Theories of Causation


· The Activity Theory (Descartes, Malebranche, Locke, Berkeley, Reid)


· The Entailment Theory (Malebranche, A. C. Ewing)


· The Regularity (or Constant Conjunction) Theory (Hume, Mill)


· The Singularist Theory (Miss Anscombe)



The activity and entailment theories are objective, necessitarian theories; Hume’s CC theory is a subjective necessitarian theory; Miss Anscombe’s singularist theory is an objective non-necessitarian theory.


David Hume: Enquiry, Section II, “Of the Origin of Ideas”


What Descartes, Locke, and Berkeley called ideas, Hume calls perceptions.


Hume divides perceptions into impressions and ideas.


Impresssions have a greater degree of force and vivacity than ideas. They are the more lively of our perceptions.


Perceptions can be simple or complex.


Hume’s Copy Principle: (see p. 191, middle of RH column for Hume’s more careful wording) Every simple idea is copied from a corresponding simple impression.


[Not “every idea is a copy of an impression”]



Hume’s two arguments for the copy principle (192-3)



Problem of the missing shade of blue. (p. 192)


David Hume: Enquiry, Section VII, “Of the Idea of Necessary Connexion”


Part I: Hume criticizes rival theories, especially the activity and entailment theories.


Part II: Hume argues for his own CC theory and gives 3 different “definitions” of cause (p. 213)


Hume against the activity theory (especially Malebranche)


· God has to act constantly in the world. (Lack of foresight?) Scientists do not discover causes. Our sensations are caused, not by physical objects, but by “a particular volition of our omnipotent Maker.” (“fairy land”) On M’s version, not even human beings have any active power of their own to will their actions.


· We have no idea of power derived from our experience of willing our own actions. (See the attack on the entailment theory)


Hume against the entailment theory


· The logical argument: for any pair of causally related events, C and E, it is no logical contradiction to suppose that C occurs but E does not follow.


We have no idea of causal power from our experience of volition.


· If we were aware of causal power from willing, then we would know, prior to experience, which organs we can move and which not.


· If we were aware of causal power from willing, then we would know what event in our brain is the immediate effect of a volition, the first link in the chain leading to, say, lifting an arm.


Hume’s analysis of the complex idea of causation


A is the cause of B if and only if:


· Temporal priority: A occurs before B


· Constant Conjunction: A’s are always followed by B’s


· Necessity: a simple idea derived from a subjective feeling, a simple impression, caused by our past experience of the constant conjunction of A’s and B’s



Hume’s three “definitions” of cause (p. 213)


“Similar objects are always conjoined with similar. Of this we have experience. Suitably to this experience, therefore, we may define a cause to be [E1] an object, followed by another, and where all the objects similar to the first are followed by objects similar to the second. Or in other words [E2] where, if the first object had not been, the second never had [i.e., never would have] existed. The appearance of a cause always conveys the mind, by a customary transition, to the idea of the effect. Of this also we have experience. We may, therefore, suitably to this experience, form another definition of cause, and call it, [E3] an object followed by another, and whose appearance always conveys the thought to [causes the thought of?] that other.”


E1: the regularity theory: all A’s are followed by B’s


E2: a very different, much stronger theory: causes are subjunctively necessary for their effects. It’s not just that all A’s are followed by B’s but, more importantly, that if A were not to occur, B would not follow. B could not occur without A.


E3: a subjective account: the regularity theory plus a controversial psychological theory


Thomas Reid’s “day/night” objection to Hume’s regularity theory: not all unbroken sequences are causal sequences


Mill’s attempt to repair Hume’s theory and answer Reid’s objection


E1*: the sequence of B’s following A’s must be not only invariable but unconditional. An unconditional sequence is one that is “invariable under all changes of circumstances.”


If we can imagine A occurring without B, then A is not the cause of B. In other words, Mill requires that a cause be sufficient for its effect.


Logically sufficient?


Causally sufficient?


 Anscombe versus Hume on Causation

Anscombe:Singularist Theory 

Hume: Constant Conjunction, or Regularity Theory

Causal relations can be perceived directly in a single case; they are not inferred from many similar cases.
Causal relations cannot be perceived directly in a single case; they can only be inferred from many similar cases.
Causation is a simple concept (relation) that cannot be analyzed (nor does it need to be). 
Causation is a complex concept (idea) that can be analyzed.
Causes do not necessitate their effects: even though A is the cause of B, A can occur in a particular case without producing B.
Causes necessitate their effects: if A is the cause of B then in all cases, if A occurs, B must also occur.
Causal relations are not instances of universal generalizations
 Causal relations are instances of universal generalizations that admit of no exceptions
Causal relations are fully objective: they exist in nature whether we are aware of them or not.
Causal relations are not fully objective: they involve a crucial element, the idea of necessity, which is purely subjective in origin.  The only objective components of causation are temporal priority and constant conjunction.