Consciousness - An Introduction (ch17)

What views are there on the consciousness of machines? - Chapter 17

Thinking and awareness

Could a machine be aware of something? This question is more difficult than the question of whether machines can think. If you ask the latter question, you can go two ways. You can say that it depends on how the word 'think' is defined. You can also say that you want to design an experiment that shows that a machine can think. This is what Turing has done. It is more difficult to know whether a machine has consciousness.

First of all, there is no clear definition for the word 'consciousness'. In addition, there is no experiment (such as the Turing test), with which it can be investigated whether machines have an awareness. If we assume that consciousness has to do with subjective experiences, then only the machine itself can know whether it has a consciousness.

Some people believe that if you ask a robot if he has a consciousness and he says that this is so, it is likely that he pretends to have a consciousness, because he has been programmed in this way. A functionalist will say that robots have a consciousness because they can perform certain tasks as robots. According to functionalists, the consciousness depends on being able to act. Consciousness is not a cause according to this view of task performance. An inessentialist does not believe that machines or robots can have a consciousness because there is no inner experience with machines. If both visions (both those of the functionalist and the inessentialist) were legitimate then there would still be no simple test to examine the consciousness of a machine.

Counter arguments

There are several arguments to indicate that machines could never have a consciousness. First, it feels unnatural to think that a machine can have a consciousness. You can also assume a dualistic perspective. For example, based on your religion, you can say that God has only given a consciousness to humans. It is also possible to adopt a non-religious dualistic perspective. In that case you can say that the mind is something that is separate from the brain . Since this is not the case with machines, they could never have a consciousness.

You could say that the machine is a zombie (and therefore has no consciousness), that God has given a soul or mind to the machine or that a machine simply has a consciousness. It is also possible to say that robots have no consciousness, because only living organisms can have a consciousness. You could say that the function of neurons can never be replicated with robots. Another argument is that organisms grow and learn before they have developed a consciousness. 

Searle believes that brains are the cause of the mind. Surprisingly, however, he does not call himself a dualist. He states that consciousness has a neurological basis, but that consciousness must be experienced. He believes that brains have the ability to create experiences. Searle does not think that brain tissue is necessary for consciousness. He believes that other organisms can also have a consciousness, but only if they, like the brain, can cause a mind with experiences.

A fourth argument against the consciousness of machines is that for some things a consciousness is needed and that machines can not execute these things because they have no consciousness . An example is developing a sense of humor and enjoying an ice cream. There is also the vision that is called 'Lady Lovelace's objection'. A machine could only do what is asked of him. A machine itself could never be creative.

The Chinese room

Turing was an opponent of the idea that we can only know what our own mind is.

We could also really understand other types of intellects (for example, machines). Searle did not agree with this and thought up the well-known Chinese room thought experiment. According to him, ' Strong AI' does not exist, because using the right program by a machine is not enough to develop a capacity for understanding.

The thought experiment of Searle goes as follows. Imagine: you are in a closed room and you get to see Chinese reading pieces there. You do not understand Chinese yourself and you can not speak it either. There is also an English book with rules in the room. People outside the room give you two Chinese reading articles about a story, together with a questionnaire that you have to fill in about what you have read. The English book with rules shows you what reaction you should give to the questions. At a certain moment this goes so well that people outside the room think that you can read and understand perfect Chinese. Afterwards, people outside the room will give you an English story to read. You answer these questions yourself because you are English yourself. Yet your answers to the Chinese story and the English story to the outside world are equally good,but there is a crucial difference: you really understand the English stories, while you do not understand the Chinese story. In the case of the Chinese room you behave like a computer that receives input from the English book to give the correct answers.

In short:The conclusion of Searle is that a computer itself can never really understand anything . He can not connect meanings to the input he receives. So there is no question of a capacity for understanding. Searle thinks that man has intentionality and the machine does not. Intentionality is about focusing attention on something to understand it. According to Searle, intentionality is a subjective issue and therefore related to consciousness.

Criticism

There has been a lot of criticism on Searle. So there has been the ' brain simulator reply' . This means that there can be a program that can simulate the way neurons fire in Chinese brains. Searle says that perhaps that is possible, but that you can not design intellect or consciousness with that. Chalmers believes that people can interact with other objects in a concrete way, while that does not apply to computers. He states that implementation is an important concept in this respect. Having the right computer program is not enough to create a consciousness. It is about applying the right program.

There are also differences of opinion about what Searle's thought experiment really proves. Some people say that Searle proves that computers can not have real understanding, while others think the experiment does not prove anything. Finally, there is the argument that there are things that machines can not do. If we can do these things, it means that we are more than just machines and that we have something special, namely a consciousness. Turing also calls this the ' mathematical objection' . The first part of this argument is correct. There are indeed things that a Turing machine can not do, but the ' incompleteness theorem' states that all systems (including human systems) are ultimately not complete. But does this really apply to people?

Turing states that humans can easily make mistakes and that man is not necessarily superior to a machine. Penrose argues that arithmetical understanding goes beyond just counting and that a consciousness is needed for this. Searle believes that consciousness can be imitated in machines (although it will never take human form), but Penrose does not even believe in it. Penrose thinks that understanding something is something very different than calculating something. Penrose believes that consciousness arises through coherence in microtubes . These are proteins that resemble tubules and can be found in almost all cells of the body. Because of these microtubes we feel that we have a self and a free will. Penrose's theory becomes ' objection reduction' called.

Kurzweil is an opponent of the vision of Penrose. Machines may not be able to solve all problems, but people certainly do not, he thinks. People can only make estimates of problems according to Kurzweil, but computers can do this too. Grush and Churchland are also opponents of Penrose's vision. Microtubes are located throughout the body and not only in the brain.

It is said that no evidence has been found that microtubes have anything to do with consciousness. Moreover, it is suspected that effects from one microtube can not be transferred to the other microtube to explain the unity of consciousness. Yet Penrose's theory remains popular because he does not explain consciousness in terms of neurons. Most people find it a strange idea that consciousness only comes from neurons that work together. Penrose uses principles from quantum physics to defend his theory. For many people that sounds much more appealing than just a statement based on the activity of neurons.

Resources:Blackmore; Susan. (2010). Consciousness, Second Edition An Introduction. Abingdon, Oxon: Taylor & Francis.