There are some things we have in common with computers. When we reason, we use logic. There are also logic circuits built into computers. I remember learning about these in my first engineering class in college. There's AND, NAND, OR, NOR, and so on. There are specific outputs given specific inputs. They work in a way that's not unlike deductive logic.
Recently, I was reading through some comments (I don't even remember where), and somebody described what computers do as reasoning. While I can see why I person might think of it that way, that is not how I think about reasoning. To me, reasoning is a conscious process. If you don't have consciousness, then you don't have reason.
A computer with logic gates behaves mechanistically and blindly. They don't actually think. When we reason, we do not just passively process information. We draw conclusions from premises by mentally "seeing" that the conclusion follows from the premises. There's an intuition involved in reasoning because it is by intuition that we recognize the logical relationship between various statements and propositions. In order to reason, we have to process and understand the meaning of the information we receive in a way that computers don't.
I am tempted to say the difference between what we do when we reason and what computers do when they process information using logic gates is even more apparent when we move away from math and deductive reasoning and more into inductive reasoning. I'm not sure, though. On the one hand, the conclusions of inductive arguments are not logically required in the same way the conclusions of deductive arguments are. On the other hand, there are algorithms that allow computers to form generalizations. On the third hand, those algorithms have to be able to be reduced to deductive processes. Otherwise, you wouldn't be able to code them. I'm not sure that's true when people reason inductively. It might be, though.
Often when we reason inductively, a lot of that reasoning is subconscious. For example, when we have negative experiences, we automatically anticipate the same negative experience under similar circumstances. We start avoiding those circumstances because of this anticipation, but we don't have to explicitly think anything like, "Every time I have been faced with these circumstances in the past, it has resulted in unpleasantness; therefore, I should expect the next time I run up against these circumstances, it will also result in unpleasantness; therefore, I should avoid those circumstances." Any baby or animal can learn through experience that fire is hot, for example, without having to go through a set of propositions and a conclusion. If humans come to conclusions through subconscious processes, and we consider that "reasoning," then is consciousness really necessary for reasoning?
I still say yes. Everything about this subconscious way of coming to a conclusion still requires consciousness. There's the conscious experience of feeling the heat from the fire, the conscious experience of dreading future contact with the fire, etc. Any conclusion we reach results in a belief, and a belief is something that requires consciousness. A belief does not have to be expressed in words. A dog probably has no idea how to express the thought, "Fire is hot," but he still knows it's true. Language isn't necessary for belief or thought, but consciousness certainly is.
That is not to say we have to constantly be thinking about or giving mental attention to a belief in order to have a belief. If I were presently consumed with thoughts about pizza, and diamonds were the furthest thing from my mind, I would still have a belief that diamonds were hard. You don't have to be presently thinking about something in order to have a belief about it. A belief can be stored like a memory where it can be recalled, but it doesn't have to be right in front of our mental gaze.
I know I've rambled a bit. I'm just thinking out loud. The bottom line is that I don't think computers reason, at least not in the usual sense of the word. While there are similarities between what computers do and what minds do, I think the major difference is in whether the process is blind and mechanistic, like a computer, or whether it involves intuitively "seeing," as well as understanding, like a mind. The way we draw conclusions by thinking things through is not how computers arrive at outputs, even when those outputs are expressed in words.
No comments:
Post a Comment