Hume and the Problem of Induction

Enlightenment philosopher and empiricist David Hume proposed the problem of induction. He begins by questioning causation, which is a relationship between two objects that go beyond the memory and the senses. 

For example, after we have seen a soccer ball (assuming it is properly inflated) hitting a solid wall, we infer that the soccer ball will bounce back after hitting the wall—the act of the soccer ball hitting the wall causes the soccer ball to bounce back. Hume claims that, to make this claim, we must use a posteriori reasoning (reasoning which uses empirical evidence) and not just pure a priori reasoning (pure reason): Hume claims that we make this inference because every other time an inflated soccer ball has a solid wall, it has bounced back (induction). 

However, to be able to make this inference (that the relationship between the soccer ball hitting the wall and it bouncing back is causation), we must make another assumption: the universe will continue to function in the future as it has in the past. If the ball has bounced back every other time, then it will continue to bounce back in the future. This is the assumption that the laws of nature must remain the same, or the Uniformity Principle (UP). If the UP is reasonable, then we can infer causation from induction, but the UP is unreasonable, then we are not justified in believing that one event causes another—one merely occurs after another, and there is no causation involved. 

Next, Hume goes on to show that this assumption cannot be supported by reason. Hume believed that there are two types of reason: demonstrative reasoning, concerning the relationships among ideas (and moral reasoning), and that “concerning matter of fact and existence”, in other words, induction made from the past on to future events. 

The first type of reasoning, demonstrative reasoning, cannot support the UP because demonstrative reasoning can only be used for things and ideas which cannot be false. Therefore they also cannot make any contradiction that the course of nature might change, and that objects of past experiences might act differently in the future. The other type of reasoning (that concerning matters of fact and experience) Hume calls probable reasoning, because it relies on the previous assumption that the laws of nature must remain the same. Therefore any argument presented through probable reasoning must be circular, as its conclusion would be used to support one of its own premises. 

Thus, because no type of reasoning can support the assumption which is the basis for causation from induction, causation cannot be determined from induction made from experience. The ball bouncing back merely occurs after the ball hitting the wall; the ball hitting the wall does not cause the ball to bounce back—there is no necessary connection between the two events. Adding another layer to the example, we might infer that a properly inflated soccer ball traveling towards the wall will bounce back. In this case, we make the assumption that the ball will continue traveling until it hits the wall and the assumption that the ball will bounce back. We believe that the ball traveling towards the wall will cause the ball to hit the wall, and that the ball hitting the wall will cause the ball to bounce back. However, Hume’s problem of induction states that neither of these inferences can be made, since causation cannot be inferred from past experiences (induction). Because of this, we can merely say that these events happen after each other, and that there is a high probability, but not certainly. 

Why it matters

The problem of induction matters because assumption made as the basis for causation is the same assumption made for modern science. Modern science bases itself upon induction, and the laws of the world which are derived from measurements and experiments (induction). Because of this, science, in order for it to be applied and to produce a truth that persists in the future, must assume that whatever we measured/experienced in the past must happen the same way in the future. Yet this assumption is not supported by reason—science, with all its ‘rationality’, makes a fundamentally unreasonable assumption! 

As science broke away from philosophy in the 17th and 18th centuries, people seemed to have forgotten the origin of science—natural philosophy, in a time when science (as well as many other subjects) was not its own field, but rather a subcategory of philosophy. They seemed to have forgotten their root and origin in philosophy, which roots not in purely empirical evidence, but also reason. Indeed, as scientific work of the current era continues, certain fundamental problems still need solving, and one of them includes the problem of induction. As science derives its methods from philosophy, so through philosophy science must justify its methods—if modern science solves the problem of induction, is science still reasonable, or is it just circular reasoning hidden in uncertain induction?

The Is-ought problem

Hume also presented the is-ought problem (also known as Hume’s law), which is the law that moral judgements cannot be made from non-moral premises (facts). This is similar to the problem of induction, as many moral arguments rely on normativity and continuity as well, but in a moral sense instead. For example, the argument that because something has happened continuously in the past, then it ought to happen in the future is erroneous (the problem of induction states that it is uncertain that it will happen that way in the future). The is-ought problem could be described as the moral version of the problem of induction. 

Science and Hume

The Is-ought problem can also be applied to modern science. As the field of science grows and certain scientists and scientific discoveries enter into the world of politics, people begin to expect more and more of science. They not only want science to tell them what there is in the world and how it works, but also how these discoveries tell people what they ought to do. But perhaps without any knowledge of philosophy, they would not realize the limitations of science, both in its fundamental concepts, or in its ability to provide the essential grounding for moral conclusions. 

The world seems to be in necessity of a reconnection with philosophy, as modern science, with its ever-quickening development, needs people to review and perhaps revise its roots and fundamentals. Even if one is at the front lines of exploration and new scientific developments, philosophical knowledge would still allow them to stay within the bounds of science, and not step into the modern, messy world of politics and society.