the dungeon has...

Is Hume right about causation?

If I am to consider whether Hume’s ideas of causation are correct, I must first discuss exactly what is meant by causation and, indeed, what exactly Hume’s views were on the matter. Causality according to the standard dictionary definition is the relationship between cause and effect, or the principle that everything has a cause. One point to raise then would be whether cause and effect are necessarily connected, could an effect exist without a cause? Cause and effect appear to always be easiest to explain in terms of ball games, if purely for the fact that every movement of the balls could be traced back to an original cause. So, when I am playing Bar Billiards and I cause one ball to hit another, the latter ball would be expected to move due to my action of hitting the first ball into it. But that this is all explainable by science we already know, since examples of the law of conservation of energy, or the executive toy “Newton’s Cradle”, come immediately to mind. However here I am simply giving an example of cause and effect.

Hume’s ‘association of ideas’ theory is generally considered to be a causal one, and it seems a good place to begin. The claim is that if both F and G perceptions have always been associated in my past experience then if I have an F impression I cannot help having or expecting a G idea. So if “one particular species of event has always, in all instances, been conjoined with another, we make no longer any scruple of foretelling one upon the appearance of the other”[i] and therefore it could be considered that it is only a matter of experience which gives us the idea of a necessary connection between cause and effect. Hume goes on to say in his Enquiry:

“But there is nothing in a number of instances, different from every single instance, which is supposed to be exactly similar; except only, that after a repetition of similar instances, the mind is carried by habit, upon the appearance of one event, to expect its usual attendant, and to believe that it will exist. This connexion, therefore, which we feel in the mind, this customary transition of the imagination from one object to its usual attendant, is the sentiment or impression from which we form the idea of power or necessary connexion.”[ii]

To take this literally, would in some way seem to entail that Hume thinks that there is no such thing as a necessary connection in nature, and that it is only by the force of habit that we consider cause and effect to be contiguous. Whether this actually is his view shall be considered later.

If we consider first the stance that there is no such thing as a necessary connection, we must obviously start to consider how this affects our view of the world. Scientific induction works on the following principle:

“If a large number of As have been observed under a wide variety of conditions, and if all those observed As without exception possessed the property B, then all As have the property B”[iii]

So it goes that whilst using this scientific method (also taking into account the Baconian ideal of the observer having no preconceptions about what he is observing), we see the same occurrence under many different conditions, we could consider it to be a property or a power of the object. However, do any number of instances of a law being fulfilled in the past mean that it will be in the future?

It can be seen that there is a problem with presuming that one thing will always go with another, just because it always has done in the past. This is most clearly illustrated by an analogy given by Bertrand Russell:

“The man who has fed the chicken every day throughout its life at last wrings its neck instead, showing that more refined views as to the uniformity of nature would have been useful to the chicken.”[iv]

The chicken in this example will have experienced being fed first thing in the morning, over many years, under many different circumstances, and may (if such intentionality can be ascribed to chickens) have begun to feel that it was a law of nature. Purely because it had always known things to be this way, it would expect them to go on being the same – although in this example such expectations were misguided.

The above example, however, only serves to highlight, the difficulty we can have in going from a finite number of observations to a universal generalisation or law. One famous example is that of black swans. Millions of swans have been observed in America and Europe and all of them were white, therefore by induction we have proved the statement ‘all swans are white’. However, even after all of these sightings there remains the possibility of an ‘unobserved’ non-white swan. In fact, in Australia there are black swans. So although we can see from the principle of induction that it may make it more likely that all swans are white, it does not make it necessarily so. There remains the difficulty of going from the specific to the general. However the argument could be reformulated as ‘It is probable that all swans are white’, in which case the argument would not be falsified by the observation of black swans.

But does everything reduce therefore to such observation laws? Nelson Goodman suggests otherwise:

“But the fact that a given man in this room is a third son does not increase the credibility of statements asserting that other men now in this room are third sons. Yet… our hypothesis is a generalisation of the evidence statement. The difference is that in…[this]… case, the hypothesis is a merely contingent or accidental generality. Only a statement that is lawlike – regardless of its truth or falsity or its scientific importance – is capable of receiving confirmation from an instance of it; accidental statements are not.”[v]

From this we could consider the difference between law-like and accidental generalisations. The theoretical chicken had made an accidental generalisation based purely on his own experience, considering his feeding to be a universal law. It would still remain however to determine which generalisations are law-like and which are not. An obvious example is that of gravity. It has always worked in a set manner, experienceable under a number of different conditions and appears to be a universal occurrence. It is something so large as a concept that it is hard to consider it not working. However if we consider all the times that it has yet to act on the world, our number of observations reduce to next to nothing, and by sheer probability it seems to become less likely that it will continue to act in the way that we expect. Still, reducing things to probabilities can cause problems. One of my favourite examples is from the play Rosencrantz and Guildenstern are Dead by Tom Stoppard, where two characters have the unusual experience of having tossed a coin a few hundred times and each time it has landed heads. One character remarks:

“A spectacular vindication of the principle that each individual coin spun individually is as likely to come down heads as tails and therefore should cause no surprise each individual time it does”[vi]

So probability can be seen to validate the unusual and reduce our belief in the usual if used in the right way. However it is not the conjunction of probability and induction that is the concern here. One argument for believing that experience based on induction will continue to work is that it always has done so in the past, so it is likely to continue that way also. However this ends up as a circular argument as the justification is also inductive.

Another problem which I shall raise briefly is that of prediction based on inductive evidence. The problem is best illustrated with Nelson Goodman’s (1906) example of grue. This imaginary colour-word is a combination of green and blue dependant on the time that it is examined. Goodman argued that if something is examined before the year 2000 AD and is found to be the colour green it would also correctly conform with a statement that they are grue. This means that we would be correct in inducting that all emeralds are green and also that all emeralds are grue. Therefore, emeralds discovered before the year 2000 will look green, whilst those examined afterwards will look blue. It would be more probable to predict that all emeralds will appear green no matter when they were examined, therefore this illustrates a main problem of inductive reasoning. Not all predictions based on empirical evidence are one hundred percent reliable, as Russell’s chicken demonstrates, but also that there is a possibility of other predictions which will satisfy the evidence and observations.

When Hume talked of a necessary connection certain ambiguities arose. Primarily, the idea of necessary connection can make sense if taken as separate parts of a statement:

“‘Connexion’ makes sense in ‘I have severed my connexion with the Communist Party’, as does ‘necessary’ in ‘Mathematical truths are necessary’; and so the disputed phrase has a complex meaning which is caught by the definition of ‘x is a necessary connexion’ as ‘x is necessary and x is a connexion’.”[vii]

However, to believe that this is what Hume meant is foolish. It is easier to regard it as a complex whole, that it is possible to deny the existence of, then to consider that Hume was even trying to rule out statements of fact. As Jonathan Bennett said: “…Hume is not interested in denying that ‘x is necessarily connected to y’ could be taken to mean, say, that x is a mile away from y.”[viii]

An interesting point which is raised by Galen Strawson is that of Hume’s meaning when he said that our ideas of a necessary connection come through repetition. It could be considered that Hume was not denying the existence of causal connections, but simple making a statement about what we can know. He claims that readers automatically move from Hume’s epistemological claim that:

“(E1) all we can ever know of causation , in the objects, is regular succession
with the positive ontological claim that
(O1) all that causation actually is, in the objects, is regular succession”[ix]

and that from this we would have less of the prediction problems that I have dealt with previously, as it states that although we cannot know for certain that causation exists in itself, there is no evidence to disprove it either. The first claim could be considered to be easily arguably true, however the latter, if true, would raise so many problems after all “…it would take a very stubborn inductivist to put his hand in a fire many times before concluding that fire burns”[x].

Continuing on that line, although we can see that it is “…therefore by EXPERIENCE only, that we can infer the existence of one object from that of another”[xi], it could be considered that we cannot know the existence of causation or the non-existence of a necessary connection. Since semantically all we mean, and all we could manage to mean, by causation is an observable regular succession of objects or occurrences. So, via that, it could be taken that causation is nothing more than regular succession. If one therefore takes the stance that ‘necessary connection’ as a concept is doubtful or at least unprovable, then it would appear that Hume’s discussion on this topic is justifiable. However, if it is claimed that Hume is denying the possible existence of a necessary connection between cause and effect, this would raise problems for everyday life. So, although it may be that all people mean when they talk of causation is a regular occurrence, they would find it difficult to act in the world if they did not expect everything that usually happens to happen. After all, if we did not believe that the sun would rise tomorrow, or that trains would work, we would never get anywhere.

- - - - - - - - - -

[i] David Hume, An Enquiry Concerning Human Understanding, Oxford University Press 1975, p75
[ii] David Hume, An Enquiry Concerning Human Understanding, Oxford University Press 1975, p75
[iii] A.F.Chalmers, What is this thing called Science?, Open University Press 1982, p5
[iv] Bertrand Russell, The Problems of Philosophy, Oxford University Press 1988, p35
[v] Nelson Goodman, Fact Fiction and Forecast, Harvard University Press 1983, p73
[vi] Tom Stoppard, Rosencrantz and Guildenstern are dead, Faber and Faber 1967, p13
[vii] Jonathan Bennett, Locke, Berkeley, Hume, Oxford University Press 1971, p258
[viii] Jonathan Bennett, Locke, Berkeley, Hume, Oxford University Press 1971, p258
[ix] Galen Strawson, The Secret Connexion, Oxford University Press 1989, p10
[x] A.F.Chalmers, What is this thing called Science?, Open University Press 1982, p16
[xi] David Hume, A Treatise Of Human Nature, Oxford University Press 1978, p87

TrackBack URL for this entry: http://bluejoh.com/cgi-bin/mt/mt-tb.cgi/411