Computer and the Mind of Man, Part 1, Logic by Machine (1950)

[Music] man has described himself in many ways but if we concentrate on man in relation to his control over his environment no description is more apt than the description of man as a tool making animal and in the short history of mankind the majority of the tools which man has made can be thought of as an extension of his muscle power that is the ability to perform work fast and in greater quantity however in the mid 1940s of the 20th century a different kind of tool was invented a tool for extending certain of the powers of man's mind this tool is the electronic computer [Music] is it is the fast reliable and tireless performance of a variety of arithmetic and logical operations that gives the computer its great utility and power but merely looking at a computer won't tell us very much about what it actually is doing neither will this tell us anything about the Revolutionary material and intellectual effects of such machines we can easily see the material and intellectual effects of say machines for transportation we know that the modern jet aircraft represents a great increase in speed over the earliest aircraft we also know that modern airplanes have made the world smaller and changed our way of thinking about ourselves and our world and future means of transportation will bring even more rapid and radical changes but even the difference between the speed of an ox cart and the fastest rocket is small when compared with the difference in speed between calculation by hand and calculation by computer for example the first electronic calculator to be completed could do the work of 50,000 people working by hand scientists when they speak of a great change in speed or size prefer to speak in terms of a unit of measurement called an order of magnitude meaning 10 times as much Dr Richard Hamming a research mathematician for the Bell Telephone Laboratories in a paper presented before a meeting of the American Academy for the advancement of science put it this way the computer Revolution is often compared with the famous Industrial Revolution in importance and scope the Industrial Revolution effectively Freed Man from being a beast of burden the computer Revolution Will simly free him from dull repetitive routine the computer Revolution is however perhaps better compared with the cernic or the darwinian Revolution both of which greatly Chang man's idea of himself and the world in which he lives before getting into the main part of this paper is is necessary to discuss briefly the idea of a change in a technology change is often measured in units of an order of magnitude meaning roughly a factor of 10 10 times as much it is a common observation that a change of an order of magnitude and Technology produces fundamentally new effects as an illustration consider the following example modern jet planes are about one order of magnitude faster than W Brother's first plane another example the fastest missiles are somewhat more than two orders of magnitude faster meaning about three time 300 times faster automobiles are used at speeds around one order of magnitude faster than a horse and wagon each of these have produced whole new effects indeed it is said that the automobile has produced even a change in our morals computers have improved in speed by at least 6 hour of magnitude a millionfold in order to understand the factor of a million consider the following two situations first that you have only $1 and second that you have a million dollars you can readily see that in the two situations there are fundamentally different effects you adopt a different view of yourself in the world in which you live along with the change in speed there has been a great increase in reliability so we now do much longer computations than were practicable by hand finally with the increase in speed there has been a corresponding decrease in the cost something more than 1,000 times cheaper it is as if suddenly automobiles cost $2 to $4 houses 20 to $60 and the changes in the computer technology are still going on these then are the we hardly need to be reminded that we live in a world that is becoming more complicated and more crammed with information every day one description for this vast quantity of data on on everything from the lifetime earning records of an individual to the beeps and pulses relayed to Earth from a space satellite uses that overworked word explosion this time an information explosion the computer is an invaluable tool for processing these millions of bits of information in accurate fast and economical fashion in accordance with rules and instructions provided by the human programmer in the most gigantic of all recordkeeping jobs the Social Security System more than 1 million personal records can be processed in one day this manufacturing plant is entirely computer controlled in accordance with rules for decision- making stored in the machine's memory this chemical plant for making polyisoprene was designed by a computer in a recent example a computer in 16 hours and out of 16,000 possible designs selected the design that most closely approached the ideal design as defined by the engineers and this actual firing of the multi-million dollar Saturn rocket was simulated on this computer many times before the actual firing was authorized and these simulated firings which helped eliminate many of the problems in the function in of the rocket cost only a fraction of what an actual launching would have cost each of these operations recordkeeping and accounting control design and simulation is achieved through the manipulation of numbers according to instructions and rules given to the computer by the programmer but what are these rules and what is the relationship between numbers and the real things they are said to represent these are some of the questions that we put to the scientist and we've already seen Dr Richard hemming research mathematician at the Bell Telephone Laboratories well I would say that at present at the Bell Telephone Laboratories we do about 10% of the experiments on a computer and about 90% in the laboratory I would expect that in time we will do 90% on the computer and 10% in the lab speed cost and effort favor the computer over the laboratory approach this Advantage is possible because we construct a mathematical model rather than a material model I use the word construct because the mathematical equations are the construction rather than a materialistic physical model it is perhaps fortunate we have found particularly in the field of physical sciences that our predictions based upon the mathematical models agree very well with what we observe in the physical world the 19th century physicist Heinrich Herz described the concept of model in these words we make for ourselves internal pictures or symbols of external objects and we make them of such a kind that the necessary consequences in thought of the internal pictures are always pictures of the necessary consequences in nature of the object symbolized the idea of a mathematical model is fairly easy to understand a very simple trivial example is the numbers you calculate on your check stubs you combine various numbers by addition subtraction and they correspond in a very real sense to the amount of money that you have in the bank the mathematical model is the numbers you manipulate the amount of money you actually have in the bank is the physical world let us consider now a slightly more complex example this is a common experience for most of us the front end of the car dips sharply when we come to a sudden stop in technical language there is a transfer of weight from rear to front in the action of braking a mathematical model of this transfer of weight can reveal exactly how much weight is on both wheels and different Road surfaces and braking conditions will result in different weight distributions but the mathematical model for the transfer of weight from rear to front will remain the same in this mathematical model of Behavioral automobile we can use a computer to simulate what would happen in actual practice we have actually found that our mathematical models of automobile traffic predict very well many of the effects which we observe returning to our more General remarks about the use of computers we first use computers to simulate situations in engineering situation which we had been doing before by hand the computers allowed us to do much larger and more complex problems but this ignored the order of magnitude effect which we spoke of we are now beginning to use machines in entirely new ways and entirely different problems and this is the exciting part this is the intellectual aspect of applying the machines to completely new ideas that has so excited many people in the field one of the uh characteristics of human beings is that among other things that they do is that they solve problems and what it means of course to solve a problem is is uh being able to not only get an answer to uh a question that arises in a particular situation but then to find other cases that are similar to the initial one so that we are in effect in a position to solve not just one concrete problem but a whole class of them this is Professor Ernest Neel leading logician and philosopher from Columbia University once a domain has been developed to to the point where uh answers can be uh given from given set of assumptions uh simply by following uh some uh set of instructions or some rules then it's not entirely evident that one should be thinking of the significance of every step that one is performing is it also possible ble to instruct machines to follow the rules of thought indeed I think this is of course uh the basis for having uh machines or computers uh since their uh entire task insists in following a set of rules or programs uh so that uh in a very uh short uh span of time uh they are able to come out with an answer uh that would have taken human beings an extraordinary length of time to produce it seems to me that the computers because they abl to ask new questions also enable us to get entirely new answers they do not answer all the old questions but because the questions are new the answers are also new and very exciting there is a strong tendency to speak of the machine as solving the problem when in fact really it is the program which describes to the machine what the machine is to do this is overlooked and I think a great deal of confusion arises from this it is not that we do not have adequate machines to solve our problems many times but rather we lack adequate descriptions of how to solve the problem and this is a very important point to understand as we spread out and learn more and more about our techniques of solving problems we will be able to do wire and wire class of problems and is not the machine so much as is our lack of ideas controls mathematics has been called the one universal language of mankind it is perhaps our most powerful tool for problem solving and in terms of subject matter pure mathematics and logic are one and the same the subject matter consists in the relations generally expressed by if then necessarily computers have been called machines for Logic the human Problem Solver must first Define the problem he must also provide the machine with instructions as to the steps necessary to solve the problem and then the machine will carry out these instructions accurately and with fantastic speed all of this of course requires very precise logical thinking on the part of the problem solver it is the machine's power to carry out these logical operations that gives it its Great Value as a tool for extending man's thought logic is not the science of how we we think thinking involves much more than logic it is essential to logic and Mathematics that given a set of premises the conclusion will follow in all cases results from a computer are the necessary conclusions arising from the data and rules of operation provided the Machine by the human programmer and yet these conclusions can accurately reflect and foretell real life situations what is the nature of this remarkable correspondence between the abstract symbols of mathematics and logic and the real perceptible world of objects and events we asked Professor Neel to comment on this well I I would like to uh say something about that because this has been uh I suppose for many many centuries a problem that has occupied uh men uh and uh because of the apparent uh agreement of what we might call reality or nature nature with the language that we employ whether it is ordinary English or whether it is the language of mathematics uh the apparent agreement uh between reality and the symbolic representation uh men have speculated about uh sometimes about the origin of the world uh something about uh the necessity of there being some particular uh kind of creative agency and so if one asks why is it that the consequences of assumptions that we make are in agreement with n in nature the answer is because nothing can count as an instance of the assumptions that we make unless the assumptions themselves are confirmed the electronic computer in the service of the human Problem Solver reflects back to him the consequences of the assumptions he employs when he builds a mathematical model of a real life situation uh so that rather than saying as for example uh the late uh s James jeans maintain that the fact that mathematics is used in exploring nature uh he maintain that this shows that uh there must have been a Divine creator that moreover that the Creator was a mathematician in terms of this uh simple reminder of the sorts of things that enter into applying any formal system to any segment of nature uh one would have to say the reason why nature conforms uh to our assumptions uh is not that there was a mathematician who created it but simply that nothing would count as an instance of the assumptions unless it also conformed to whatever consequence was deduced from it the problem solving power of mathematics and logic combined with the speed and accuracy of a machine for carrying out these operations is the basis of the computer Revolution but does this mean that a computer can produce a new idea or make an original contribution to knowledge we have this question of both professor nego and Dr Hamming it depends entirely what you mean by a new idea in some senses the machine only produces what you expect on the other hand you do not foresee all combinations and the Machine often produces combination which you had not foreseen and have the impression of novelty well the question is to whether machines or computers are capable of uh original thought or of making uh original contributions to knowledge uh is a very much debated one but before one can profitably discuss it it seems to me it's very important to distinguish between two senses in which uh a statement or an idea might be said to be original the two senses are psychological sense of originality and a logical sense of originality and I think uh perhaps the best way of making the distinction clear is first with an illustration uh when a child is given an elementary arithmetical problem for example to find the sum of two numbers uh when he gets the answer uh the conclusion is something that is novel or new to him so the psychologically of course he gets the sense of having made a discovery uh nevertheless the answer is is something that is necessarily or logically implied in the premises or the assumptions uh and the process of thought here simply is to unravel or make explicit what was already contained in a logical sense in the premises I think i' risk uh uh going out on a LM here and say that it seems to me that uh at least on the basis of our present conceptions of the capacities of uh computers as well as the actual on that have been constructed that uh uh computers do not come up with logically new ideas and uh unless the nature of the Beast changes very radically that they cannot come up with logically new ideas given the assumptions that underly the present theory of computers many people are attempting to use computers to replace man I am much more interested myself in using man and machines as a working combination I believe that some of the future is already fairly clear is obvious that the machines can do routine logical consequences much more rapidly and much more accurately than a human can therefore I believe we will gradually push the detailed logical Parts the methods which we use onto the machine and we will reserve for ourselves the creation of new ideas and the direction of the research thus man machine combination will find a dividing line passing gradually from man over to machine machine taking more and more of the burden as we learn how to use the machines and as the machines become more powerful and we will find ourselves left in the more creative aspects of the problem deciding what it is worth attacking what are the problems which we wish to solve these kind of questions we do not know how to give the computer and seem best reserved for man at present I think we have to say that uh in this respect uh the human mind is capable of doing things that uh none of the machines that are uh available today is capable of doing this certainly doesn't foreclose the question whether sometime in the future uh machines might not be devised so that uh uh exactly this this kind of a thing will be credit to machines but at present and this is all that one can say at the moment uh there is there are clear lines of demarcation between capacities of the human mind in respect to the capacity to make contributions to knowledge which is our original Dr Heming will you give me some of your summary thoughts on this computer Revolution well the word revolution is used very freely these days and yet I think there's some meaning to the words the computer Revolution we have got an increase in computers of a million pole decrease in costs by a thousand as we said before this is a very real change but the question still arises will this revolution continue it seems to me in view of the political situation in the whole world that no responsible group can oppose the coming computer Revolution therefore it seems to me we have a real one which will continue now believe the reports we hear of this revolution are very misleading they often underestimate the speed which is going on and the importance but worst of all they tend to emphasize material aspects how much we will have of this and that how much Leisure what wonderful new things we will have and they fail discuss the impact it will have our view of ourselves and our world Outlook this I believe is the most important part of the computer Revolution I do think that uh as a consequence of these increased uh complexities in society and the uh obviously greater uh degree of control that will have to be uh uh exercised if we are not going to run into each other constantly uh it does uh requires to think seriously as to how in the future uh we or perhaps more specifically our children and grandchildren uh will be able to realize the potentialities that we would like our own Society to enable us to develop we shall have to find ways of training ourselves how to enjoy the Leisure that we presumably are going to have the greater Leisure that we're going to have in such a way that instead of being frustrating and making life a Perpetual boredom will enable us to engage in a creative activities this seems to me uh to be the central problem that faces uh western civilization in so far as the material Goods of Life are going to be increased but it does uh raise a fundamental uh issue as to how what we sometimes called the spiritual aspect of of life or I would prefer to call it the creative aspect of individuality will be given uh a room for development now there's no simp simple answer to this question if there is any mystery in the computer Revolution Let's be sure that the mystery is located in the right place in spite of its magnetic memory complex circuitry and so on the mystery is not in the machine the sense of awe and mystery that we may experience when we see a computer in operation is really awe at the accumulated months and years of thinking by human problem solvers reflected back to us at Fantastic speed the computer scientist once remarked that we have machines that compute with the speed of light but with the intelligence of the earthworm well it is true that the computer can be called a kind of brain although a very limited one in order to solve any problem the computer must first be instructed by a human programmer who has painstakingly and logically analyzed the problem the computer is then given the problem in the form of numbers or instructions pertinent to arithmetic it is the arithmetic logical task that the computer is organized to do once instructed it can do as much arithmetic in a minute as a man in a lifetime a man in a lifetime a lifetime of all mankind is but a brief moment in the long history of this Earth of ours and only yesterday in the history of mankind has man made any significant advance in his control over his Earthly environment computers machines for logic may change Chang this more than any other of man's inventions these machines which have been with us less than a millionth of a second in terms of the temporal span of man's history have already given promise of deep and far-reaching change in our way of life and way of thinking they literally accelerate into milliseconds of time our ability to perform logical arithmatic and control tasks but finally the future accepted in terms of Trends and probabilities is the great unknown and the uses to which we put ourselves in relation to the marvelous Machines of our own invention is perhaps the central moral and intellectual challenge of our own brief moment in time [Music] [Music] this is NE National educational television