ideas" or a "matter of fact"?

2. The Critical Phase: Deconstructing Causal Necessity

When we say "A causes B", we typically mean more than just "A happened, and then B happened." We imply a necessary connection between A and B; that given A, B must follow. But where does this idea of necessary connection come from? Hume uses his fork to dismantle the traditional answers.

Is Causation a Relation of Ideas?

Could we figure out causes and effects just by reasoning, a priori? Hume's answer is a definitive no.

The core argument is that effects are distinct events from their causes. You cannot deduce the properties of an effect by pure analysis of the cause. If you had never seen water before, you could not deduce from its transparency and liquidity that it could drown you. Since the contrary of any causal claim is conceivable (we can imagine striking a match and it not lighting), the connection cannot be a matter of logical necessity.

So, causal knowledge is not a "relation of ideas." It must, therefore, fall on the other prong of the fork: it must be a "matter of fact" derived from experience.

Can Experience Justify Causal Inference?

This seems obvious, right? We learn that fire burns from experience. But Hume's genius was to show that this is not as simple as it seems. His argument has two stages.

1. We Never Perceive a Causal Power

First, Hume argues that when we observe a causal interaction, we do not actually experience the "power" or "necessary connection" itself.

As the video explains, we see one event follow another, but we never see the "tie" that binds them. All we observe is a constant conjunction of events. After seeing event A followed by event B many times, we begin to call A the cause and B the effect. But the repetition adds nothing new to the objects themselves; it only changes something in our minds.

2. The Problem of Induction

This leads to the deepest part of Hume's critique. If all we have is past experience of constant conjunction, how do we justify extending that experience to the future? Our reasoning looks like this:

1. In all past instances, I have observed that event A (e.g., eating bread) was followed by event B (e.g., being nourished).
2. Therefore, in the next instance, event A will be followed by event B.

Hume asks: what is the justification for this "therefore"? The inference relies on a hidden premise: The future will resemble the past. This is often called the Uniformity Principle.

But how do we justify the Uniformity Principle itself?

• We can't prove it with demonstrative reasoning (as a relation of ideas), because it's not a logical contradiction to imagine the course of nature changing dramatically.
• We can't prove it with probable reasoning (as a matter of fact), because any such argument would have to assume the very principle it's trying to prove. We would be arguing: "In the past, the future has resembled the past, therefore the future will continue to resemble the past." This is viciously circular.

Hume concludes that there is no rational justification for our causal inferences. Reason cannot bridge the gap from past experience to future predictions.

3. The Constructive Phase: A "Sceptical Solution"

Having shown that reason cannot be the source of our belief in causation, Hume offers a "sceptical solution." If not reason, then what? His answer is custom or habit.

Hume's solution is a psychological one.

• After repeated observations of A followed by B, our mind develops a habit or a customary transition to expect B when we experience A.
• This feeling of expectation—this "determination of the mind" to move from the impression of the cause to the idea of the effect—is itself a new kind of impression, an impression of reflection.
• This internal feeling is the origin of our idea of necessary connection.
• Because our minds have a "great propensity to spread itself on external objects," we mistakenly project this internal feeling of necessity onto the world, believing that the connection exists between the objects themselves.

For Hume, then, causation is not a feature of reality that we discover, but a feature of our own psychology that we project onto reality. This leads him to his famous two definitions of a cause:

1. The Philosophical (Objective) Definition: "An object, followed by another, and where all the objects, similar to the first, are followed by objects similar to the second." This definition speaks only of regularity and constant conjunction.
2. The Natural (Subjective) Definition: "An object followed by another, and whose appearance always conveys the thought to that other." This definition refers to the psychological habit of inference in our minds.

For Hume, we have no coherent idea of causation beyond these two definitions.

Conclusion

Hume's analysis is a watershed moment in philosophy. He doesn't deny that we think and act as if causes exist; he argues that the foundation for this belief is not reason, but a fundamental, non-rational instinct of human nature.

Key Takeaways:

• Hume's strict empiricism (the Copy Principle) and his "fork" (relations of ideas vs. matters of fact) provide the tools for his critique.
• He argues that causal reasoning is not justified a priori (by reason) because the contrary is always conceivable.
• It is also not justified a posteriori (by experience) because we never observe a "causal power," only constant conjunction, and any attempt to project past conjunctions into the future relies on a circular justification (the problem of induction).
• Hume's "sceptical solution" is that our belief in causation arises from custom or habit. The idea of "necessary connection" is an internal impression of the mind's expectation, which we project onto the external world.

For our course, the implications are profound. The Aristotelian argument rests on metaphysical principles like "every change is the actualization of a potential by something already actual" and "whatever is moved is moved by another." Hume reduces these grand principles to mere summaries of past experience: "we've observed that changes are preceded by other things" and "we've observed that moving things are preceded by other things." By denying that we have any empirical or rational access to a real "causal power" or "necessity" in the world, Hume attempts to saw off the branch on which the entire Aristotelian proof for the Unmoved Mover sits.

In our next lesson, we will see how Immanuel Kant, famously "awakened from his dogmatic slumber" by Hume, tried to save reason from this devastating critique. Kant accepted Hume's premise that necessity isn't found in experience but argued that it is a structure the mind brings to experience, setting the stage for a very different kind of challenge to the classical arguments for God's existence.