The purpose of this paper will be to describe the nature and function of Turing Machines as well as the account of mental and physical states which such a machine is supposed to provide. We will then use this account to demonstrate how it is that a Turing Machine is an instance of a Functionalist account of the mind.
A Turing Machine is an extraordinarily simple device which consists of two functional objects: a tape and a scanner/printer into which the tape is fed. The tape upon which a Turing Machine runs is a string of squares of, ideally, infinite length and a single square’s width. Within each of these squares is a symbol, including the possible blank square, upon which the scanner/printer, having scanned the square, can perform a total of three possible functions: 1) erase, 2) print another symbol or 3) leave as is. The scanner/printer, however, can itself perform a total of four basic tasks: 1) scan, 2) erase/print/leave-as-is, 3) move left a square and 4) move right a square.
Nevertheless, while the actual tasks that the scanner/printer may perform with the tape are extremely simple, this should not prevent us from recognizing where the real complexity of the system lies, namely is the machine table which decides what the scanner/printer will do after having scanned a square. The machine table can be thought of as a table of columns and rows which are filled with instructions for the scanner/printer to perform in the case that it should scan a particular symbol (each rows corresponds to a particular symbol) while the machine is in a particular state (each column corresponds to a particular state which the machine can be in). Thus, the total number of possible instructions which a machine could be given will be equal to the total number of possible symbols which could be read multiplied by the total number of states which the machine can possibly be in, just as the number of cells in a table is simply the number of rows multiplied by the number of columns.
Thus, while the tape upon which the Turing Machine run is of infinite length, the machine itself “is a device with a finite number of internal configurations, each of which involves the machine’s being in one of the finite number of states, and the machine’s scanning a tape on which certain symbols appear.” (Putnam, 21)
Before moving on to consider the relation which has been claimed to exist between a Turing Machine and the mind, it will be of profit to get a better grasp of the role which these computational “states” play in the functioning of a Turing Machine. Each sequence consists of four essential parts: the scanning of the symbol, the printing of the correct output symbol, the movement which the scanner/printer is to make (left or right) and the state into which the machine is to enter before the next sequence begins. It is essential that we understand the importance of the state which the machine is in, for the same input symbol can result is very different output (printing, moving, change of state) behavior by the machine. Thus, an input symbol only can serve as such in the context of a particular computational state, and has no such meaning outside of such a context.
We are now in a position to fully understand the claim which Hilary Putnam makes with regards to Turing Machines and the mind. According to Putnam, mental states simply are the computational states into which a Turing Machine, or any other computing machine for that matter, can be in. In other words, according to such a model our physical selves are the scanner/printer which receives input and delivers output to the tape, which is meant to represent the external world. Accordingly, to enter the mental state “pain” is to enter a computational state in which certain inputs from the world, presumably by way of sensory input, produce certain behaviors, behaviors which differ from that which the exact same input would produce were we in the mental state of “euphoria.”
This should not, however, be confused with the claims of the behaviorists that mental states are simply behavioral states, to take the definition of crude behaviorism, or the disposition or tendency toward certain behavior, although this latter definition does seem strikingly similar. The difference between the two is that mental states are not behavior at all, be it actual or dispositional. According to Putnam, mental states are not the behavior which derived from the input received in a particular computational state, but are instead the computational states themselves. It is in equating mental states with computational states rather than the behavior which such states partially determine that Putnam is able to argue that “all of the [various issues and puzzles that make up the traditional mind-body problem] arise in connection with any computing system capable of answering questions about its own structure, and have thus nothing to do with the unique nature (if it is unique) of human subjective experience.” (Putnam, 20)
Having thus described the nature of Turing Machines, as well as the relation which Putnam sees between them and the traditional problems surrounding the mind, we will now conclude with a brief discussion of Functionalist accounts of the mind, such as the one we have just been discussing. While the tape and the scanner/printer in the case of the Turing Machine are meant to be taken as literal in that they are to be seen as actually tokens of interactor/interacted (machine/tape) types, the machine table is not to be taken as such in that such a table is meant only as a description of what the machine is doing. There need be no literal table which the machine is constantly searching within itself. “A ‘machine table’ describes a machine if the machine has internal states corresponding to the columns of the table… Any machine that is described by a machine table … is a Turing machine.” (Putnam, 22) Similarly, Putnam is not suggesting that there is an actual “brain-table” which our brains are constantly searching in response to its constant flow of input. Instead, such tables are only meant as a description of what the brain is doing, the table is meant to describe the function of the brain.
Thus, while a Turing Machine need something which serves the physical function of the tape as it does something which serves the physical function of the scanner/printer, it does not need anything which physically serves as a machine table for the does not physically exist. The machine table is simply a description of the functional states which the machine can be in just as ‘paper-weight’ is a description of a functional state which a rock, piece or glass or just about any other physical object can be put in. Just as the computational states can be instantiated in the physical medium of the machine/tape system, so too a mental state can be instantiated by means of neurons, silicon, or even an army of chickens pecking in the proper patterns.
A mind, according to Functionalism, is whatever object performs the function of a mind, and is simply not limited by the physical medium in which such a function happens to be instantiated in any given case. The case of the Turing Machine is a clear example of Functionalism in that computational states and mental states are equated with each other due entirely to the fact that the two are seen are performing the identical function as each other, regardless of whether such states are instantiated in a brain or not.
Bibliography
Putnam, Hilary: “Minds and Machines” as reprinted in Minds, Brains and Computers (Blackwell Publishers Ltd, 2000)
