Hume refuted the theory that "causality" is true and inevitable in "Theory of Human Nature" and later "Research on Human Cognition". He pointed out that although we can observe that one thing comes with another, we cannot observe the relationship between any two things. According to his skeptical epistemology, we can only trust the knowledge gained from our observation. Hume believes that our concept of causality is just an idea, that is, we expect one thing to happen with another. "We don't know causality, only know that some things are always linked, and these things have never been separated in past experience. We can't see through the rationality behind the connection of these things. We can only observe these things themselves and find that these things are always classified by our imagination through a frequent connection. " (Hume, 1740: 93) Therefore, we can't say that one thing makes another. We only know that one thing may be related to another.
Hume put forward the word "constant combination" here, which means that when we see that one thing always "causes" another thing, what we see is actually that one thing always "frequently contacts" with another thing. Therefore, we have no reason to believe that one thing does cause another thing, and these two things may not always be "linked" in the future (popkin &: Strolling, 1993: 268). The reason why we believe in causality is not because causality is the essence of nature, but because of our psychological habits and human nature (popkin &; Stroll, 1993: 272).
This statement put forward by Hume strongly refutes causality. Some philosophers after Hume, such as Bertrand Russell, completely abandoned the concept of causality and only regarded it as a superstition. But what comes from this is the question of causality-where does our understanding of causality come from? What kind of connection can we recognize? This question was later debated by the German philosopher Kant.
Hume claimed that human beings (and other animals) have an instinct to trust causality, which comes from the habits formed in our nervous system. We can't get rid of this habit for a long time, but we don't have any arguments, and we can't prove that this habit is correct through deduction or induction, just as we know nothing about places outside the world.
It is worth noting that although the theory of "frequent contact" is generally considered to be put forward by Hume, other philosophers may have put forward similar concepts before Hume.
Medieval philosopher Maimonide quoted several Islamic philosophers who also didn't believe in causality. He wrote in the book A Guide to Confused People: "In short, we shouldn't say that' this' is the reason for' that'." From the perspective of these Islamic philosophers, everything created by the creator is independent and unconnected, so there is no certain connection between these things. In The Theory of Human Understanding, Hume believes that all human thinking activities can be divided into two types: the pursuit of "the relationship between ideas" and "the question of facts". The former involves abstract logical concepts and mathematics, mainly intuitive and logical deduction; The latter focuses on the study of the real world. In order to avoid being influenced by any actual truth we don't know or facts we didn't perceive in past experience, we must use inductive thinking.
The principle of inductive thinking is to assume that our past behavior can be a reliable guide to our future behavior (this is sometimes called the principle of natural consistency). For example, if according to past experience, the sun always rises in the east and sets in the west, then inductive reasoning will tell us that the sun may still rise in the east and set in the west in the future. But how do we explain our ability to make such inferences? Hume believes that it is impossible for us to interpret our thinking ability as the product of reason, because reason can only come from two ways, and neither of them can be the basis of our reasoning and thinking:
Argumentation or intuition: this kind of thinking is basically transcendental, and we can't prove that the future will be consistent with the past with transcendental knowledge, because the obvious fact that can be thought out (logically) is that the world is no longer consistent. Hume did not clearly distinguish between the general principle of natural consistency and a "concrete" principle of consistency here. A philosopher may argue (or perhaps be a Kantian school) that, in fact, it is really hard for us to imagine that the world does not operate in a "definite" form; But Hume's key point here is that even any "specific" unified principle in natural operation may stop running in the future. Therefore, inductive thinking cannot be based on transcendental knowledge.
Induction: We can't resort to the successful experience of inductive reasoning in the past to prove the reliability of inductive reasoning, because it will constitute a circular argument.
Hume went on to conclude that our thinking ability does not have a rational basis, because there is no form of rationality to prove this ability. It should be noted here that Hume is not advocating the following points: he is not advocating that induction is unreasonable because it does not belong to deduction (Hume is not a so-called "deductionist").
As Hume said in the section "On Skepticism and Reason", he argued that if reason can form our thoughts without foundation, and if thoughts are made up of reason from beginning to end, then we can't believe anything, including any truth derived from intuition or deduction.
In addition, Hume does not advocate that induction is infeasible, nor does he advocate that induction cannot draw reliable conclusions. On the contrary, Hume argues that this inductive thinking is not actually composed of reason. Another key point of Hume's theory is that although Hume is pessimistic about the possibility that induction belongs to rational thinking, he still believes that inductive reasoning has an excellent and magical ability to foresee the future. In order to solve the problems we face in understanding inductive reasoning, Hume put forward "nature" as the answer to the problem. Nature determines that there will be more similar things we expect in the future, and "this way of thinking allows us to infer possible results through the same reasons, and vice versa." This way of thinking is an indispensable condition for all human beings to survive in the world. But we can't believe the wrong inference made by our reason, which is not only dull in thinking, but also very easy to make mistakes in our lives since we were born (On Human Understanding, 5.2.22). Hume's statement is probably the closest to the theory of evolution of that era (pre-Darwin era) about human inductive thinking ability. Hume also highlights the main differences between himself and all atheist thinkers here, and presents his side as a naturalist thinker completely. Hume pointed out that we usually assume that we are the same as five years ago, although we have changed in many ways. We were the same person five years ago and now. We also think about how much time can change a person's heart without changing ourselves. However, Hume denied that there are differences between the mysterious self and a person's various personalities. When we begin to introspect, we will find: "we can never have any consciousness unless we rely on some kind of feeling;" "People are just a collection or package of many different feelings, and these feelings always alternate with each other at an unimaginable flow speed."
Obviously, in the process of our thinking, our thoughts are always changing, and our imagination can easily change from one idea to another similar idea, and the characteristics of the idea itself are enough to form connections and associations. Similarly, our feelings are bound to change constantly, and the changed feelings will be similar to those before. Imagination must be a way of thinking cultivated through long-term habits. With the change of space and time, new ideas are constantly emerging [15].
It is worth noting that from Hume's point of view, these feelings belong to nothing. On the contrary, Hume compared the human soul to a republic, which maintained its noumenon not by any permanent core concept, but by various different, ever-changing and interrelated concepts. Therefore, the individual's noumenon is just a loose connection composed of a person's various personal experiences.
In short, for Hume, it is not important whether "ontology" exists, but what matters is the causal relationship, series relationship and similarity between various feelings. Most people will think that some behaviors are more "reasonable" than others. For example, swallowing aluminum foil is a strange move for most people. However, Hume denied that reason played any important role in promoting or rejecting specific behaviors. After all, rationality is only the calculation of concepts and experiences.
According to Hume, what really matters is how we feel these behaviors. Hume's theory is regarded as the basic principle of modern instrumentalism, which holds that the rationality of an action should depend on whether it can achieve its predetermined goals and desires, no matter what these goals and desires are. Reason only acts as a medium and tool, telling us what kind of behavior can achieve our goals and desires, but reason itself can never in turn guide us to choose what kind of goals and desires.
Therefore, if a person wants to swallow aluminum foil, reason can tell him where to find aluminum foil. There is nothing irrational about "eating aluminum foil" or "wanting to eat aluminum foil" (of course, unless a person has a strong desire for health or sensory ability, reason will tell him not to do so). But today, many people think that Hume has actually reached the realm of nihilism here, and point out that one can actually deliberately obstruct one's goals and desires without violating the rational principle ("I want to eat aluminum foil, let me stick my mouth up"). Such behavior must be quite abnormal, but because reason does not play any role and cannot be used to evaluate behavior, such behavior will not violate reason.