Why David Hume should terrify you; The Scandal of Induction

David Hume was a Scottish philosopher from the period of the Enlightenment (18th century) who is considered one of the greatest philosophers of all time. He is described as a very pleasant and mild-mannered person who supposedly kept his upbeat attitude even on his deathbed. Apart from the French philosopher J.J. Rousseau, who, due to his poor mental health in his later years, accused Hume of trying to plot murder against him, scarcely is there anyone who would be afraid of David Hume. If one would be given a unique opportunity to share a dinner with any philosopher in history, to choose him would probably be one of the best choices. With this in mind, why and how should Hume terrify us?

The revolutionary importance of Hume’s philosophical arguments is illustrated by the fact that many of them remain “unsolved” to this day*. The consequences of some include an entire restructure of certain pieces of foundational western knowledge. Such is the case with many scientific presuppositions which were present in the west for almost 2000 years and were demolished by Hume. From the time of Aristotle (up until Hume), the process of the scientific method was largely played out (and formalized through the Baconian method by the great philosopher Francis Bacon) as the following; 

• the scientist observes a certain phenomenon in the world

• For the sake of this argument, let’s say that he is trying to determine the color of swans. Based on previous personal experience, the scientist hypothesizes that swans are white

• However, he can’t verify his theory if his sample is too small. Thus, he must make multiple observations of various swans, possibly in numerous different locations, with different factors involved (e.g., climate, habitat, etc.)

• Bear in mind, though, that the scientist can't check every swan in the world, so as a rule of thumb, the larger the sample the better

• Consequently, by a structured and organized method, he gathers large amounts of data by which he proves or disproves his initial theory

• Seeing that all of the swans explored in this hypothetical story were indeed white, our scientist infers from this particular subset of swans that all swans are white 

In turn, Hume saw this process as faulty insofar as there is no rational reason for safely assuming that all swans ought to be white merely because a large amount of them is, in fact, white (it is obvious from the formulation why this problem is often referred to as the is-ought gap). Hume concluded that science makes a logically impossible leap in the last step of the aforementioned process; it gathers a general conclusion (all swans are white) from a number of particular cases (several individual swans were white) while having no rational backing to do so. Interestingly, the aforementioned swan example is actually a real historical account, which further proves his point; the Europeans thought that all swans were white for a long period of time until they found black swans in Australia. In other words, they had a theory that was being verified continuously each time they saw a white swan, but only one black swan was needed to disprove the theory. That erroneous last step is a case of inductive reasoning (as opposed to deductive which makes conclusions on particulars from the general), hence Hume’s theory is known as the scandal of induction. Such reasoning is used through science in evaluating what Hume refers to as “matters of fact” ( i.e., things and events in the world). There is a difference, he points out, between knowledge of the world, the so-called synthetic knowledge, which we gather a posteriori (through our experience), and analytic knowledge which we can gather a priori (regardless of our prior experience). The former is used in empirical science while the latter with formal sciences (mathematics and logic). The statement “all bachelors are unmarried”, to give an example, is a case of analytic knowledge because the second part of the statement is necessarily true. We know it without ever meeting or seeing a bachelor - it is not logically possible for a bachelor not to be unmarried. On the other hand, empirical science deals with synthetic knowledge such as “the billiard ball will move once it is hit by another ball”. While this statement may be true, it is not necessarily true; it is a case of what is, not what must be (the bachelor must be unmarried by definition). Empirical science,  in other words, deals with establishing the descriptive (i.e., how things are). However, there is an unbridgeable gap between how things are and how they will/ought to be (the normative). Thus, empirical science can only be used to explain those things that were experienced. For this, Hume gives an example of a billiard ball hitting another to show that the effect of this action is learned, not deduced; 

“It must invent or imagine some event, which it ascribes to the object as its effect, and it is plain that this invention must be entirely arbitrary. The mind can never possibly find the effect in the supposed cause, by the most accurate scrutiny and examination. For the effect is totally different from the cause, and consequently can never be discovered in it. Motion in the second billiard ball is a quite distinct event from the motion in the first. nor is there anything in the one to suggest the smallest hint of the other…”  - An Essay Concerning Human Understanding

We know that the ball will move once it is hit because we have experienced the ball moving in the past. We generalize that the ball will always move once hit due to the false assumption that the future will resemble the past. But the past only shows what had already happened, not what must or will happen. Habitually experiencing some event does not guarantee that such event will continue happening. Now, it might be argued the past has previously helped us in judging what will happen in the future. In fact, it does so daily when we, for example, choose one route to work in favor of the other because it had previously shown to have less traffic. However, the same method must be applied here as well; the fact that the future has previously resembled the past (that we could determine which route to take based on past experience) does not guarantee that it will continue to do so. This just means that we justify our irrational belief that the future will resemble the past on the grounds of the future resembling the past in the past - thus, making the same error only one level up. We cannot grab to the particular instance of the past (which may have shown that the future indeed resembled the past) to get to the general (that future always resembles the past) just as we can’t determine that all swans must be white because some of them are. Like our European ancestors and their black swan, it is entirely possible (and likely) that one day we will be stuck in traffic cursing our naive faith of predicting the future on the basis of the past.

This theory was undoubtedly revolutionary as it caused the restructuring of the foundations of the scientific method, and reshaped our idea of what science is, but what is so terrifying about it? Namely, the inevitable consequences that follow; not only is science proved to be impotent in regards to making any predictions or universal comprehensions, but the human knowledge as such is revealed not to be able to grasp anything resembling an absolute/unquestionable truth. We can only know the world based on what the world “decides” to present to us. We are certain that an object will fall if we let go of it solely because it always did. We have described with physics and mathematics why it falls, but the fact that it will fall was established only after we’ve seen it continually. Likewise, in a hypothetical situation in which the object remains standing in thin air, the only thing we could do is to admit that we were wrong beforehand and gather new conclusions as to why it didn’t fall this time. From an epistemological standpoint, we can’t even be certain of that. Certainly, a thought that brings uneasiness.

Thus, there are certain existential conclusions from Hume’s theory that must be noted. To recapitulate, experience doesn’t guarantee future knowledge. Despite this, habit combined with the human mind’s need to simplify causes the mind to make shortcuts and conclude that certain events must have certain consequences - what would in existential terms mean that order* between events can be found. The mind tries to find connections between unconnected events in order to put rational boundaries on the irrationality of being. It must do so as the complete randomness with which existence seems to operate (chaos) is in direct opposition to what the human mind wants - even more, needs (order). To illustrate this existential turn on Hume’s philosophy further, a quote from Blaise Pascal might be helpful which states that there is no rational reason one should be here rather than there or now rather than then. It merely is like that. We are, as Heidegger pointed out, thrown into existence with no apparent logic or plan behind it. Consequently, we tend to think that the connections between events in our existence (i.e., cause and effect) are reasoned deductively rather than inductively. We think that a certain effect must follow its cause as this brings peace to our minds through the allegedly established order. In such a way we perceive ourselves as something like the masters of nature; we think we know the necessary effects of causes and with it, how the world will unfold rather than admitting our dependence on nature in establishing any modicum of firm knowledge on it. Essentially, our knowledge of existence is based on habit (the inherent need/passion of finding meaning), not on rationality. Thus, the famous Hume’s quote notes that reason is always a slave to passions. Nature guides us and directs our knowledge, not the other way around. Perhaps one day nature will “decide” to simply keep the object in the air, leaving us with our mouths open and minds blown. Consequently, if the certainty of gravity isn’t certain, what is? How can we find our way through the chaos if there is, arguably, zero hard ground to stand on? It is a question to ponder on, and it is definitely a terrifying one.

 

*The only one who came close to resolving Hume’s scandal (i.e., who arguably did so) was Karl Popper. Through his theory of falsifiability (with which we define science to this day), Popper answered Hume's predicament by agreeing that it is impossible to gather universal conclusions from particular cases, but adding that the particulars CAN disprove the universals (i.e., falsify them). In other words, Popper realized that science can't work in the Aristotelian way of gathering many pieces of individual data which prove a theory, but must rather try to falsify the theory with everything in its power. Thus, the theory that underwent the largest number of falsifying experiments and remained standing can be considered the best one so far (until a better one comes along).

* Order would in this regard be defined as events unfolding in a predictable way in which the person wants them to unfold (chaos being the opposite) - known vs. unknown territory