Causality

From Aristotle's four causes and the occasionalists' God to Hume's constant conjunction, Kant's category and Lewis's counterfactuals — the hidden glue of the world.

The question

When we say one thing causes another, what really links them — a power, a habit of mind, or nothing but constant conjunction?

Causality seems the most solid thing in the world — the fire burns, the ball moves the other ball — yet no notion has proved harder to justify. Aristotle multiplied causes into four kinds; medieval thinkers argued whether created things truly have causal powers or whether God alone acts. Hume then delivered the great shock: we never observe a necessary connection, only one thing regularly followed by another, so causation is a habit of expectation projected onto the world. Kant replied that causality is a category the understanding must supply for experience to be possible at all. Later thinkers reduced it to laws, questioned whether physics needs the notion, analysed it through counterfactual dependence, or denied inherent causal power altogether. Together these positions expose the fragile inference on which science and everyday life alike depend.

13 thinkers

Aristotle

384–322 BCE

Ancient Greek (Peripatetic)

To know a thing is to know its causes, of which there are four: the material (what it is made of), the formal (its essence or form), the efficient (what brings it about), and the final (the end for which it exists). Explanation is incomplete until all four are given, and nature acts for ends, not by chance alone.

Physics, Book II; Metaphysics, Book V.

Avicenna (Ibn Sīnā)

980–1037

Islamic Aristotelianism

Avicenna defends a robust causal order flowing necessarily from the Necessary Being: causes truly produce their effects, and a cause must be simultaneous with its effect's existence, sustaining it, not merely preceding it. The whole cosmos is a chain of essential causes ultimately grounded in God, from whom being necessarily emanates.

The Book of Healing (Metaphysics).

Al-Ghazālī

1058–1111

Islamic theology (occasionalism)

'The connection between what is habitually believed to be a cause and an effect is not necessary.' When cotton meets fire it is not the fire that burns but God, who creates the burning on the occasion of contact. What we call causation is only God's customary ordering of events, which He could suspend at will — the ground of the possibility of miracles.

The Incoherence of the Philosophers (Tahāfut al-Falāsifa).

Averroes (Ibn Rushd)

1126–1198

Islamic Aristotelianism

Answering Al-Ghazālī, Averroes defends real, intrinsic causal powers: to deny that fire burns by its own nature is to deny knowledge itself, for science depends on the necessary connection of causes and effects rooted in things' essences. God works through, not instead of, the natures He has created.

The Incoherence of the Incoherence (Tahāfut al-Tahāfut).

Nicolas Malebranche

1638–1715

Rationalism (occasionalism)

Created things are merely 'occasional causes': the real, efficacious cause of every event is God. Since a true cause is one between which and its effect the mind perceives a necessary connection, and we find this only in an omnipotent will, only God's volitions truly bring anything about; bodies and minds are occasions on which He acts.

The Search after Truth (1674–75).

Baruch Spinoza

1632–1677

Rationalism

Everything follows from the necessity of the divine nature as conclusions follow from premises: nothing is contingent, and every event is fully determined by prior causes in an unbroken order. To understand a thing is to grasp it through its cause; there is no free will and no chance, only the necessary self-unfolding of one substance.

Ethics (1677), Part I.

Gottfried Wilhelm Leibniz

1646–1716

Rationalism

By the principle of sufficient reason, nothing happens without a reason why it is so and not otherwise. Yet substances (monads) do not causally interact; each unfolds its own states according to its inner nature, and the appearance of causation is God's pre-established harmony, synchronizing all monads like perfectly coordinated clocks.

Monadology (1714); Discourse on Metaphysics (1686).

David Hume

1711–1776

Empiricism

We never observe a necessary connection between cause and effect — only contiguity, temporal priority, and constant conjunction. The 'power' or 'necessity' we ascribe to causes is a feeling of expectation the mind projects onto the world after repeated experience. Causal inference rests on custom, not reason, and cannot be rationally justified.

A Treatise of Human Nature (1739–40); Enquiry Concerning Human Understanding (1748).

Immanuel Kant

1724–1804

German idealism / critical philosophy

Roused by Hume, Kant argues that causality is not derived from experience but is a pure category of the understanding that the mind applies to make experience possible. 'Every event has a cause' is a synthetic a priori principle: it holds necessarily of all appearances, precisely because the ordered, objective world is constituted by our own cognitive forms.

Critique of Pure Reason (1781), 'Second Analogy of Experience'.

John Stuart Mill

1806–1873

Empiricism / inductive logic

Mill treats causation as invariable, unconditional succession discoverable by experience, and codifies the methods for finding it — agreement, difference, residues, concomitant variations. A cause is the whole assemblage of conditions, positive and negative, that regularly precedes an effect; his methods remain the backbone of experimental inquiry.

A System of Logic (1843).

Bertrand Russell

1872–1970

Analytic philosophy

Russell provocatively argued that the notion of cause is 'a relic of a bygone age', absent from advanced physics, which uses functional relations and differential equations rather than causes and effects. The everyday concept, with its asymmetry and necessity, does not map onto the time-symmetric laws of mature science.

'On the Notion of Cause' (1912).

Nāgārjuna

c. 150–250 CE

Madhyamaka Buddhism

Nāgārjuna subjects causation to a fourfold critique: a thing cannot arise from itself, from another, from both, or from neither. There is no inherently existing cause producing an inherently existing effect; causation is real only as dependent origination — mutually conditioned arising empty of any self-existent causal power.

Mūlamadhyamakakārikā, ch. 1 (Examination of Conditions).

David Lewis

1941–2001

Analytic metaphysics

Lewis analyses causation in terms of counterfactual dependence: c causes e roughly when, had c not occurred, e would not have occurred, cashed out through comparisons across possible worlds. This captures the intuition of causal difference-making while dispensing with mysterious necessary connections in nature.

'Causation' (1973); Counterfactuals (1973).