Someone recently bought our

students are currently browsing our notes.


Causation Notes

Philosophy Notes > Epistemology and Metaphysics Notes

This is an extract of our Causation Notes document, which we sell as part of our Epistemology and Metaphysics Notes collection written by the top tier of University Of Oxford students.

The following is a more accessble plain text extract of the PDF sample above, taken from our Epistemology and Metaphysics Notes. Due to the challenges of extracting text from PDFs, it will have odd formatting:

Causation General comments on causation

1. We are looking for a way of expressing the truth conditions of causal relations (C caused E if...) in terms that do not involve causal concepts themselves, so that any definition is non-circular.

2. What are the relata of causation? What kind of things are we looking to demonstrate as related in an account of causal relation? An account of causal relata should tell us their category, role and number - what they are, how many they are, and what they do. The standard view: the causal relata are events, there are two of them, and they are cause and effect. Thus when a cue ball knocks a nine-ball into a pocket, we say that event E , the cue ball hitting the nine-ball at angle x, caused event C, the nine-ball rolling into the pocket. However there seems to be a case for four causal relata instead of two - see below.

3. Complete vs. contributory causes: a contributory cause is an event which is a cause of a given effect, but not necessarily its complete cause at any given time. A complete cause is the sum of all contributory or partial causes of a given effect at a given time; the 'sufficient' cause, which necessitates the occurrence of the effect. Not all contributory causes can be described as 'events': the spark that begins a house fire is not the only cause of the fire, since the presence of flammable gas or oxygen in the room also contributed to the fire, however there being oxygen in a room cannot be considered an 'event'. These are better described as 'standing conditions' or 'states of affairs'.

4. Two required premisses implicit in theories of causation:
? C and E are actual
? C and E are wholly distinct events (so that my raising my arm is not a cause of my raising my hand) Six theories of event causation

1. 2.

3. 4.

5. 6.

Constant conjunction INUS SCA Probabilistic SCA Lewis' 1973 CCA Lewis' revised 2000 CCA

1. Hume's 'constant conjunction' C is a cause of E iff:
? C is an event of type T1 and E is an event of type T2
? Events of type T1 are always followed by events of type T2 So what makes one event the cause of the other is the fact that events of the first kind are 'constantly conjoined' with - which is to say, universally followed by - events of the second kind. Problems

1. Which features of C and E are relevant to determining the type of event involved? We know at least that C is an event of a type which has as a property being the cause of event E, or events of that type, but this brings us back to the circularity we are endeavouring to avoid.

2. The account is too broad: by Hume's account, day causes night and night causes day. Not only are both unpalatable conceptions of causation, but by this account the causation relation is symmetric, and intuitively we consider that it should be asymmetric.

3. The analysis is satisfied by coincidental occurrences - an old lady opens her window every morning at 8.30am,

and this is always temporally succeeded by workers arriving at the factory across the road at 8.45am.

4. Similar to 3, the account is satisfied by epiphenomena - a descent in air pressure is followed by a fall of mercury in a barometer, then later by a storm. By Hume's account, the falling mercury caused the storm.

5. Presupposes generality in causation - couldn't causation be singular?

2. Mackie's INUS condition Take a short-circuit that causes a house fire: we would not want to say the short-circuit was a necessary condition for the fire, since any number of accidents could have caused such a fire. Similarly, we would not want to label it a sufficient condition, since it is possible that the short-circuit may have occurred and not caused a full-blown house fire. Thus there is a set of conditions which when combined with the sufficient but unnecessary instance of the short-circuit constituted a complex condition which was sufficient, but not necessary, for the house fire. The shortcircuit is an insufficient but necessary part of a condition which is itself unnecessary but sufficient for the resulting fire. Mackie suggests that when we talk of one event causing another, it is this kind of condition (what he calls an INUS condition) we often have in mind. The disjunction of all the minimal sufficient conditions of this result constitutes a necessary and sufficient condition: the formula "AB!C or DE!F or G!H!I or..." represents a necessary and sufficient condition for the fire, each of its disjuncts represents a minimal sufficient condition, and each conjunct within each minimal sufficient condition represents an INUS condition. To generalise, we replace the conjunction of terms joined with A with the single term x, and all other minimal sufficient conditions with the single term y, and then define an INUS condition thus: C is an INUS condition of E iff, for some x and some y,
? (Cx or y) is a necessary and sufficient condition of E
? C is not a sufficient condition of E
? x is not a sufficient condition of E. Thus an account of causation is yielded: C is a cause of E iff,
? C is an INUS condition if E
? the x factors are present
? the y disjuncts are absent Note: a) the definition leaves the possibility open that A might be a conjunct in all of the minimal sufficient conditions, in which case A would become a necessary condition of P. Whether necessary in this way or not, A remains either way an INUS condition. b) the requirement that x by itself should not be sufficient ensures that A is not a redundant part of the minimal sufficient condition Ax.


1. Priority - let Cx be a drink emerges from a machine. Let y be no drink emerges from the machine, and coins are returned, and E be coins are fed into the slot. By this account, a drink emerging from the machine causes coins to be fed into the slot, which is naturally unpalatable!

2. Epiphenomena - the INUS conditions account still falls foul of the barometer-storm example.

3. Causal pre-emption - Billy and Suzy are throwing stones at a bottle. Billy's throw is on course to smash the

Buy the full version of these notes or essay plans and more in our Epistemology and Metaphysics Notes.