In this essay I will be attempting to explain what the necessary connection between cause and effect entails, and whether or not it is logically founded. I will be looking at Plato, Aristotle, Russell, Hume and Kant respectively to set out the problem with causality and what kind of conclusions can be made about it and its necessary connection.
So to begin: what is a cause and an effect, and also what is the necessary connection between them? A cause is typically defined as ‘a person or thing that produces an effect’, (1) while an effect is defined as ‘a change which is a result of an action or other cause’. (2) Necessary is defined as ‘needing to be done or present; essential’. (3) Causality is a threefold concept consisting of:
1. Continuity in space and time.
3. Necessary connection.
Anthony Quinton describes causality as:
‘It is, plainly enough, a relation: a complex, threefold one, composed of contiguity in space and time, succession and necessary connection, Neither contiguity nor succession is, in fact, essential to causation. There can be action at a distance and cause and effect can be simultaneous.’ (4)
So what we mean when we talk of cause and effect is an action or thing which takes place (cause) which then leads to another event or thing (effect). For example: when the trigger of a gun is pulled (cause) the bullet being propelled (or failing to be propelled due to a malfunction) is the effect. While when we say there is a necessary connection between cause and effect, what we mean to say is that when the trigger of the gun was pulled it was necessary that something was to come to be or fail to come to be. Which is then defined as the effect. So the effect is a result of the cause, and without the cause there is no effect. (5)
This connection between cause and effect is often cited as being necessary by laymen, philosophers and scientists. To say that there is no necessary connection between cause and effect seems to fly in the face of our experiences of reality. For example, the Greek philosopher Plato argued in his Timaeus that the world is built with order and regularity. There is a necessity here because of what has to be, thus causality could be explained via necessity and mathematics. (6) While Aristotle in his Physics & Metaphysics says that no events are causeless, while admitting that not all events can have their causes found, because not every event has the sort of regularity which Science requires to find causes. (7)
Corresponding with Plato’s and Aristotle’s view of causality and its necessary connection, the philosopher Bertrand Russell put forward his views in his essay On the notion of cause, with applications to the Free-Will problem. In this essay he says that causality is a relic of bygone ages and Science leads the way forward out of this philosophical minefield. (8) How it does this is not exactly made clear, other than induction and experience. Science’s view is that a brick hitting a window could be either way – a window causing a brick to hit it, or a brick being caused to hit it. ‘The impression from which our idea of necessary connection is derived is not of sensation, but of reflection, that of feeling compelled in expecting the broken window on perception of the brick flying towards it.’ (9) But we should not forget that physics is the representation of the world, and should not be mistaken for the world itself. (10) Russell seems to ignore this and doesn’t consider that our experiences can be incorrect and have founded Science. His reliance on Science is unfounded, as will be pointed out as this essay unfolds.
While Plato’s, Aristotle’s and Russell’s views may make sense, they are not questioning whether cause and effect have a necessary connection, but merely assuming it does by assumed necessity and/or empirical suggestion. They place trust human intuition, in the way the world appears to be and the scientific method respectively. Denying the necessary connection between cause and effect is said to fly in the face of our experience of reality as I have said above. But, this does not mean our views of reality and experience are well founded. David Hume was the philosopher to put it clearest when it came to this problem. (11) He believed that when one experiences a cause and an effect there is no necessary connection which is observable, we merely form the idea by experience. So if one was to experience the same effect after the same cause, then we would assume that cause and effect have a necessary connection. (12) The problem with this, as Hume points out: is that this doesn’t prove anything. When we form the concept of necessary connection between cause and effect we are assuming the universe works in a regular way (13) and that our experiences are reliable. (14)
This is put simply by David Lewis:
‘A latter-day Humean, David Lewis (1941-2001), suggested we understand Hume’s idea as follows. The world is like a vast mosaic of unconnected matters of fact: just one little thing and then another. But when we look at that mosaic, we might be able to discern a pattern. Nothing has made there be a pattern. It is as if the mosaic tiles had been placed in a bucket, shaken up, and then thrown out on the floor. But even if nothing but pure chance has made the tiles land where they have, there still can be a discernible pattern. It could be, as a matter of fact, that whenever there is a blue mosaic tile, there is a red one next to it. Nothing makes this be the case. Nothing compels or necessitates it. It’s just a fact about the mosaic.’ (15)
In order to explain this further I need to go into Hume’s fork. Hume’s fork is the separation of two types of statements:
1. Matters of fact: a posteriori (after experience) – Empirical: The sun will rise tomorrow.
2. Relations of ideas: a priori (before experience) – Noncontradictory: bachelors are unmarried men. Mathematics.
Using Hume’s fork we can see the necessary connection between cause and effect in a new light. The standard account of cause and effect runs like this: A causes B, therefore B has always been seen to follow A. Thus, B will always follow A. This implies that nature is uniform, and there is never enough evidence for that conclusion because it is a matter of fact – a posteriori problem. It cannot be justified by logic, since there is no contradiction in saying that A does not cause B, and so on. And it can’t be proved from experience, but merely suggested, because we cannot witness every event to ever say that logically there is a connection between cause and effect: ‘One event follows; but we never can observe any tie between them. They seem conjoined, but never connected.’, (16) never mind say there is a necessary connection. Regularity fails to reveal any necessity in causation, and also fails to reveal any necessity between cause and effect. Our inductive method is flawed, and we form a notion of causality by induction. The inductive method is flawed as it relies on itself to prove itself practically, but cannot logically. Making a blind faith in Science; like the one expressed by Russell; a logical problem. Science is founded on probability, regularity and induction, while Hume is directly attacking this.
Ayer makes this clear in his Hume: A very short introduction:
‘That the principle cannot be demonstrated follows evidently from the assumption, which we have seen that Hume is entitled to make, that the members of the two classes of instances are logically distinct. As for its being probable, we run up against the fact that the ascription of probability draws its force from past experience. In Hume’s words, ‘probability is founded on the presumption of a resemblance betwixt those objects, of which we have had experience, and those, of which we have had none, and therefore ’tis impossible this presumption can arise from probability’ (T 90).’ (17)
Working from Hume’s Fork was the philosopher Immanuel Kant. He altered Hume’s fork as such:
1. a posteriori – Same as Hume’s.
2. a priori – Same as Hume’s.
3. Synthetic – arithmetic – requires a priori concepts, but also requires experience to figure out. = Cats (concept) are all black (requires experience to prove.)
Kant’s example: 7 + 5 = 12 – requires working out to know. 12 does not contain 7 and/or 5 in the definition. While 7 contains 5, 5 does not contain 7. We need experience to work out this equation.
A priori propositions follow the principle of contradiction: that is to say that they cannot be true if it contradicts itself and its definition. (18) All a posteriori propositions are synthetic. The distinction between a priori and synthetic is that an a priori proposition is decided by the principle of contradiction. While a synthetic judgement is not. It is decided by experience: how the proposition relates to the world. (19)
For Kant the necessary connection between cause and effect was a human intuition given to us by experience of reality. (20) In this sense causality is objective, while our experience of it is synthetic. Thus, causality is an a priori notion, but is only known a posteriori (experience). Meaning that causality is a synthetic fact which stems from an a priori fact combined with the experience of cause and effect. So, for Kant cause and effect does have an objective necessary connection. (21)
If causality is a human intuition placed on reality to make sense of it, and thus translates reality into a human experience: could the causal laws and laws of nature commonly known as physics merely be reduced to a human intuition and translation of reality so one can understand it? If this is possible then this would reduce causality and the necessary connection between cause and effect to a human matter of fact: an unprovable concept, which is only ever suggested by experience, but never proved. Which leaves it open for the future to disprove it by failing to proceed in a regular pattern which fits our idea of causality. Or would this merely reflect our ignorance of this universe and how it works? Ducasse expresses the problem with causality as a category in Kantian terms:
‘[I]t is admitted to be true that “experience is a fact,” is it in that case still true that without causation as a category, no experience is possible? I have in the preceding pages tried to show that neither Kant nor Professor Kemp Smith has succeeded in proving that it is true. And it cannot be proved true for the same reason as in the other case, namely, the meaning of “experience” in which the second premise is admittedly true is obtainable inductively only (by analysing the characters of whatever is called experience in the cases in which the proposition is admitted to be true), and therefore that meaning is never statable otherwise than as a provisional description, theoretically always liable to modification. So that theoretically, we never can know for certain what the proposition “Without causation as a category, no experience is possible” strictly means, nor therefore prove it.’ (22)
The problem here is two fold:
1. We could well be correct to think causality exists, but we shall never prove that until every experience possible has been experienced with no other experiences left to have. Which is never going to happen.
2. We could be placing a human category onto reality to make sense of it, while causality doesn’t exist as anything other than a human construct and our way of thinking projected onto reality and reflected back at us.
Either way, we have no logical foundation for thinking that cause and effect have a necessary connection. Theoretically cause and effect have no proved necessary connection. However, practically it does have a connection, although not necessary: i.e. via induction from experience. This is theoretically flawed and contains no certainties. In this sense Hume was correct in his criticisms of cause and effect, and its necessary connection. While Kant made a respectable effort to try and find a practical proof to further legitimise the necessary connection. Although this failed, it was a respectable attempt and pursuit: one that was also an endeavour to legitimise Science not by theoretical foundations in logic, but by practical application of experience and the assumption of regularity which Kant believed stemmed from an a priori knowledge of concepts. Which were then proved via experience resulting in synthetic knowledge based on the objective and necessary. (23) Thus if successful would give Science a metaphysical foundation making it a legitimate system both theoretically and practically. However, Kant is unable to overcome Hume’s criticisms (24) which means he cannot make its theoretical foundation solid, but does leave us with a practical application of cause and effect and its concept. Although, this still leaves the necessary connection of cause and effect as a redundant feature of the concept of causality. This is to say that maybe there is a necessary connection between a cause and its effects, but this necessary connection can only be practically applied at this time for it lacks the logical proof to become a legitimate feature of cause and effect. (25) As Hume had made clear in his own time: trying to prove cause and effect’s necessary connection with a methodology which requires and relies on cause and effect to function is somewhat absurd.
1. Oxford English dictionary. Edited by Maurice Waite. (Seventh edition, Oxford university press, 2012.)
4. Ray Monk & Frederic Raphael. The great philosophers: From Socrates to Turing. Anthony Quinton, Hume. Causation. (A Phoenix paperback, 2001.) Page 229.
5. ‘A part of the common concept of causality is that the cause and the effect are necessarily connected—that if the cause occurs, the effect must occur; it cannot but occur.’ – John Hospers. An introduction to philosophical analysis. 5, What is and what must be. Freedom and necessity. 3. Causality. 3. Necessary connection. (Routledge, fourth edition, 2007.) Page 142.
6. Julia Annas. Plato: A very short introduction. Chapter 7, The nature of things. Chaos and order. (Oxford university press, 2003.) Pages 77-78.
7. ‘Aristotle does not admit causeless events into the natural world. But he does allow that not all events are amenable to scientific knowledge; for not everything exhibits the sort of regularity which science requires.’ – Jonathan Barnes. Aristotle: A very short introduction. Chapter 12, Causes. (Oxford university press, 2000.) page 91.
8. ‘The causal conceptualization of the world is thus an ignorant and prescientific one and, in a famous passage, Russell says: ‘The law of causality, I believe, like much that passes muster among philosophers, is a relic of a bygone age, surviving, like the monarchy, only because it is erroneously supposed to do no harm.” – Stephen Mumford & Rani Lill Anjum. Causation: A very short introduction. Chapter 1, The problem: what’s the matter with causation? Bertrand Russell flexes his muscle. (Oxford university press, 2013.) Page 11.
9. Ray Monk & Frederic Raphael. The great philosophers: From Socrates to Turing. Anthony Quinton, Hume. Causation. (A Phoenix paperback, 2001.) Page 231.
10. ‘The world is not a number, nor an equation. It is a concrete particular inhabited by physical objects and some of them appear to be causally related to others. Physics sometimes forces us to rethink and revise common sense, which may be perfectly legitimate. But it should not follow automatically that because a theory works out mathematically, within a model, the world is exactly like that model or like the maths.’ – Stephen Mumford & Rani Lill Anjum. Causation: A very short introduction. Chapter 1, The problem: what’s the matter with causation? Metaphysics and better physics. (Oxford university press, 2013.) Pages 12-13.
11. ‘Underlying the view that rationalizations were not causal explanations and that reasons were not causes of actions was the Humean view of causation, according to which cause and effect are discrete, separate ‘objects’, so that their relation as cause and effect derives merely from the fact that they instantiate a regularity or law of nature.’ – Carlos J. Moya, The philosophy of action: an introduction. 10, Davidson’s causal theory of intentional action. (Basil Blackwell Ltd, Polity press, 1990.) Pages 105-106.
12. ‘[A]s Humeans insist, causation doesn’t involve necessity but allows complete contingency.’ – Stephen Mumford & Rani Lill Anjum. Causation: A very short introduction. Chapter 4, Necessity: do causes guarantee their effects? Free will. (Oxford university press, 2013.) Page 45.
13. ‘If none of us can see causal connections, why then do we believe that they are real? Hume had a view on this. The main reason why we think that the first event caused the second is that it is part of a pattern.’ – Stephen Mumford. Metaphysics: a very short introduction. Chapter 5, What is a cause? Fighting the cause. (Oxford university press, 2012.) Page 47.
14. ‘Hume’s argument concerning evidence runs like this:
I see something happen several times.
I therefore expect it to happen again.
I get into the mental habit of expecting it to happen.
I may be tempted to project this mental habit out onto the external world in the form of a ‘law’ of physics.
So, for example, ‘A causes B’ could be taken to mean ‘B has always been seen to follow A’.’ – Mel Thompson. Teach yourself: Philosophy. 01, the theory of knowledge. Hume (1711-76). (Of the teach yourself series, http://www.teachyourself.com, 2006.) Page 28.
15. Stephen Mumford & Rani Lill Anjum. Causation: A very short introduction. Chapter 2, Regularity: causation without connection? Playing billiards with Hume. (Oxford university press, 2013.) Page 18.
16. David Hume. Enquiries concerning Human Understanding and concerning the Principles of Morals. Section VII. Of the idea of necessary connexion. Part II. (Oxford university press, third edition, 1985.) Page 74.
17. A. J. Ayer. Hume: A very short introduction. Chapter 4, Cause and effect. (Oxford university press, 2000.) Pages 85-86.
18. ‘Contrast these two judgements (1) ‘all effects are caused’ and (2) ‘all events are caused’. The first is analytic and is true a priori, merely by virtue of the meaning of the words. By definitions, to be an ‘effect is nothing other than to be caused’. The statement is necessarily true, but it is also trivially true. It is an analytic a prior judgement.’ – Robert Wicks. Kant: A complete introduction. Section two: What can we know. 2. Kant’s theory of judgement: questioning the foundations of empiricism. (Teach yourself series, 2014.) Pages 41-2.
19. ‘Kant argued that certain features of experience, including space, time and causality, were not in themselves features of the external world, but were imposed by the mind on experience. This was a revolutionary way of looking at the theory of knowledge and at metaphysics. Take the example of time. When I see a sequence of things, I say that time is passing and that one thing follows another. But where is that time? Is it something that exists ‘out there’ to be seen? Is time there to be discovered? Kant argued that time was one of the ways in which the mind organizes its experiences; it is part of our mental apparatus.’ – Mel Thompson. Teach yourself: Philosophy. 01, the theory of knowledge. Kant (1724-1804). (Of the teach yourself series, http://www.teachyourself.com, 2006.) Page 29-30.
20. ‘Now the judgement, ‘all events are caused’ is another matter. This judgement is thought-provoking. To be an ‘event’ does not logically entail being ’caused’, for although it is remote and virtually unimaginable, it is not contradictory to assert that something could happen spontaneously, ‘out of nowhere’. For Kantian reasons that we will later see, every human being has a difficult time imagining this. Nonetheless, admitting the mere possibility of an ‘uncaused event’ is not the same as proposing an idea like ’round square’, ‘wooden iron’, or ‘reddish green’. The idea of ‘uncaused effect’ may be a contradiction, but that of an ‘uncaused event’ is not.’ – Robert Wicks. Kant: A complete introduction. Section two: What can we know. 2. Kant’s theory of judgement: questioning the foundations of empiricism. (Teach yourself series, 2014.) Page 42.
21. ‘We often judge that an event, say A (or of type A), causes another event, say B (or of type B). We do, in other words, apply the notion ’causes’ to what we perceive. But do we also abstract this notion from our perceptions? True, we perceive that A followed by B and that they are spatially and temporally near to each other. We may remember that A has so far always been followed by B and we may even have developed a habit of passing from the thought of A to the thought of B. All this, Kant holds, does not add up to the notion of causality, which is a kind of necessary connexion. It is, according to Kant, a general notion which, although applicable, is not abstracted from perception. It is an a priori concept.’ – S. Körner. Kant. Chapter one, The plan of the critical philosophy. 4. Sense, understanding, reason. (A Pelican original, 1967 reprint of the 1955 original.) Pages 28-29.
22. Curt John Ducasse. Causation and the types of necessity. Part I, Causation-Critical. Chapter III. Kant and Professor Kemp Smith. (Dover publications, Inc., New York. 1969.) Pages 35-6.
23. ‘The general conditions of objective experiences include not only its a priori conditions but also the empirical laws of nature. What follows from them is causally but not logically necessary. That all men must die is causally necessary because deducible from the empirical laws of nature; but it is not logically necessary, since the negation of it, the statement that there exists an immortal man – although false – is not self-contradictory.’ – Körner, S.. Kant. Chapter four, The system of synthetic A priori principles. 3. The synthetic A priori principles of the understanding. (A Pelican original, 1967 reprint of the 1955 original.) Pages 90-91.
24. ‘[A]ccording to Hume’s principles, there can be no question of discovering some method for assigning probabilities, which is logically certain to give correct results. Such a method demands a priori knowledge of principles of uniformity, the possibility of which Hume has already denied in his treatment of causal inference.’ – A.H. Basson. David Hume. 4. The understanding. (A Pelican book, 1958.) Page 80.
25. ‘The only explanation seems to be that the propositions that ‘every event has a cause’ and that ‘the course of nature continues always uniformly the same’ were regarded by Hume in the light of natural beliefs. They cannot be proved, but nature is so constituted that we cannot avoid accepting them.’ – A. J. Ayer. Hume: A very short introduction. Chapter 4, Cause and effect. (Oxford university press, 2000.) Pages 87.
– Annas, Julia. Plato: A very short introduction. (Oxford university press, 2003.)
– Aristotle. Physics.
– Aristotle. Metaphysics.
– Ayer, A. J.. Hume: A very short introduction. (Oxford university press, 2000.)
– Barnes, Jonathan. Aristotle: A very short introduction. (Oxford university press, 2000.)
– Basson, A.H.. David Hume. (A Pelican book, 1958.)
– Russell, Bertrand. On the notion of cause, with applications to the Free-Will problem. Taken from: Readings in the philosophy of science. On the site: http://www.aristoteliansociety.org.uk (New York, Appleton-Century-Crofts, Inc.)
– Ducasse, Curt John. Causation and the types of necessity. (Dover publications, Inc., New York. 1969.)
– Harré, R. & Madden, E.H.. Causal powers. (Basil Blackwell, Oxford 1975.)
– Hospers, John. An introduction to philosophical analysis. (Routledge, fourth edition, 2007.)
– Hume, David. An Enquiry concerning Human Understanding. (Oxford world’s classics, 2008 reissue.)
– Hume, David. Enquiries concerning Human Understanding and concerning the Principles of Morals. (Oxford university press, third edition, 1985.)
– Hume, David. Treatise of human nature. (1738.)
– Kant, Immanuel. Critique of pure reason. (Palgrave Macmillan, translated by Norman Kemp Smith, revised second edition ((With new introduction by Howard Caygill) published 2003.)
– Kant, Immanuel. Kant’s Prolegomena and Metaphysical foundations of natural science. (Translated by Ernest Belfort Bax. Second edition revised. Bohn’s Philosophical library, 1891.)
– Körner, S.. Kant. (A Pelican original, 1967 reprint of the 1955 original.)
– Monk, Ray & Raphael, Frederic. The great philosophers: From Socrates to Turing. (A Phoenix paperback, 2001.)
– Moya, Carlos J.. The philosophy of action: an introduction. (Basil Blackwell Ltd, Polity press, 1990.)
– Mumford, Stephen. Metaphysics: a very short introduction. (Oxford university press, 2012.)
– Mumford, Stephen & Anjum, Rani Lill. Causation: A very short introduction. (Oxford university press, 2013.)
– Oxford English dictionary. Edited by Maurice Waite. (Seventh edition, Oxford university press, 2012.)
– Plato. Timaeus.
– Price, H. H.. Hume’s theory of the external world. (Oxford university press, 1940.)
– Thompson, Mel. Teach yourself: Philosophy. (Of the teach yourself series, http://www.teachyourself.com, 2006.)
– Wicks, Robert. Kant: A complete introduction. (Teach yourself series, 2014.)