A CRITICAL APPRAISAL OF THE EMPIRICISM OF DAVID HUME AND THE PROBLEM OF INDUCTION
Background to the Problem
Once the task of the rationalists in establishing “certain knowledge” derived from reasoning seemed set, Locke’s attempt to revive empiricism was necessitated. This allowed him to introduce into the scene the concept of tabula rasa or blank slate upon which experience imprints. This in turn seemed to have established the necessary foundations for Berkeley’s immaterialism to be set so that he showed how Locke’s empiricism could lead to skepticism of the external world and eventually, immaterialism. When David Hume took the chair of the empiricist tradition, his own aim was to show how science was misinformed in their supposition that induction was the principle according to which laws of science could be derived. He sought to show the fault in the reasoning by declaring that induction was only based upon a uniformity in nature, one that is wrongly presupposed to be valid. The conclusion was that Hume became a radical skeptic for we eventually see how his position will lead to saying knowledge cannot be attained.
1.1 The Philosophy of David Hume
Attempts of the continental rationalists to assert with some force of conviction the claim that it was only through reasoning that ultimate knowledge could be derived, as can be seen in the rationalist theories of Descartes, Leibniz and Spinoza. They unanimously, directly or indirectly subscribed to notions of innate concept that helped them define their rationalism. Lockean empiricism then appeared to shatter whatever ‘certain’ foundations they had purportedly built. This was followed by Bishop George Berkeley’s immaterialism. Consequently, David Hume wrote his famous Enquiry that laid bare his own conception of the distinction between knowledge derived from reasoning and experience. Though the foundations for this has been gradually laid since the time of pre-Socratic philosophy, and most especially in Descartes.
When John Locke developed his theory of empiricism, he divided our knowledge as being derived from sensations and reflection[i]. He claims that knowledge from reasoning and the mind came from reflection while sensation had to do with knowledge acquired from the external world through experience[ii]. Berkeley had come in a very controversial manner to undermine Lockean theory of perception. Hume however came also with a very different objective in mind to wit: “David Hume was fed up with the sorts of abstract philosophical systems that had been constructed by Cartesians and other such metaphysicians. He sought to criticize these (Cartesian) systems by demonstrating that they rested upon nothing other than “sophistry and illusion.”[iii] The major point he wanted to argue is that as long as beliefs and philosophies are ungrounded in observation and sensory experience, they must remain little more than superstitious fantasies that has no relation to reality.”[iv] This gives us a hint into his latter conclusion of being an empiricist, since he claims that any philosophy not grounded in observation will not amount to anything. It means in essence that any system that is not built on evidence gotten through the senses should be rendered to the flames he was himself an empiricist after all.
Just like Locke had done, Hume divided the contents of the mind into two, he “…distinguished among the contents of the mind ‘impressions’ and ‘ideas’. The first correspond(s) to what we should call sensations and perceptions, the second to what we should call concepts, or ‘meanings’. When I perceive a horse, I have a particular impression (in this case a visual impression); when I think of a horse, I summon up an idea: this idea belongs to a class which together constitutes the meaning (for me) of the word ‘horse’.”[v] Just like Locke had done, Hume distinguished between these two contents of the mind, for him this distinction lies in their respective ‘force’ or ‘liveliness’. The impression is received through the senses, and is vivid and forceful during the moment of its reception. Idea on the other hand is what remains thereafter, when the liveliness and force of the impression have dwindled. However, Hume also describes ideas as ‘copies’, ‘representations’ and ‘images’ of impressions: they are “the faint images [of impressions] in thinking and reasoning.”[vi]
Hume once again, like Locke made, a distinction between simple and complex ideas, claiming that “all our simple ideas in their first appearance are derived from simple impressions, which are correspondent to them, and which they exactly represent”[vii]. Complex ideas on the other hand are built from simple ideas; so that all ideas can be traced to the impressions from which they were derived. This implied that no term is meaningful unless there is an impression from which its meaning can be inferred. The meaning of everything that can be said consists in its empirical content. Laying the foundation for the famous Humean line… “all ideas in the mind have corresponding sense impressions”[viii]For instance, the notion of complex ideas in Locke can be seen here, whatever obtains in the mind, must have been first in experience, even and including dreams where we see the mind exerting its function in being able to “abstract” several ideas it got from experience.
Hume went ahead to divide all significant propositions into two kinds: empirical and logical. In the first case the empirical derive what meaning they have from experience; in the second case they speak only of the relations between ideas.[ix] In his An Enquiry Concerning Human Understanding the division was articulated thus:
“All the objects of human reason or enquiry may naturally be divided into two kinds, to wit, Relations of Ideas, and Matters of fact. Of the first kind are the sciences of Geometry, Algebra, and Arithmetic...[which are] discoverable by the mere operation of thought...Matters of fact, which are the second object of human reason, are not ascertained in the same manner; nor is our evidence of their truth, however great, of a like nature with the foregoing.”[x]
The famous Humean division between relations of idea and matters of fact is regarded to as ‘Hume’s Fork’[xi] which we shall attempt to describe in detail.
1.1.1 Hume’s Fork
As above mentioned, Hume divided all significant propositions two, while the first was knowledge from reasoning, known as knowledge from relations of ideas in Hume’s philosophy; the other is knowledge from experience known as knowledge from matters of fact.
22.214.171.124 Relation of Ideas
According to Hume, knowledge from relation of ideas is all knowledge that is derived from reasoning. For him, knowledge from relations of ideas is certain, true. It can only be an analysis of the ideas that we have, ideas that must have been gotten from only one other source. “…first objects of reason, are propositions whose truth can be known by simply inspecting the ideas to see if the asserted relation holds. They are truths that are true because of the meanings of and the logical relationships between the ideas”[xii], thus we see that knowledge from the relation of ideas is pertaining to the notion of ‘ideas’, and due to that has to do with the mind, since ideas are the only constituents of the human mind, therefore knowledge from the relation of ideas is knowledge derived the human mind, its demonstrative certainty and truthfulness is only a consequence of its being processed by the mind on receiving it from experience.
Propositions that are used to express knowledge from relations of ideas are not just true, they are said to be necessarily true. The introduction of this notion of ‘necessity’ is because of the fact that the otherwise of a statement from relations of ideas cannot be derived except that we encounter a contradiction, all knowledge from relation of ideas are either necessarily true, or necessarily false once we attempt to derive its otherwise. Another way to express this is to say that the denial of knowledge from relation of ideas will involve a contradiction.
Examples of knowledge from relation of ideas are found in mathematics, algebra, and geometry (what passed for ‘the sciences’ during Hume’s time). Mathematics truthfully asserts that when a proposition is given, and its opposite is assumed, a contradiction is generated. We shall consider examples of knowledge from relation of ideas;
A square has four sides
This is an example of knowledge gotten from the relation of ideas. To claim that a square does not have four sides is to contradict the term.
Every whole is bigger than its part
This is also an example of knowledge from relations of ideas. To claim it’s otherwise is to generate a contradiction. For then we shall be claiming that a whole is not bigger than its part. Other examples include; 2+2=4, every whole is bigger than its parts, etc.
These sorts of statement are absolutely certain, according to Hume. But its certainty still leads us to the conclusion of its inability to tell us absolutely nothing about the state of the world outside of our minds. The truths of mathematics, geometry, and logic are truths that hold only in the abstract world of thought. Nowhere in the actual world of concrete reality will you find perfect circles or triangles, but it will always be certain.[xiii] Which implies that Hume thought since these kinds of statements are ideas and therefore restricted to the mind regardless of their certainty, they would always remain confined to the walls of the mind, they may be true and certain but they have nothing to do with experience and therefore are said to be uninformative as they do not give us any new knowledge of the external world. They only break down (analyze) what was gotten through experience. So, Hume agrees with the rationalists that there are certain, demonstrable truths. But what we find out in his writings is that he differed with them regarding their importance. He believed that “Relations of Ideas” were empty, devoid of information about matters of fact and existence.[xiv] Since they are confined to the mind and cannot refer to things that exist outside of it.
126.96.36.199 Matters of Fact
They are the second objects of human reason. The kinds of statements derivable from here are ‘true’ because they correspond to a direct sense experience. Each matter of fact is contingent; its negation is distinctly conceivable and represents a possibility and therefore does not imply a contradiction. Since dependent upon experience, its negation is not only conceivable but also possible. Stated differently, they are propositions whose ‘truths’ can only be known by experience. By making an inspection of the world to see if they are true or false. In distinction from the relation of ideas therefore, their own truthfulness cannot be derived from a mere linguistic concepts used to express them, experimentation has to be conducted before they can be checked.
As articulated in a historical passage by W.T. Jones, according to him, all our knowledge of what, is called ‘objects’ is merely knowledge of spatiotemporal relations among simple ideas or impressions (plus an act of the imagination). This kind of knowledge according to Jones, Hume called ‘knowledge of matters of fact.’ We can only know that, a particular data is followed by some other data, or is above it, or is below it. We can never know for certain that it as a matter of necessity has to be that way; this is because every fact could always be otherwise. Its contrary being matters of fact is still possible; it can never imply a contradiction. That the sun will not rise tomorrow for instance is no less intelligible a proposition, and implies no more contradiction than the affirmation that it will rise. An attempt to show that it is false therefore will only amount to a vain feat.[xv]
Examples of knowledge from matters of fact include the following;
Today is Saturday
This is an example of a statement from matters of fact. We see that it can be false without any notions of contradiction, today can be Monday without any contradiction being generated.
Heavy objects fall when left unsuspended
This is another example of statements from matters of fact. It is totally dependent upon experience and does not have anything to do with the relations between the concepts involved in the assertion.
1.2 The Problem of Induction
In his distinction between relation of ideas and matters of fact, we see that Hume has shown that relation of ideas though certain, really relies on matters of fact. This is because it is merely an analysis of the relation between concepts which however deductively certain do not inform us of anything more than matters of fact. Matters of fact as explained is based upon experience: experience, of course, of the external world, because this experience is not a priori, it is not seen to be certain. The consequence of this is that we cannot be totally certain of the knowledge we get from experience, not only because its denial does not generate contradiction but also because it is based upon the principle of cause and effect, also known as the “causal principle”.
What the causal principle says is that every event has a cause, and every cause must have an effect. It is from here that matters of fact derives its nature. To justify matters of fact, we would require a justification of the principle of cause and effect. There is something wrong with the principle of induction, for it only assumes a connection between all causes and all effects and from there universal propositions due to this relationship are constructed. A universe is defined. But the principle itself is problematic because it falsely assumes that an event once observed to have occurred will always give rise to a conclusion. Now the conclusion that was resulted into is also falsely assumed to be the conclusion to always be derived whenever the instance that effects the conditions for the cause obtains. This is because matters of fact are not certain like relations of ideas.
In the natural sciences, the scientist observes some data in the external world, notes it down, forms a hypothesis from there, and then proceeds to perform an experiment based on the formulated hypothesis. Now once the experiment is concluded, a conclusion is of course generated so that he makes a general, universal statement that ought to take care of all the instances in which whenever the experiment is formulated, similar conclusions as presently generated will always be generated. For Hume, there is no justification for making such inferences, for all that was observed were just particular instances in the present moment. Nothing in this moments imply that the same conclusions as were presently discovered would be the same as will always be discovered, so there really is no justification for making such an inductive conclusion.
Whenever we perceive, all we perceive is the present moment as presented by experience. But we live our lives with the mind of being able to infer the future from the present. Having noticed in the past that certain conclusions proceed from certain occurrences, we assume that similar conclusions will obtain whenever such occurrences do. The point was the relation between what we experience. For Hume our experience is based upon the principle of cause and effect, that every event is a cause which has an effect, another event that proceeds from it. According to Hume, if we could have noticed an event happening and another supposedly proceeding from it, then we ought to be able to have noticed as well a connection between the same events. Since it is based upon experience, then to claim justification for connecting two events happening after the other, we ought to be able to account for a necessary connection that can be empirically verifiable and connects these two.
1.2.1 The Justification of Induction
Despite the condition of the existence of a necessary connection between cause and effect, there are some other conditions that ought to obtain. The first is the spatiotemporal priority which is the condition that every cause must be prior or come before its effect. It is in its nature to cause the other and thus could not possibly come any other time but before. The second condition is the spatiotemporal contiguity that both events must be close to each other in space, such that we can due to proximity in distance claim that this event caused the other. And the third is the spatiotemporal simultaneity which holds that cause and effect must be close to each other in time, they must occur almost at the same time. One noticeable thing about this criterion is that they are all discoverable through experience, thus kind of justification that we seek for the last arm of the causal principle is the idea of necessary connection.
Now, it has to be a feature of experience, but experience cannot show us this connection for all we observe is just an event ‘causing’ another event with the first three conditions holding in quick succession. This is no sufficient reason therefore to assume that such event will always cause such conclusions. We do not have any empirical evidence of a connection between two events therefore. Indulging us a bit, Hume asks that this connection should be established through relations of ideas. But if such connection is discoverable through the relation then it must be one of certainty as is the character of the relation. This however is not the case, for it is not a necessary truth that the sun will rise tomorrow as the possibility of its not rising is not only conceivable but also possible. Cause and effect “…seem conjoined, but (are) never connected”[xvi] therefore the connection cannot be discovered from either means. If this is so then it implies that it cannot be discovered at all since all we have at our disposal are the two means; relations of ideas and matters of fact.
“At most, it is a habit of our minds. Because it is always possible that the future state of affairs in the world will change, our habit of thinking in terms of cause and effect really says nothing about the world itself. It only tells us about the things that we ourselves anticipate on the basis of what we have already observed about the world around us.”[xvii]
[i] Hume, An Enquiry Concerning Human Understanding, Section IV, Part I, 2nd ed., ed. L.A. Selby- Bigge: Oxford: Oxford University Press, 1902. Pg. 25
[ii] Marmysz, John: The Path of Philosophy: Truth, Wonder and Distress. Wadsworth Cengage Learning Inc Boston, 2012; p. 191