Humean Causality and Presuppositionalism

David Hume’s Understanding of Causality

Back when I was explaining to Chris Bolt the importance of an objective approach to induction, I pointed out one of David Hume’s key flaws in formulating his skeptical conclusion about induction. Specifically, I explained that Hume’s conception of causality as a relationship between “events,” as opposed to a relationship between an entity and its own actions, is a crucial premise in his infamous argument for skepticism in induction (see for instance here and here). The reason for this should be simple to understand: on Hume’s conception, there is no inherently necessary relationship which connects one event to another which happens to follow it.

As Hume himself put it:

All events seem entirely loose and separate. One event follows another; but we never can observe any tie between them. They seem conjoined, but never connected. (An Enquiry Concerning Human Understanding)

Hume’s conception of causality, as expressed here, essentially guarantees a skeptical analysis of induction. (Hume made other mistakes as well, and these simply doomed the Scottish thinker’s understanding of induction all the more.) Since “all events seem entirely loose and separate,” and causality is understood essentially as a succession of events - one called cause, and the one which happens to follow it called effect - it’s just a happy accident that touching the hot surface of a stove (the “cause”) results in the experience of pain (the “effect”). On this view, touching the hot surface of a backyard grill or molten iron could just as easily result in pleasure, a musical performance, paper production, or anything else, since they “seem… never connected.”

The implications which such a view of causality has for induction are hard to miss. For instance, since on Hume’s view “all events seem entirely loose and separate” and “we never can observe any tie between them,” a single instance of one event following another is never sufficient to give us knowledge of a necessary connection between two events such that we can be assured that they are causally related (or “connected,” as Hume puts it). In other words, for Hume, observing one instance of cause and effect is insufficient to validate what some have come to call “the inductive principle,” that is, “the principle that future unobserved instances will resemble past observed instances” (James Anderson, Secular Responses to the Problem of Induction; cf. also Brian Knapp, “Induction and the Unbeliever,” The Portable Presuppositionalist, pp. 121-122). Hume tells us this explicitly when he states:

When any natural object or event is presented, it is impossible for us, by any sagacity or penetration, to discover, or even conjecture, without experience, what event will result from it, or to carry our foresight beyond that object which is immediately present to the memory and senses. Even after one instance or experiment where we have observed a particular event to follow upon another, we are not entitled to form a general rule, or foretell what will happen in like cases; it being justly esteemed an unpardonable temerity to judge of the whole course of nature from one single experiment, however accurate or certain. (Op. cit.)

Why is “one instance or experiment” not enough for us “to form a general rule, or foretell what will happen in like cases”? Because, according to Hume, “all events seem entirely loose and separate” and, he says, “we can never observe any tie between them.” This seems strange to me, as the pain that I experience when I touch a hot stove top does seem very much “connected” to my touching the hot stove top in the first place. But Hume says that “we never can observe any tie between” the one and the other. Apparently we’re supposed to take his word for this.

But Hume doesn’t think we’re entirely lost on this. There is a way around this, as he explains:

But when 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, and of employing that reasoning, which can alone assure us of any matter of fact or existence. (Ibid.)

So touching a hot stove top and experiencing pain only once is not sufficient, says Hume, to draw the general inference that touching it again will result in the same experience. But, he says, if this happens “always, in all instances,” there is apparently no objection against considering such a rule. In other words, in order to formulate the general rule that touching a hot stove will cause pain, we need to know that in “all instances” of touching a hot stove – that is, not 50%, not 75%, not 98% of the time, but 100% of the time – the one touching it experiences pain.

The question at this point becomes: how do we know when we have knowledge of “all instances” of “one particular species of event”? Since we are not omniscient, it can reasonably be supposed there will always be the possibility that some instances of a particular species of event lie outside of our knowledge, and among them may be instances which do not resemble the instances which we do know about. Hume gives no indication as to how one could confidently be assured that he has “all instances” to work with, or how one can reliably conclude that there are no instances outside of his knowledge which defy the norm of those which he does know. Since no one individual has knowledge of “all instances” of touching a hot stove, Hume is essentially saying that no one meets the minimum requirements needed to conclude as a general rule that touching a hot stove will cause pain.

Hume’s own terms arbitrarily require omniscience as a standard, even though there is no such thing as omniscience. But without it, his conception of causality dooms us to the skepticism for which he is so famous.

To make matters worse, Hume seemed to believe that the imagination serves as some kind of final court of appeal in settling the difficulty. In his search for a “connexion” between one event (the cause) and another (the effect), Hume explicitly pointed to what one “feels” in his imagination, and only when repetition of the conjoined events in succession is involved:

It appears, then, that this idea of a necessary connexion among events arises from a number of similar instances which occur of the constant conjunction of these events; nor can that idea ever be suggested by any one of these instances, surveyed in all possible lights and positions. 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. Nothing farther is in the case. Contemplate the subject on all sides; you will never find any other origin of that idea. This is the sole difference between one instance, from which we can never receive the idea of connexion, and a number of similar instances, by which it is suggested. The first time a man saw the communication of motion by impulse, as by the shock of two billiard balls, he could not pronounce that the one event was connected: but only that it was conjoined with the other. After he has observed several instances of this nature, he then pronounces them to be connected. What alteration has happened to give rise to this new idea of connexion? Nothing but that he now feels these events to be connected in his imagination, and can readily foretell the existence of one from the appearance of the other. (Ibid.)

Essentially, Hume is saying here that, without a sufficient number of iterations (Hume indicates no specific quantity here), the imagination does not have enough input from the world to consider one event “conjoined” with another, to be “connected” – i.e., to be causally connected. If you touch your finger to a hot stove and feel the pain of a burn one time, that’s not sufficient to attribute a causal connection between the one action of touching your finger to a hot stove and feeling pain. For Hume, you have to do this numerous times (how many, he does not say) for you to finally “get it” (or “feel”) in your imagination that touching a hot surface and experiencing pain are causally “connected.”

To make sure we understand this correctly, Hume states it again as follows:

In all single instances of the operation of bodies or minds, there is nothing that produces any impression, nor consequently suggest any idea of power or necessary connexion. But when many uniform instances appear, and the same object is always followed by the same event; we then begin to entertain the notion of cause and connexion. We then feel a new sentiment or impression, to wit, a customary connexion in the thought or imagination between object and its usual attendant; and this sentiment is the original of that idea which we seek for. (Ibid.)

Streminger interprets Hume accordingly:

Since we never perceive a propter hoc, but only a post hoc, causal connections have to depend on the creative imagination. (Hume’s Theory of Imagination)

Notice what is being affirmed here: that something metaphysical (“causal connections”) depends on what the imagination creates. The role of man’s imagination, says Streminger (interpreting Hume), is “to structure the stream of [one’s] perceptions” (Ibid.). Presumably, a “stream” of perceptions has no structure apart from what the imagination imposes on them. Or, they do have a natural structure, but, perhaps, the structure provided by the imagination has greater epistemological import as a faculty by which those perceptions are assembled into content amenable to inductive inference. Either way, the sequence here seems to be perception plus imagination yields the material needed to generate inductive conclusions. Causal connections in themselves are ultimately subjective phenomena on Hume’s view. As John Vickers puts it, “the objectivity of causality, the objective support of inductive inference, is thus an illusion” (The Problem of Induction, from The Stanford Encyclopedia of Philosophy).

In this way, I wager that David Kelley is correct in his assessment of Hume when he interprets Hume to be saying (note that Kelley is not presenting the Objectivist view here):

The mark of an axiom, of a self-evident truth, is that we cannot deny it without contradicting ourselves. And the test of whether a proposition is contradictory, is whether we can imagine it. We cannot imagine a case where A equals not A. That’s why the law of identity, A is A, is self-evident. It cannot be denied without contradiction; we cannot imagine it being false. But we can imagine that fire soothes on Monday, Wednesday and Friday, and burns on Tuesday, Thursday and Friday, Sunday being a day of metaphysical rest. The law of causality can be denied without self-contradiction, so it cannot be an axiom. And even if it were an axiom, finally, that wouldn’t do us any good. Axioms, says Hume, are merely relations among our own ideas. They are arbitrary constructs that we make true by the way we define our own terms. And therefore they don’t tell us anything about reality. (Universals and Induction)

To be sure, Hume held some seriously defective views. To summarize, Hume’s view of causality entails the following points:

1. Causality is a relationship between “events”

2. There is no inherent necessity connecting one event to another (they “seem entirely loose and separate” and “we can never observe any tie between them”)

3. A single instance is never sufficient to inform a general principle

4. Exhaustive knowledge of all instances is needed to form a general principle (repetition)

5. The “necessity” of cause and effect relationships is ultimately grounded in the imagination

Given underlying assumptions as strange as these, it should be no wonder why Hume could only come to the conclusion about induction for which he is so well known. Indeed, it could rightly be said that the problem of induction is the problem of seeking an adequate justification for inductive generalization when one’s conception of causality is so expressly geared against it. Absent from Hume’s analysis, so far that I’ve been able to find, is a defense of the view that causality is a relationship of events, or of the premise that “all events seem entirely loose and separate” and that “we can never observe any tie between them.” And yet, as should be clear, these are key assumptions in Hume’s skeptical argument.

Hume and Presuppositionalism

Now it is important to note that, when presuppositionalists raise the problem of induction in their apologetic challenges against non-believers, they very typically cite Hume as an authority on induction. And in doing so, they express no concern for or dissatisfaction with Hume’s conception of causality which underlies his skeptical conclusion about the nature of inductive inference. On the contrary, presuppositionalists generally point to Hume as if he were an unquestionable authority on the matter. In this way presuppositionalists tacitly endorse the premises in Hume’s argument, among them, sadly, his conception of causality.

When engaging presuppositionalists who attempt to raise the problem of induction as a debating point, it is a good idea to ask them whether or not they think Hume’s argument for inductive skepticism is sound. This will answer many questions at the outset. For instance, if the apologist concedes that Hume’s argument is not sound, then he should acknowledge the fact that raising the issue to begin with is futile. An argument admitted to be unsound is not one which needs to be taken seriously. If the apologist insists that Hume’s argument is sound, we should invite him to reconstruct the argument and argue for its premises. Moreover, if he truly thinks Hume’s argument is sound, then he needs to live with its skeptical conclusion, and this would be counterproductive to his apologetic agenda.

In my experience, however, apologists who invoke the inductive version of presuppositionalism typically resist stating for the record whether they think Hume’s argument is sound or not. (See for instance my 08 Oct. 2009 comment in this blog and my exchanges with Paul Manata in the comments section of this blog.) This is quite telling, for it suggests that apologists who raise the issue are unwilling to take much of a stand on it. It becomes all the more revealing when it is pointed out that Hume’s conception of causality is faulty, and apologists subsequently distance themselves even further from taking a stand.

Swallowing Hume’s conclusion about induction in whole without examining his premises seems to be commonplace among presuppositionalists, even in the case of those who should know better. Greg Bahnsen, for instance, who holds that “the causal principle is seen to be intelligible only within the Christian framework of thought,” cites Hume as an “unbeliever” who was “both brilliant and honest about the matter,” pointing to Hume’s “devastating critique of causal reasoning in An Inquiry Concerning Human Understanding” as “the foremost example of this” (Van Til’s Apologetic: Readings & Analysis, pp. 618-619, including 619n.143). Hume is said by Bahnsen to number among those “unbelievers” who “have openly conceded that they have no rational basis for believing that the future will resemble the past” (Ibid., p. 619). Bahnsen summarizes the essence of the Humean concession as follows:

We may have observed that event B followed event A many times in the past, but to know that B necessarily follows A (i.e., that the relation is causal), calls for reference to a metaphysical principle (namely, the future will be like the past) for which the unbeliever has no warrant or right. As Bertrand Russell was driven to conclude: “The general principles of science, such as the believe in the reign of law, and the belief that every event must have a cause, are as completely dependent upon the inductive principle as are the beliefs of daily life. All such general principles are believed because mankind have found innumerable instances of their truth and no instances of their falsehood. But this affords no evidence for their truth in the future, unless the inductive principle is assumed.” Assumed? But that is what was supposed to be proved! Russell was aware of his defeat: “Hence we can never use experience to prove the inductive principle without begging the question. Thus we must… forgo all justification of our expectations about the future.” (Ibid.; quoting Russell’s “On Induction,” in The Problems of Philosophy (1912; reprint, London: Oxford University Press, 1959), 69, 68.)

What is noteworthy here is that neither Bahnsen nor Russell ever seems to question Hume’s conception of causality, an element which is integral to the skeptical conclusion for which Hume is so famous.

Similarly, in his essay “Induction and the Unbeliever” (The Portable Presuppositionalist, pp. 118-142), presuppositional apologist Brian Knapp cites Hume as one raising the question of how we can justify induction (cf. pp. 122-123), but never questions the premises underlying Hume’s argument for the view that “induction is hence unjustifiable” (p. 129, citing this Stanford Encyclopedia of Philosophy article). Nor does Knapp express any concern for the particular understanding of causality assumed by Hume in framing the famous problem.

So the tendency to miss the significance of Hume’s conception of causality in generating his skeptical conclusion in regard to induction does not appear to be an isolated incident, but rather a widespread habit. It is therefore unlikely to occur to presuppositionalists that a worldview which avoids the defects in Hume’s conception of causality may very well be immune to the systemic skepticism to which it leads. Since Hume’s conception of causality plays a central role in establishing his skeptical view of induction, a philosophy which holds to an alternative understanding of causality is unlikely to suffer from the same vulnerability.

In a forthcoming blog, I will explain just how the Objectivist view of causality does in fact avoid the pitfalls which plague Hume’s problem of induction. However, it is my prediction that presuppositionalists will ignore these points as if they did not pertain, and continue to insist that only Christianity somehow solves the problem of induction (even though it lacks an objective understanding of causality). But don’t be surprised when at this point they fail to present arguments and instead prefer simply to rattle off questions in rapid succession in the manner of “How do you know….?” over and over and over again, while ignoring answers given to those questions.

by Dawson Bethrick