The following paper, which I wrote while I was in high school, seems to me in some respects to adumbrate extropianism. In addition to _Cybernetics_, which I mention explicitly & which was then recent, I believe that by that time I had read Orwell's _Road to Wigan Pier_ & Vonnegut's _Player Piano_, both of which I regarded as _reductiones ad absurdum_ of the notion that there was a difference in kind between humans & machines. It is also possible that I had read _Erewhon_.
I thank Jim Fehlinger for using his scanner to transcribe this old typescript to ASCII.
||: The universe is not only queerer than we suppose, but :||
||: queerer than we can suppose. :|| ----------------------------------------------------------------------
Free paper due end winter term, 1953
MACHINE, MAN, AND THE FUTURE
Since the advent of the use of heat engines and other power machines in the 19th century the replacement of more and more of human muscle power with machine power has become among us a quite acceptable ideal. Indeed, if we ever succeed in avoiding the prospect of a future of successive disastrous wars, it seems quite probable that most of the physical work of the world will be done by mechanical devices.
But the invention and quick development of automatic control devices showing similarities in capability to human beings, such as electronic calculators, has not met with such great philosophical welcome. While some wild theorists like L. Ron Hubbard have vastly overemphasized the similarity between calculators and nervous systems, attempting to reduce psychology to electronic engineering, most of us seem to side emotionally with Prof. Aiken of Harvard, who, it would appear, turns white with rage at the mere suggestion that a machine can "think." Others, such as Watson of I. B. M., less interested in mere playing with words, give us the more reasoned defense that machines cannot replace scientists because machines can only handle data which scientists have previously organized and inserted. Wiener's famous statement on the effects of such machines on society, from his book Cybernetics, is worth quoting here:
". . . The first industrial revolution . . . was the devaluation of
the human arm by the competition of machinery. . . . The modern
industrial revolution is similarly bound to devalue the human brain in
at least its simpler and more routine decisions. Of course, just as
the skilled carpenter, the skilled mechanic, the skilled dressmaker
have in some degree survived the first industrial revolution, so the
skilled scientist and the skilled administrator may survive the
second. However, taking the second revolution as accomplished, the
average human being of mediocre attainments or less has nothing to
sell that it is worth anyone's money to buy."
"The answer, of course, is to have a society based on human values
other than buying or selling. . . ."
Unfortunately these defenses of the "organic" mind against the ultimate encroachment by artificially constructed ones must also in the long run fail. Let us take them one by one.
The common statement that "machines can't think" either is nonsense or
is not true. It is not even necessary to quibble over the definition
of "thinking" to see this. These things are obvious. There is no
need to look on thought as a supernatural phenomenon. Thought, or the
"soul" (the only meaningful definition for which is the capacity for
thought) is not a bit of Godly (or diabolical!) vapor floating about
in the head, nor is it (as some of our earlier "materialists"
asserted) a fluid which the brain secretes "as the liver secretes
bile." Thought is a kind of perfectly natural behavior of the nervous
system, the result of certain relations between neurons and energy
impulses, all explainable by the laws of physics. One can see no
logical reason to assume that the cells of the nervous system are the
only devices which can behave in that particular way. As a matter of
fact, the principle present structural difference between a neuron and
an electric relay seems to be that the neuron can store information
not just by taking an impulse and passing it along but also by raising
its "threshold value," that is, the minimum strength of an incoming
stimulus which will cause it to respond by sending along another
impulse to the next cell. Wiener thinks this is the basis of longterm
memory, so it is important. But there is no reason we cannot
build such a thing into a tube or transistor, or get the same effect
by some other method of storage if it seems advisable. It therefore
follows that anything a man can do, a machine can, in theory, do also.
The difference between them is not a fundamental structural one. So
any position in any society "based on human values" (or any other
values, for that matter), whether or not of buying or selling, could,
theoretically, be occupied by a machine.
Similarly the idea that a machine can only handle carefully prepared and fed in data may be disposed of. Already in automatic factories calculators work from data which are fed directly to them from gauges without human intervention, and express their results by turning valves with similar independence. The reaction of a calculator to a situation, moreover, need not be standardized, but can be made subject to individual circumstances, so that it is not possible to predict at the beginning of a run what the mechanism will be doing at a given point. The reactions of the machines will, it is true, depend on what they start out with and what they get thru their receptors later, but the same is true of us or of anything that reacts -- for we may be sure that the random factor which exists in us will also exist in such complicated machines, not only because of the "noise" factor which exists in any communication system, but because some sort of device for producing random impulses is often useful in ratiocination. The situation in which it takes a genius to prepare material for a calculator is, in all, quite temporary.
As to Wiener's provocative statement, it, like the others, is perfectly true for a while, but in the long run he seems to have missed the point. The reason certain skilled "manual" workers have survived the first industrial revolution is not that a machine could not make the same motions as (say) a dressmaker, but that a machine cannot have the requisite delicacy and conditionality of control. This, it seems, will be eventually supplied by our electronic relay systems. And, if a machine can be built with enough units of the right sort, it can doubtless replace the scientist too. The question of whether such a thing will happen depends on two things: technological advance and human psychology. The first, if our technical civilization can be preserved, is rather to be taken for granted: the hugely increased reliability and hugely decreased size and power requirement of the new transistors as compared to electron tubes are by no means the last step in the progress in such components, and it looks as if larger and larger control mechanisms will become possible. To discover accurately the relationships between the brain cells in a human being in order to reproduce them accurately in a machine with necessary modifications will be our biggest job, and is far beyond the present state of biological knowledge, but it is quite probable that this problem too will be eventually solved.
But it remains to be seen how people will take this gradual usurpation of their duties by their creations -- this not so much referring to unemployment in the present sense (which if we ever get a reasonably sane society will not exist), but because the jobs will have nothing to do on them, with the machines doing their manufacturing and even their thinking for them at the push of a button. By the time it were decided that machines could push buttons more efficiently too, a large number of people probably would have thought of trying out the latest explosive on the whole shebang. This, however, would not do any good. To deliberately deny the machines work in order that people might have it would be of no use: it might keep people busy, but it wouldn't keep them happy unless they were very, very stupid, and the race would become so decadent after a few generations of doing work in essence just as artificial as digging a hole and filling it up again, that it would not be worth preserving anyway. There must be another alternative.
There is another alternative. It is the only one that pleases me. Men must be neither master nor servant of the machine: he must become the machine. He must ultimately desert brain cells for transistors, muscles for motors. He must build machines similar in structure to his brain and equipped with the proper receptors and effectors, and not let them do things for him, but be he. Each person must have the relationship between his nerve cells reproduced as a relationship between artificial components. Then the "organic" human race will pass away.
Some people now will not like this prospect. They will say that it
means the death of true humanity. But it does not. Humanity will
live on because the machines will be just as much people as we are.
They will, as a matter or fact, be far better people than most of us
are today. They will consume, grow, learn, control, and reproduce.
Therefore by any sensible definition they will be alive and they will
be human. What, after all, makes a man a man? To use an iron lung or
a wooden leg does not make one less human than someone who uses the
"natural" equipment. Neither will using electronic brains; on the
contrary, they, unlike wooden legs, will be more efficient than those
for which they are substituted. The greatest thing man has is the way
he learns. This will continue, will even be improved, That is the
important thing.
People will also say that the mere fact that the brain is using disconnected, mobile, radio-controlled receptors and effectors instead of eyes, ears, hands and feet will make necessary great structural differences in the brain, so that it will not be human. It is true that there will have to be great changes. But they will all make man more powerful and more adjusted. We are rather different from the Cro- Man, but that does not make us less human. Quite the contrary, we are more human, because we can learn better. Similarly, the machines are the next step. But let us not quibble over the definition of "man." What is worth preserving will be preserved.
There are many other reasons that this step is advisable, not to say necessary. One of the most important is our evolution. If we ever establish a peaceful civilization, humanitarianism will completely eliminate the natural selection on the basis of "survival of the fittest" which has been the basis of the evolution of species so far. The result will be degeneration. Already in civilized countries the intelligence of the masses is slowly falling. The biologist Carter, noting that human beings look more like foetuses of the lower animals than something more advanced, has warned that there seems not much hope for further evolution of the body. Conversion to artificial components would obviously solve this difficulty. Instead of having to rely on the slow, wasteful, and (for us) inoperative process of random mutation and natural selection, we will be able to plan reliable circuit changes and improvements in instruments and put them into effect immediately. It will not, of course, be quite as simple as changing a tube in a radio is, but the process will be more similar to this one than it will be to biological evolution such as we have (?) today.
The results of this planning would be marvelous. First, all the petty diseases which bother us today could be gotten rid of. We would have no appendicitis because we would have no appendices, and if we did, they would be made of steel and the toughest microbe would wear his teeth out on them, and we would have no fear because every reaction would be in essence planned and an automatic one not a result of policy could not exist. Second, we would not be limited by the ridiculous restrictions which our bodies now impose on us: such and such an atmosphere at such and such a temperature, seeing only a small amount of space at one time: we could see by radio thru distant cameras all over the solar system at once if we wished. And, last but certainly not least, we would be practically immortal, making permanent learning possible as well as long interstellar voyages. The new possibilities for enjoyment and freedom which this wonderful versatility would give us are quite uncountable.
Then, too, we can see the great results of eliminating the troublesome dichotomy between man and machine. No more would we just sit and let machines do everything for us: we would be doing it ourselves again, just as before the "Machine Age" -- but with what different results! It is true that we would do many things more or less automatically, but so do we now our breathing, and no one accuses us of not doing our breathing for ourselves! Man would again be his own master, not a slave to his creation, as Thoreau has called him, but his own creation.
The crisis of the gathering of knowledge is another important reason we must eventually switch to machinery. Knowledge begets more knowledge in a way which is delightful but just now rather terrifying as the holdings of libraries go up geometrically and the organization of the vast amounts of specialized information is even now getting hopeless, specialists in "distant" fields being often unaware of each other's work tho they might be of immense use to each other. Only the flexibility offered by a machine gives any hope of solving this urgent problem and making science efficient again.
Especially inasmuch as it will undoubtedly be several centuries before our society becomes sane enough to make the necessary advance for this move, to object, therefore, to this wonderful possibility of reform is no more than silly sentimentality (and where would we be if the apes had had it?) We are now foolish, and may make the machine our destruction, but if we avoid that, it can be our destiny. Rather, then, we must work to hasten the day when man can make himself thus most able to live, to learn, and to control as long as there is aught in the universe for him to do it with; until the wa"rmeto"d marks his end concurrently with that of the present order of the universe as well.