Can Computers Actually Think?
A big question that has kept computer architects, and well, pretty much everybody who has seen a sci-fi film, awake at night is “can computers actually think?” Often this thought can lead to many more questions and visions of robots with human levels of intelligence. One philosopher at the University of California, Berkeley has doubts about the traditional method for detecting “intelligent computers.” John Searle criticizes the Turing Test by presenting a thought experiment, known as the “Chinese Room Argument.”
Essentially the Chinese room Argument claims that the Turing Test fails to distinguish between machines that have intentionality - which he talks about in his paper “Minds, Brains, and Programs.” He founds his argument on two principles “(1) Intentionality in human beings (and animals) is a product of causal features of the brain…” and “(2) Instantiating a computer program is never by iteslf a sufficient condition of intentionality.” Long story short, the Chinese Room Argument is an example of “…how a human agent could instantiate the program and still not have the relevant intentionality.”
I think that Searle makes a good point with the Chinese Room Argument, but that it also has logical comparison flaws. First, a human has little in common with a central processor computer chip. Perhaps they both have the ability to carry out tasks, but they are so different that merely substituting a human for a computer chip in the Chinese Room Argument is bound to have problems.
One critical counterargument explained in the reading assignment summerizing the Chinese Room Argument is “while the man in the room does not understand Chinese, he is not the whole system. He is but the central processing unit, a simple cog in the large mechanism that includes…the whole room…that understands Chinese.” This is a great counter, however, this too suffers from logical flaws! (ex. is this critic claiming that the brain [central processing unit] is a “simple cog?” I think the brain is a lot more complicated than that!) However, it does raise another interesting point - is the “room” all that is needed for thinking, or can that merely be thought of as the body or brain itself? (i.e. needing the computer chip to actually do the “thinking” - whatever that may be) Of course, this question is left for the Cognitive Scientists to figure out.
In any case, I can’t help think of the notion of transitivity when it comes to the concept of Intentionality. There MUST be levels of it, for Searle even says himself, the kind of intentionality that he discusses in his paper applies to human beings AND dogs. Logically we can assume that humans and animals have different levels of consiousness and thus different notions of Intentionality. Since we (generally) think of humans as being more intelligent than animals, it’s easy to visualize a transitive spectrum wherein humans lie on a point ABOVE animals.
If one accepts that Intentionality is necessarily transitive, than one can imagine that complicated computers carrying out tasks and solving problems lie at some point on the transitive scale. Whether or not these computers can THINK is a different issue. Processing information involves no personal identity trouble for a machine, at this point. Once again, it seems cognitive science holds the key to the door of understanding, at least on this issue.
