Further discussion of this point would carry us into the work of social and cultural
science which cannot be followed here. Instead let us consider certain other illustrations.
What was said above of stimulus and sensation is applicable to physiology and the
biological sciences no less than to psychology. It has been tried, for example,
by postulating sums of more and more special apparatus, to account for meaningful
or, as it is often called, purposive behaviour. Once more we find meaninglessly
combined reflexes taken for granted although it is probable that even with minute
organisms it is not true that a piece-stimulus automatically bring about its corresponding
Opposing this view is vitalism which, however, as it appears to Gestalt theory,
also errs in its efforts to solve the problem, for it, too, begins with the assumption
that natural occurrences are themselves essentially blind and haphazard - and adds
a mystical something over and above them which imposes order. Vitalism fails to
inquire of physical events whether a genuine order might not already prevail amongst
them. And yet nature does exhibit numerous instances of physical wholes in which
part events are determined by the inner structure of the whole. 
These brief references to biology will suffice to remind us that whole-phenomena
are not "merely" psychological, but appear in other sciences as well.
Obviously, therefore, the problem is not solved by separating off various provinces
of science and classifying whole-phenomena as something peculiar to psychology.
The fundamental question can be very simply stated: Are the parts of a given
whole determined by the inner structure of that whole, or are the events such that,
as independent, piecemeal, fortuitous and blind the total activity is a sum of the
part-activities? Human beings can, of course, devise a kind of physics of their
own - e.g. a sequence of machines - exemplifying the latter half of our question,
but this does not signify that all natural phenomena are of this type. Here is a
place where Gestalt theory is least easily understood and this because of the great
number of prejudices about nature which have accumulated during the centuries. Nature
is thought of as something essentially blind in its laws, where whatever takes place
in the whole is purely a sum of individual occurrences. This view was the natural
result of the struggle which physics has always had to purge itself of teleology.
To-day it can be seen that we are obliged to traverse other routes than those suggested
by this kind of purposivism.
Let us proceed another step and ask: How does all this stand with regard to the
problem of body and mind? What does my knowledge of another's mental experiences
amount to and how do I obtain it? There are, of course, old and established dogmas
on these points: The mental and physical are wholly heterogeneous: There obtains
between them an absolute dichotomy. (From this point of departure philosophers have
drawn an array of metaphysical deductions so as to attribute all the good qualities
to mind while reserving for nature the odious.) As regards the second question,
my discerning mental phenomena in others is traditionally explained as inference
by analogy. Strictly interpreted the principle here is that something mental is
meaninglessly coupled with something physical. I observe the physical and infer
the mental from it more or less according to the following scheme: I see someone
press a button on the wall and infer that he wants the light to go on. There may
be couplings of this sort. However, many scientists have been disturbed by this
dualism.and have tried to save themselves by recourse to very curious hypotheses.
Indeed, the ordinary person would violently refuse to believe that when he sees
his companion startled, frightened, or angry he is seeing only certain physical
occurrences which themselves have nothing to do (in their inner nature) with the
mental, being only superficially coupled with it: you have frequently seen this
and this combined ... etc. There have been many attempts to surmount this problem.
One speaks, for example, of intuition and says there can be no other possibility,
for I see my companion's fear. It is not true, argue the intuitionists, that I see
only the bare bodily activities meaninglessly coupled with other and invisible activities.
However inadmissible it may otherwise be, an intuition theory does have at least
this in its favour, it shows a suspicion that the traditional procedure might be
successfully reversed. But the word intuition is at best only a naming of that which
we must strive to lay hold of.
This and other hypotheses, apprehended as they now are, will not advance scientific
pursuit, for science demands fruitful penetration, not mere cataloguing and systematization.
But the question is, How does the matter really stand? Looking more closely we find
a I third assumption, namely that a process such as fear is a matter of consciousness.
Is this true? Suppose you see a person who is kindly or benevolent. Does anyone
suppose that this person is feeling mawkish? No one could possibly believe that.
The characteristic feature of such behaviour has very little to do with consciousness.
It has been one of the easiest contrivances of philosophy to identify a man's real
behaviour and the direction of his mind with his consciousness. Parenthetically,
in the opinion of many people the distinction between idealism and materialism implies
that between the noble and the ignoble. Yet does one really mean by this to contrast
consciousness with the blithesome budding of trees? Indeed, what is there so repugnant
about the materialistic and mechanical? What is so attractive about the idealistic?
Does it come from the material qualities of the connected pieces? Broadly speaking
most psychological theories and textbooks, despite their continued emphasis upon
consciousness, are far more "materialistic", arid, and spiritless than
a living tree - which probably has no consciousness at all. The point is not what
the material pieces are, but what kind of whole it is. Proceeding in terms of specific
problems one soon realizes how many bodily activities there are which give no hint
of a separation between body and mind. Imagine a dance, a dance full of grace and
joy. What is the situation in such a dance? Do we have a summation of physical limb
movements and a psychical consciousness? No. Obviously this answer does not solve
the problem; we have to start anew - and it seems to me that a proper and fruitful
point of attack has been discovered.  One finds
many processes which, in their dynamical form, are identical regardless of variations
in the material character of their elements. When a man is timid, afraid or energetic,
happy or sad, it can often be shown that the course of his physical processes is
Gestalt-identical with the course pursued by the mental processes.
Again In I can only indicate the direction of thought. I have touched on the
question of body and mind merely to show that the problem we are discussing also
has its philosophic aspects. To strengthen the import of the foregoing suggestions
let us consider the fields of epistemology and logic. For centuries the assumption
has prevailed that our world is essentially a summation of elements. For Hume and
largely also for Kant the world is like a bundle of fragments, and the dogma of
meaningless summations continues to play its part. As for logic, it supplies: concepts,
which when rigorously viewed are but sums of properties; classes, which upon closer
inspection prove to be mere catchalls ; syllogisms, devised by arbitrarily lumping
together any two propositions having the character that ... etc. When one considers
what a concept is in living thought, what it really means to grasp a conclusion;
when one considers what the crucial thing is about a mathematical proof and the
concrete interrelationships it involves, one sees that the categories of traditional
logic have accomplished nothing in this direction. 
It is our task to inquire, whether a logic is possible which is not piecemeal.
Indeed the same question arises in mathematics also. Is it necessary that all mathematics
be established upon a piecewise basis? What sort of mathematical system would it
be in which this were not the case? There have been attempts to answer the latter
question but almost always they have fallen back in the end upon the old procedures.
This fate has overtaken many, for the result of training in piecewise thinking is
extraordinarily tenacious. It is not enough and certainly does not constitute a
solution of the, principal problem if one shows that the atoms a of mathematics
are both piecemeal and t the same time evince something of the opposite character.
The problem has been scientifically grasped only when an attack specifically designed
to yield positive results has been launched. Just how this attack is to be made
seems to many mathematicians a colossal problem, but perhaps the quantum theory
will force the mathematicians to attack it.
This brings us to the close of an attempt to present a view of the problem as
illustrated by its specific appearances in various fields. In concluding I may suggest
a certain unification of these illustrations somewhat as follows. I consider the
situation from the point of view of a theory of aggregates and say: How should a
world be where science, concepts, inquiry, investigation, and comprehension of inner
unities were impossible? The answer is obvious. This world would be a manifold of
disparate pieces. Secondly, what kind of world would there have to be in which a
piecewise science would apply? The answer is again quite simple, for here one needs
only a system of recurrent couplings that are blind and piecewise in character,
whereupon everything is available for a pursuit of the traditional piecewise methods
of logic, mathematics, and science generally in so far as these presuppose this
kind of world. But there is a third kind of aggregate which has been but cursorily
investigated. These are the aggregates in which a manifold is not compounded from
adjacently situated pieces but rather such that a term at its place in that aggregate
is determined by the whole-laws of the aggregate itself.
Pictorially: suppose the world were a vast plateau upon which were many musicians.
I walk about listening and watching the players. First suppose that the world is
a meaningless plurality. Everyone does as he will, each for himself. What happens
together when I hear ten players might be the basis for my guessing as to what they
all are doing, but this is merely a matter of chance and probability much as in
the kinetics of gas molecules. - A second possibility would be that each time one
musician played c, another played f so and so many seconds later. I work out a theory
of blind couplings but the playing as a whole remains meaningless. This is what
many people think physics does, but the real work of physics belies this. - The
third possibility is, say, a Beethoven symphony where it would be possible for one
to select one part of the whole and work from that towards an idea of the structural
principle motivating and determining the whole. Here the fundamental laws are not
those of fortuitous pieces, but concern the very character of the event.