G5AIAI - Introduction to Artificial Intelligence

This course is run at the The University of Nottingham within the School of Computer Science & IT. The course is run by Graham Kendall (EMAIL : gxk@cs.nott.ac.uk)

The Chinese Room

If the Turing Test was passed Turing would conclude that the machine was intelligent. John Searle argued that behaving intelligently was not enough to prove that a computer was intelligent. To demonstrate this he devised a thought experiment which he called the Chinese Room (Searle, 1980). It worked as follows.

The system comprises a human, who only understands English, a rule book, written in English, and two stacks of paper. One stack of paper is blank. The other has indecipherable symbols on them.
In computing terms the human is the CPU, the rule book is the program and the two stacks of paper are storage devices.
The system is housed in a room that is totally sealed with the exception of a small opening.
The human sits inside the room waiting for pieces of paper to be pushed through the opening. The pieces of paper have indecipherable symbols written upon them.
The human has the task of matching the symbols from the "outside" with the rule book. Once the symbol has been found the instructions in the rule book are followed. This may involve writing new symbols on blank pieces of paper or looking up symbols in the stack of supplied symbols.
Eventually, the human will write some symbols onto one of the blank pieces of paper and pass these out through the opening.

Assume that the symbols passed into the room were valid Chinese sentences, which posed questions. Also assume that pieces of paper passed out of the room were also valid Chinese sentences, which answered those questions. Searle argues that we have a system that is capable of passing the Turing Test and is therefore intelligent according to Turing. But Searle also argues that the system does not understand Chinese as it just comprises a rule book and stacks of paper which do no understand Chinese.
Therefore, according to Searle, running the right program does not necessarily generate understanding.

There were various arguments against Searle's conclusion. One of them Searle called the Systems Reply. This argued that the system as a whole understand Chinese.
Searle's argument was to remove all the elements the system and place the entire system inside the brain of the human. Therefore, the human would memorise all the rules and the stacks of paper.
Searle's now argues that the only thing that can understand Chinese is the human and if you asked the human (in English) if they understood Chinese the answer would be no.

In AIMA (Russell, 1995) it states that the real claim made by Searle is as follows

Certain kinds of objects are incapable of conscious understanding (of Chinese)
The human, paper and rule book are objects of that kind.
If each of a set of objects is incapable of conscious understanding, then any system constructed from those objects is incapable of conscious understanding.
Therefore there is no conscious understanding in the Chinese room.

AIMA goes on to say that the third assumption is invalid and puts this argument forward as to why.

If you believe it [item 3] and if you believe that humans are composed of molecules then either you must believe either

That humans are incapable of conscious understanding

Individual molecules are capable of conscious understanding.

There have been various arguments for and against The Chinese Room. These have been given the names of "The xxxxxx Reply" and they are summarised below.

The Systems Reply states that the system as a whole understands.
Searle argues that the system could be internalised into a brain and yet the person would still claim not to understand.
The Robot Reply argues we could internalise everything inside a robot (android) so that it appears like a human.
Searle argues that nothing has been achieved by adding motors and perceptual capabilities.
The Brain Simulator Reply argues we could write a program that simulates the brain (neurons firing etc.)
Searle argues we could emulate the brain using a series of water pipes and valves. Can we now argue that the water pipes understand? He claims not.
The Combination Reply argues we could combine all three of the previous suggestions.
Searle argues that this does not progress the argument. As we know how the android operates Searle claims it is not carrying out strong AI

Each of these replies is iscussed in more detail in Searle's paper (Searle, 1980) and I suggest you read it. You might also want to take a look at (Copeland, 1993), chapter 6 and the places to visit listed below.

Places to Visit

http://www.ptproject.ilstu.edu/pt/chinovrv.htm (Chinese Room Overview)
http://www.utm.edu/research/iep/c/chineser.htm (Chinese Room Overview)
http://www.ptproject.ilstu.edu/pt/chinrob1.htm (The robot reply)
http://www.ptproject.ilstu.edu/pt/chinsys.htm (The Sytems Reply)

References

Copeland, J. 1993. The Curious Case of the Chinese Room in Artificial Intelligence : A Philospohical Introduction. Blackwell
Russell S. and Norvig P. Artificial Intelligence : A Modern Approach, Prentice Hall, 1995, ISBN 0-13-103805-2
Searle, J.R. 1980. Minds, brains and programs. Behavioural and Brain Sciences, 3: 417-457, 1980.
Turing, A.M. 1950. "Computing Machinery and Intelligence." From Mind LIX, no. 2236, Oct. 1950 : 433-460.

Last Updated : 11 Sep 2001