Home | Philosophy | General

Hume's Problem of Induction
By: Mitchell

Hume’s problem of induction is a problem, which still pervades philosophical thinking today. It is, in essence, a problem of how we may ‘know’ from prior experience, that the past is conformable to the present and future. How, for example, we may know that tomorrow the sun will rise.

This problem flies in the face of common sense, which tells us that we know the sun will rise tomorrow because it always has done, but for the rigorous and logical empiricist philosophy of Hume, this is not enough. Hume wants some form of conclusive proof or principle that can rule over such matters.

The fact that a future event will occur for whatever reason; be it due to its relation to a previous event through causality, or because of its repeated occurrence is a form of knowledge. For Hume, all knowledge falls in to one of two categories. The first is Relations of Ideas; this form of knowledge is logical in its basis and a priori in nature. It depends solely on the laws of logic and the interaction of different ideas, e.g. all bachelors are unmarried men. The second form of knowledge is Matter of Fact; here anything that is empirically found to be true is included. For example, the knowledge that I am sitting at my Apple Mac laptop right now is a matter of fact. The distinction between the two comes in the fact that a relation of ideas form of knowledge cannot logically be the contrary, e.g. bachelors cannot be married women – without entailing logical contradiction. Where as it is quite possible I could be sitting a computer in the college library right now.

If we are to accept Hume’s distinction of the forms of knowledge, where does knowledge of the future, based on the past fit in? It cannot be a relation of ideas because logically the counter of any proposition can be true; it could be that the sun does not rise tomorrow. However, it cannot be a matter of fact because knowledge of the future cannot be obtained empirically. Under these conditions, no such knowledge is possible. Whilst we may always have seen X following Y and whilst every past instance may attest to this pattern, we still cannot know, by Hume’s rules of valid knowledge, that in the next instance X will not follow, but Z will. All we can see is the case in hand, interpreted by previous experience.

In section V of Hume’s An Enquiry Concerning Human Understanding, Hume gives us the following example,
“Suppose a person endowed of the strongest faculties of reason be brought on a sudden into the world; he would observe a continual succession of objects and events following each other; but he would never be able to discover anything farther. He would not be able to reach the idea of cause and effect.

Suppose again, that he acquired more experience and saw objects and events to be constantly conjoined together; what is the consequence? He infers from the existence of one object from the appearance of the other, but never acquires any knowledge of the secret power by which one object produces the other. This inference from experience is Custom or Habit, the great guide of human life.”
It is clear from this quote that Hume feels that our insistence that there is a Uniformity of Nature, namely, an idea that what has been will resemble what will come to be, is Custom or Habit. He concedes that whilst the most rigorous endeavours of philosophy may draw us to conclude that there is in fact no reason for there to be any continuity of experience, custom and habit bring us back to such a belief. As Hume says, again in Section V of the Enquiry,
“We need not fear that this philosophy should ever undermine our reasonings of common life, for, whatever we may conclude, nature will always maintain her rights and prevail in the end over any abstract reasoning.”

But this does not rid us of the problem; all it does is assert that whilst we don’t have any firm grounds to believe, we do nevertheless believe because it is part of our nature. It is, of course, necessary for us to have some concept of induction. Were we not to we would never have survived as a species. Were primitive man unable to quickly learn that fire could burn him alive, or that Savannah predators were keen to consume him, he would never have survived and so it is easy to see that such a faculty is of primary evolutionary importance. But even still, this does not tell us how such a belief is valid and justifiable beyond the realms of daily pragmatism and evolutionary survival.

There are, however, various possible solutions to this problem. In particular the eminent 20th Century philosopher Karl Popper proposed the following,
“…there is no induction, because universal theories are not deducible from singular statements. But they may be refuted by singular statements, since they may clash with descriptions of observable facts.

Moreover, we may speak of better or worse theories in an objective sense even before our theories are put to the test: the better theories are those with the greater content and great explanatory power (both relative to the problems we are trying to solve). And these, I showed, are also better testable theories; and- if they stand up to tests- the better tested theories”
Popper is proposing a new look at the old views on induction, where rather than induction being a process whereby we look at what has happened and posit what will happen in the past, it becomes more of a feedback loop of probability and testability. We form an idea based on the past; the more evidence for it the more concrete the theory. If it is never contradicted then it may as well be true. Of course, by Hume’s reasoning, it is not in fact true, only highly probable and previously always the case. But for our terms, if something always happens then we have no reason to think the contrary, and for practical usability, this is the most important thing. However, if experience shows us it is in fact NOT always the case, the theory becomes weaker. With every counter example it is further weakened and may eventually be discarded.

This gives us a more ‘common sense’ version of induction. Where experience slowly weighs up on the scales of belief. If something continual happens or is continually followed by something else, then the scales tip in the favour of believing it. If counter examples are found, they add to the other pan and may slowly tip the scales towards disbelieving. It seems, then, that induction, in practical terms, depends on evidence gathering, but this can have its own difficulties. Testing a piece of copper wire and seeing that it is conductive adds weight to my theory that copper is conductive, also the fact that if Person X is standing in a room and is an only child weight is added to my theory that all the people in this room are only children. However, there is a distinction in the form of knowledge the evidence is adding to that needs to be taken in to account. In the case of the wire, more testing would add further weight. Yet in the case of the only child scenario, more testing would detract from the theory. One is positing a law concerning conduction and materials, one is simply gaining knowledge of certain people and does nothing to add a law or principle to them, but simply gain knowledge of them specifically.

Here we must be careful. Inductive knowledge must be that form of knowledge, which adds a law or principle to something, not simply gains knowledge of specific instances. If it is the latter then no amount of evidence can reinforce the theory beyond it being particular occurrences. In the former case, however, further evidence will add to the theory and make it all the more attestable.

I feel however that Popper is missing a crucial element of how our everyday inductive faculties function. Whilst seeing over and over again that in winter leaves fall from trees will draw me to conclude that this winter leaves will fall from trees –this view is not held, at least in the first encounter of the evidence, with great volition. But the single occasion of touching an electrified fence or putting my hand in to a flame leaves in my mind a greater force of belief than seeing leaves slowly drop from a tree over a course of weeks. Here, I think, is a crucial point. It is not simply the amount of evidence in favour of a proposition, but also the strength of the evidence. And it is a combined result of the two that leaves us with a belief in the future. Whilst induction may not be provable by Hume’s standards, we have seen it is necessary to our functioning in this world, and looking at it from a psychological and biologically necessary angle, the strength of experience as well as the weight, is an important factor.

Another possible solution to the problem of induction is to relate it to the problem of cause and effect, which are very closely related, if not restatements of the same underlying issue. Cause and Effect, whilst believed in almost universally, has no necessary connection. We believe B will follow A because it always has done previously, but there is nothing impossible about it being followed by C or anything else. As such, how can we hold any true notion of cause an effect, namely, how can we inductively posit that B will follow A?

Here I think an appeal to Immanuel Kant’s Critique of Pure Reason is the answer, in particular Kant’s theory of the synthetic a priori statement. A synthetic a priori statement is one whose terms alone are not sufficient to prove the proposition and is also true by no appeal to empirical evidence. ‘Every event has a cause’ is one such statement, it is true by the analysis of event, that there need be a cause; it is also true by the means of empirical experience. This, I suggest, is one way of looking at the problem of induction. If we take the problem of induction, in essence, to be the problem that we cannot know by any reasonable (according to Hume) means whether the future will be conformable to the past we can accept that Hume would disagree that A causes B and will always be followed by B. However, such a statement is both empirically true and analytically true as it is of the nature of Synthetic a priori. In this way we can see that any statements made about the future can be held to be true if they are synthetic a priori statements. This does not solve the problem entirely but does limit the devastation that Hume’s problem of induction can wreak by positing some forms of knowledge as immune to his problem. However, I do feel that ultimately the problem is one of practicality and whilst we do have no way of knowing whether the future really will follow like the past and whether certain event that have always occurred will continue to occur it is, as I have shown, both practically and biologically necessary that we do possess some such faculty. A faculty ruled by judgement based on weight and force of evidence and a faculty open to being overhauled when evidence calls for it.

Bibliography

Goodman, Nelson – Fact, Fiction and Forecast (The Harvester Press Limited 1979)

Hughes Glyn - The Condensed Edition of Immanuel Kant's Critiques of Pure & Practical Reason
http://www.btinternet.com/~glynhughes/squashed/index.htm

Hume, David – An Enquiry Concerning Human Understanding (Oxford University Press 1999)

Popper, Karl – Unended Quest (Routledge, 2002)

Russel, Bertrand – The Problems of Philosophy (Oxford University Press 2001)

Article Source: http://journal.ilovephilosophy.com

Please Rate this Article

 

# of Ratings = 15 | Rating = 4.3/5

Click the XML Icon Above to Receive General Articles Via RSS!
Additional Articles From - Home | Philosophy | General
Sign Up as an Author.
Submit Articles
Member Login
Top Authors
Most Popular Articles
Submission Guidelines
Ezine Notifications
Article RSS Feeds
New Stuff
Forum
Link to Us
Contact Us

Powered by Article Dashboard