WERNER HEISENBERG (1953)

Werner Heisenberg (1901 - 1976)
3 followers
...and the number is expressed as 1 through 137.
So, with a magnitude of a little less than 1%,
the magnetic moment of the electron can deviate from the Dirac value.
Now, with the quantum electrodynamics, this deviation has been calculated,
namely, it has been calculated with an accuracy of 1 to 1,000,000,
to 6 decimal places exactly,
in which one has gone up to the square of this Sommerfeld fine structure,
that is, up to the constant.
The latest measurements, carried out by Bloch and his colleagues,
have actually confirmed this value up to 1,000,000th.
That means, all 6 decimals of this number have been confirmed by the experiment.
The last one is then experimentally somewhat uncertain.
So, one can say,
this quantum electrodynamics has actually proven itself as well as one could hope for.
There are certain trains of quantum electrodynamics,
that have not yet been fully confirmed by the experiment,
although they also represent a deviation from the classical electrodynamics
and therefore would be interesting.
According to the quantum electrodynamics,
there should be, for example, an interaction between light quantities.
A light beam should be able to be deflected or scattered by another,
but this effect is so difficult to measure, that it has not been successful so far.
So, one would not be able to doubt,
that the conditions of the quantum electrodynamics meet exactly here as well.
In this respect, in the last 10 years,
one has actually achieved extraordinary progress.
One can say, that the quantum electrodynamics is in an almost perfected form.
One can hardly imagine, that there is still something to be improved on this theory.
Nevertheless, one knows,
if one wanted to apply this quantum electrodynamics to the collision of very energetic particles,
i.e. if one wanted to ask the question,
what happens, when two electrons with enormous energies,
let's say 100 billion electron volts, collide?
Then this quantum electrodynamics would certainly give a wrong answer.
Because in reality, heavy elementary particles are created,
mesons and the like,
if according to the quantum electrodynamics, this would not come out.
Well, this is again no wonder,
because for such processes with very high energy transmission,
the separation of the electron-light quantum field from the remaining elementary particles is no longer possible.
Therefore, the forming system has to fail there.
That at this point, the validity limit for the electrodynamics lies,
can also be seen in a completely different example.
There are also other particles,
which are not so much lighter than the protons,
but are still a little lighter than the protons.
I think of the mesons, the pi-mesons.
They are about seven times lighter than the protons.
One might think, that one can also split these particles
from the protons in such a way,
that one says, I only consider processes,
in which a new creation of protons is not possible.
With such processes, I do not need to take into account the heavy elementary particles.
It is enough, if I describe the behavior of the light elementary particles as the mesons among each other.
Here, in meson physics, the distance to the heavy elementary particles is much smaller than in electron physics.
In addition, the coupling constants are much larger.
One should not expect that the separation works here,
and in fact, it does not work.
In recent years, many attempts have been made
to formulate the meson theory mathematically,
but any satisfactory agreement with the experience has not been achieved,
precisely because it is obviously not possible
to separate the mesons from the field of the other heavier particles.
So one will probably be forced,
if one drives meson physics,
to describe the whole field of elementary particles at once.
And that is, of course, much more difficult.
One cannot separate the problems,
one must, so to speak, look for the whole field of elementary particles at once.
Now, in this attempt, which has of course already been made,
and Mr. Yukawa has already spoken about a special form of such an attempt,
in this attempt, very strange difficulties arise,
which had already occurred in a slightly different form in quantum electronics.
I would like to say a few words about these difficulties.
It was purely mathematical.
First, a field equation was written,
as one was used to from classical theory,
for example, the Maxwellian theory,
applied the quantum conditions to this field equation
and assumed that one would now have to get something experimentally reasonable.
In reality, it turns out that the equations do not converge,
that one gets infinite things,
that the results become meaningless.
This was also the case with quantum electrodynamics
and could only be avoided there by a certain art,
which I do not want to discuss in more detail.
What is the reason that we initially get such infinities when quantizing field equations?
At the very beginning, one probably expected the possibility
that this would only result in an unskilled mathematical formulation of the equations,
but in the meantime it has turned out
that there is a principal difficulty
that one has to thoroughly analyze
before one can hope to come up with a solution.
The difficulty is the unification of quantum theory and relativity theory.
Or to be more precise,
it is the difficulty of bringing the uncertainty relations of quantum theory
in line with the space-time structure of the special relativity theory.
And what is this space-time structure?
If we were to describe in the future all those events
on which one can still influence in principle
and as the past all those events
from which one can at least learn something in principle,
then we would say from our daily observation
that these two groups of events,
which I have just mentioned in the future and in the past,
that they are only separated by an infinitely short moment,
which one can call the present moment.
In the special relativity theory of Einstein,
however, one has learned that the structure of space and time
is somewhat more complicated and in any case somewhat different.
If one again takes this definition of the future and the past
that I have just said,
in the future all those events
on which one can still influence in principle
and in the past all those events
from which one can learn something in principle,
then these two groups of events
are actually separated by an infinite time interval
whose size depends on the distance
at which the events take place from the observer.
And this whole space can of course now also be called the present space.
It is only important to realize
that this word present refers to an infinite time interval.
So if we think of events on a distant star,
let's say events on the star Sirius,
then between the group of future events
and the group of past events
there is a space, a time interval of the order of magnitude of a few years.
And this time interval can be called the present space.
That means that any events that occur in this time interval on the Sirius
can be thought of at the same time as events that take place here.
And are therefore also at the same time in the strict sense of the relativity theory
Now this special space-time structure of the relativity theory
has made it necessary
that the demand for causality
can only be connected in a very special mathematical form
with the theory of relativity.
Because if one wants to establish
that always the effect has the consequence of the cause,
that is, if one wants to indicate a consequent consequence of cause and effect,
then this is no longer possible in the theory of relativity
in the sense that it was possible in classical mechanics
with forces that would go to the distance.
But this is only possible
by saying that an effect can always go from one point to an immediate neighboring point.
All effects must spread at the highest possible speed,
at least not faster.
And therefore such effects are best described
with differential equations
of the type of wave equation of light,
i.e. with hyperbolic differential equations
that indicate an effect from point to point,
a so-called near effect,
and in which such effects then propagate with light speed
or at most with light speed.
Therefore, it has also been generally accepted since the special theory of relativity
that all basic laws of physics
have to be described as differential equations
of the type of relativistic wave equation.
Now, this consequence of the theory of relativity
results in difficulties with the indeterminacy relations of quantum theory.
In quantum theory, it has turned out
that one cannot know exactly from a particle, for example,
place and speed at the same time.
Either one can determine the place very sharply
and then the speed is very indeterminate
or one can determine the speed very sharply
and then the place is very unknown.
In other words, a very sharp localization of a process
has a very great indeterminacy in the impulse and in the energy as a result.
But if I now demand such a near effect,
if I say from this point here
that effects can only take place in the immediate neighborhood
and not in other spaces,
then this means an infinitely sharp localization.
So a near effect theory is a theory
in which the effect is infinitely sharply localized from one point to another.
Such an infinitely sharp localization
corresponds to an infinite impulse, an infinite energy.
So energy and impulse must be infinitely indeterminate
and these are the infinities
that have always disturbed in the mathematics of this quantum theory of wave fields so far.
So these infinities are not random things
that could be avoided by clever mathematics,
but they are directly related to the space-time structure
of the special relativity theory
and to the indeterminacy relations of quantum theory.
Now one can try to interpret these consequences
or this dilemma in some way.
And here one can point out
that although we see in the relativity theory
that we can correctly describe the relativistic behavior with such differential equations,
we are not sure that we necessarily have to describe nature with such differential equations.
So there could be some kind of far-reaching effect, at least over smaller distances,
but one would have to make sure
that this far-reaching effect adapts to the invariance requirements of the relativistic theory.
If one looks at this program,
then of course the first question is
what remains of our theory so far?
What do we have to give up? What can we retain?
We can then say
that it is obviously possible to observe from the elementary particles
at collisions and the like
what their mass is,
to observe at what speed
they leave a collision process,
how many particles come out at what speed and so on.
We can generally say
that the asymptotic behavior
of the waves or particles in large spacetime distances
can obviously still be observed.
So this must form a component of a future theory.
And then it was further established
that this asymptotic behavior
can actually be represented by certain mathematical forms
that the physicist calls the S-matrix.
And that if one
abandons all other pieces of mathematics
that with the help of this S-matrix
one can actually get a decent mathematical description
for what happens at any collision processes.
So it is possible to avoid all infinities
and to get a theory with only reasonable mathematics
if one looks at a description of the local
collision process
and is only interested in the asymptotic behavior
of the infinite.
Now, if one does that
one may perhaps also do without too much
because one has thus destroyed
the whole causal sequence of events.
I don't mean
the kind of deviation from the causality
that the quantum theory already calls out.
You know that the quantum theory
requires a certain statistical character
of the atomic processes.
But that's not what I'm talking about here.
I'm only talking about
this kind of relativity theory
that from one point
only effects with light speed
can proceed to other points.
So this kind of behavior
would then also be destroyed.
And that would be such a radical
intervention in the structure of the theory
that the whole time-relationship
would fall into disarray.
That one can no longer say
in which time order
it actually took place.
So one is in danger
of getting very crude deviations
from this concept of relativistic causality.
And on the other hand
we know that such crude deviations
do not take place.
So in recent years
one has tried
to write down theories
that also have such deviations
but only deviations in very small dimensions
from the causal principle
that is allowed by the theory of relativity
so about in dimensions
of the magnitude of the diameter
of an atomic nucleus.
This group of theories
includes for example
the non-local theory
that Mr. Yukawa talked about here.
So one has written down
such non-local theories
and has now tried
in this way
to get to a convergent mathematics
to a mathematical scheme
that is meaningful
We do not yet know
whether this program
can be fully implemented.
Certainly one can eliminate
in these non-local theories
that exist in the local theories
as in the theories
with close interaction.
But whether the non-local theories
can be implemented in such a way
that on the one hand
the full mathematical convergence
is guaranteed
and on the other hand
deviations from the relativistic causality
always only occur
in very small space-time ranges
we do not yet know
and this has to be found out
in the next few years.
The success of this extension of the theory
is not yet quite certain
and in any case
one does not yet know
whether this extension of the theory
is needed.
One can perhaps already say
that it would be desirable
that the local close interaction
would only come through
the quantum theory
because one does not have the impression
that in the area where the quantum theory
does not play a role
as in the large dimensions
that one should somehow
deviate from the usual differential equations
of the relativistic theory.
Now, when it comes to the question
whether the future theory
will be a pure close interaction theory
or a non-local theory
like in the case of the Yokawa
then of course
it is by no means
only a mathematical problem
but especially
an experimental problem.
What kind of experiments
can one bring out
which of these two alternatives
is to be chosen?
According to the quantum theory
a very precise localization
of a very rigid location
corresponds to
a very large uncertainty
of energy.
So one will only be able
about this question
of local or non-local theory
through the collision processes
of very large energy
between different elementary particles.
Nowadays we can observe collisions
in cosmic radiation
between particles
of energies
of the order of 1000 billion
electron volts or more.
And it is precisely these collisions
which have to do
with a local or non-local theory.
I would like to put three alternatives
side by side
and say briefly
how they would express themselves
in the experiment.
So one possibility would be
that it is a strict close interaction theory
and that it is a close interaction theory
with a so-called small interaction
i.e. with an interaction
which is only a small coupling
between different fields.
If that were the case
only one new elementary particle
is radiated
one or in the worst case
only very few
that this one elementary particle
but in general
can be radiated with quite high energy
that about half
of the total energy
can take part in it.
Now the experiments
certainly do not look like this.
It does not look like
that this really happens in nature.
We can secondly assume
that the theory
is a non-local theory
i.e. a theory
in which the elementary particles
are smeared over an area
let's say from the largest to the smallest
to minus 13 cm
and in this case
it is quite equal
if we assume that the interaction
is very large and very small.
One would then assume
that in any case
only relatively little energy
can be transmitted.
It may be that
depending on how things lie
and how the interaction is
either many elementary particles
arise or few
but in any case
it should never occur
or only extremely rarely occur
that very high energy
is transmitted to an elementary particle
because this smearing
over a certain finite space
means a reduction
of the probability
of high energy
and impulse transmission.
It does not look like this
although perhaps the last word
has not yet been mentioned
and there is relatively little
experimental material.
And finally we can
think of the theory
as a close interaction theory
but a close interaction theory
with a very strong interaction effect.
Then it would be to expect
that in any case
very many elementary particles
arise
but these elementary particles
relatively often
with small energy
and only relatively rarely
with greater energy.
We tried in Göttingen
some time ago
to pursue this possibility
and to describe this emission
of elementary particles
at the impact of two energy-rich particles
in a similar way
as you would describe
an explosion wave
that is, as a non-linear wave equation
but as a real close interaction equation.
And in doing so
you come to results
that seem to fit
reasonably to the experience.
I say seem to fit
because even there
the files are certainly not yet closed
and so far
there is relatively little
experimental material.
These impact processes
of very high energy
are relatively rare
and when Mr. Paul
recently showed you pictures
of such impact processes
then you must not conclude
that you find hundreds of such pictures
every day
but that they were museum specimens
that we have just presented
because they are particularly beautiful.
So these impact processes
must be examined
and analyzed in an even greater number
so that we get a clear understanding
of the type of interaction
of the elementary particles.
I should perhaps mention at this point
that other attempts have been made
to expand the mathematical possibilities.
For example, Snyder tried
to introduce a quantization
of space and time
and Mr. Dirac presented here
a possibility
to expand the relativistic quantum mechanics
through the assumption
of an absolute time variable.
So in many different places
attempts are made
to expand the mathematical forms
in such a way
that they finally fit
the area of experience
but these attempts are not yet closed.
And with that I come to the end.
I have described the difficulties
and ways out of the difficulties
and now I just want to say at the end
what we can do concretely
and what is actually being done
in the world now
to solve these difficulties.
Well, in one way
it is essentially
a problem for an experimental physicist.
We have to better examine
these processes of high energy,
we have to know
which elementary particles there are,
we have to learn
in particular their interaction,
for example, to get to know
which elementary particles
cannot be transformed into others
and we have to draw conclusions
on the symmetry of the equations
of the matter
which we can not write down
but which will certainly be given later.
Practically,
there are two possibilities for this.
The balloon ascents
which Mr. Powell talked about
and then large machines.
And maybe I should mention
in this context
that just yesterday in Paris
a European convention was signed
after which the participating states
including West Germany
decided to establish
a nuclear physics institute in Geneva
in which a very large machine
will be at the center
with which it is possible
to accelerate elementary particles
up to about 30 billion electron volts.
Once this machine is in place
it will certainly give us
extremely valuable conclusions
on the behavior of elementary particles.
As long as the machine does not exist
one will be dependent
on the cosmic radiation
or on the smaller machines
which are already running in America
and which will not exist anymore.
In addition to all this experimental work
one can of course also do something
as a theorist.
One can,
according to mathematical structures
which at least qualitatively represent
what we observe.
Qualitatively we know very well
how it goes with physics of elementary particles.
We know that there are a number
of stable and actually unstable particles.
We know that they can transform each other.
So it should be possible
at least qualitatively
It has not been possible so far
to write such schemes
which are at the same time mathematically meaningful
and which correctly represent the qualitative.
So there is a problem
which one can also hope
that it will be solved in the next few years
but that it is still not solved.
But even if this solution is difficult
it does not mean any kind of pessimism.
One can still assume
that the future physics of elementary particles
will look very simple.
But the way to that is difficult
and it is very difficult
to come up with these very special
mathematical structures
which these physics of elementary particles
will later master.
And with that I would like to conclude.
