Folder C133. Fact and Inference Continued. Feyerabend. October–December 1967

Abstract

In answer to your question, I have no particular preferences about the duplication of my preprints. Let the university make the usual charges, and keep the money for me, if this is indeed the usual practice.

OCT 10th, 1967 [Date added – CT.] Dear Jeffrey

I received your recent letter and card, for which we thank you.

In answer to your question, I have no particular preferences about the duplication of my preprints. Let the university make the usual charges, and keep the money for me, if this is indeed the usual practice.

Your comments on fact and inference are very much to the point. The question of beauty and ugliness is involved in a very deep way. Probably we would need several days discussion to get really far into these issues. However, I shall try to sketch a bit of what is implied in the relationships of beauty, fact, truth, harmony, totality, peace, etc., etc.

Let us begin with the boy who is stealing. You say that there are two inferences:

(a) The boy is stealing.

(b) The boy is a thief.

How do I justify calling (a) a valid inference and (b) a source of delusion?

It is true that (a) and (b) are both inferences. But there is a tremendous difference in their basic qualities. (a) refers to the directly observable actions of the boy, and makes no assumptions about his deep more and or less permanent structure. In other words, (a) refers only to a contingent feature of the boy’s actions. By observing these carefully enough, we can answer the question

(A) Is the boy stealing?

Note that in the step between Inference_n and Fact_n+1, Inference_n is first changed into Question_n. (This step may be largely tacit and not verbally expressed.) Question_n calls for Observation_n and the answer thus supplied is Fact_n+1.

Let us now do this with Inference(b). We ask:

Is the boy a thief?

Note however that this question contains a deeper tacit assumption; i.e., that human nature is such a simple and mechanical sort of thing that this question can have a “yes” or “no” answer. Thus, we can ask “Is this an electric motor?” By studying the structure carefully we can answer “Yes, it is an electric motor”, implying by this that when the switch is pressed, the motor must turn. But can we obtain a corresponding knowledge of the boy’s structure, which would imply that when the boy sees something that he wants, this presses the mental “switch” so that he will steal, just as the motor will turn, when its switch is pressed? Evidently, this is a vast question that cannot be answered by a few observations of the boy’s behavior. Nor can one even see how it could be answered by a detailed analysis of his brain and nervous system. Moreover, there is a vast quantity of knowledge available to everyone which suggests that the situation is more complex than this. For example, each man knows that he is similar to the boy in many basic ways. Yet, few men will agree willingly to having their own faults and weaknesses attributed to an “intrinsically wicked nature”, implying that they must always be ugly and wicked, unless it is “beaten out of them.” Rather, they ask for affection and sympathetic understanding of the deeper origins of their faults. But in thinking of the boy, they forget all this.

So what can we learn from this? Mainly that each question contains tacit assumptions, of which one has to be aware. The question “Is the boy stealing?” contains the tacit assumption that the boy is capable of a certain behavior called “stealing”. We know enough of people to realize that everyone is capable of this behavior. So this is a valid question, because its basic tacit assumptions have been adequately confirmed. The question “Is the boy a thief?” assumes tacitly that the human mind is capable of having a thieving structure. This assumption is at best an inference from very limited and inadequate data, which moreover ignores vast quantities of contrary data that are available to everyone. So it is not a fact. Question (b) therefore confuses inference with fact, not in the form of an explicit statement, but rather in the form of tacit inferences, of which we are actually unaware. This is the really dangerous kind of inference. For without knowing it, we have taken certain inferences (e.g., about human nature) to be facts, before we have posed the questions and made the observations that could establish the real fact about them (i.e., truth or falsity). This very step is the beginning of delusion. And from here follows endless confusion. For as we obtain new data, we have to shift our consciously held inferences in order to accommodate this data, without altering our unconscious tacit influences. In this way, the very evidence that could show us the falsity of our tacit assumptions becomes the source of new false assumptions that make our minds more confused than ever.

For example, the inference “The boy’s thieving proclivities can be beaten out of him” is generally found not to work. He goes on stealing. From this, one might be led to question the assumption that a human being can be a thief, who has a thieving structure. But one doesn’t realize that this is an assumption. Rather, it seems to be an “evident fact”. So instead one infers that the boy is an “incorrigible criminal”.

If every question can contain deep and largely unconscious tacit assumptions which are easily mistaken for facts (i.e., highly confirmed inferences), what is to be done to be free of delusion and confusion? Evidently, there is no verbal formula that could adequately answer this question. Rather what is called for is a deep and extensive sensitivity to what is false, disharmonious, ugly, and full of conflict. Evidently, one gets a feeling that there is something wrong with the question “Is the boy a thief or not?” If one pays serious attention to this feeling, being sensitive to all its subtle nuances, one is led to a more detailed perception of what is wrong with the question. The same holds more generally. Thus, in science one may ask “Is the electron in a certain position or not?” This question presupposes a certain “particle-like” structure that the electron probably does not actually have. When we find that the question does not have a clear answer, we should be led to question the tacit assumptions behind the question. But instead, we may make new inferences to accommodate the facts (e.g., that the electron’s structure function is incomprehensible, and that only computation of its phenomenally observed behavior is possible). These assumptions add to the confusion (as in the case of the boy).

So fundamentally, to see what is true or false (and therefore to perceive the fact) depends on sensitivity to beauty and ugliness, harmony and disharmony, etc. This evidently takes place at great depths in our mental processes. People like to say that it is “personal” and “subjective”, therefore of no concern to the study of science, which is based on fact. Yet, without it, there can be no real fact, at all. Rather, a mind that is insensitive to beauty and ugliness must ultimately delude itself.

So you are right to say that we confuse fact and inference because the mind is already in an ugly state. But it is not enough just to say so and to leave it at that. Rather, we can go further. Firstly, we can be aware of the place where insensitivity to ugliness gives rise to delusion – i.e., it enables us to accept tacit assumptions or inferences as confirmed facts. To be aware of how this it is happening will be at least a step in ending the whole process of delusion and confusion, leading to evil. And secondly we can inquire more deeply into why we are insensitive to ugliness.

As a first step in this inquiry, let us notice that when I say “That boy is a thief”, there is an implied comparison to me, meaning “I am not a thief but a good virtuous person”. The thought of this comparison induces a fleeting sensation of pleasure, security, satisfaction, etc. If one doesn’t look at it too closely, this pleasure seems to be “beautiful”, to have in it a sense of “harmony”, “well being”, “euphoria”, etc. Of course, if one does scrutinize it carefully, one sees that it is ugly, hence painful. That is why the sense of pleasure is so fleeting. For it can be “recognized” as beautiful only if one does not get a good look at it.

The real issue here is the confusion of glamor and beauty. Glamor is a superficial imitation of beauty, which “gets by” for a short time. Evidently, most of what is called “beauty” in modern society is actually glamor. And the same was true in ancient society. Thus, children were told stories about the “heroes” in which the murderous and destructive actions of these individuals were glamorized and made to appear to be noble and beautiful. Likewise, people tend to “glamorize” their own qualities, so that these appear to be beautiful, when they are actually ugly. For example, a person’s violent and aggressive behavior is glamorized as “strength of character”, while another person’s timidity and fear is glamorized as “gentleness”.

When the mind is full of false beauty, which is really glamor, then it has inevitably lost the sensitivity needed show up the deeper tacit assumptions that are in the questions, with which we establish the fact. For example a Nazi who was trying to be “factual” might say: “Are the Jews really of the same nature as Nordic Germans or not?” This very question contains the tacit assumption that people can be divided into well defined groups, which can meaningfully be compared with regard to the values in their deeper natures. This assumption makes possible the pleasant glow of euphoria and glamorized substitute for beauty, which comes when one thinks: “In a comparison between my group and other groups, my group comes out supremely superior.” In the false light of this glow, the subtle feeling that “something is wrong” can no longer be noticed. So one is trapped. One has mistaken an inference for a fact. The resulting glow of “pleasure” creates a spurious impression of beauty, and therefore the tacit inference “Everything I think is right”. Any hesitant step to perceiving the underlying ugliness interrupts this glamorized “beauty”. The interruption of “beauty” is then interpreted as ugliness. So one confuses the perception of ugliness with the ugliness that is perceived. This confusion means that the very evidence of the falsity of the structure of one’s thought becomes the means of preventing one from seeing this falsity. All evidence of this kind is re-interpreted and accommodated, in an ever growing structure of new false inferences, thus giving rise to never ending confusion.

Svetlana Stalin illustrates this nicely in her account of her father¹, who was always demanding “facts” about the people who were “scheming against him”. This very question contained the tacit assumption that every man was either a friend of Stalin or an enemy of his. So all sorts of really harmless actions were inferred to be “plots” against him. Then when Stalin’s wife questioned the actions of Beria, Stalin always asked for “facts” against him. Since Beria was the only possible source of such “facts”, it was plain that none could ever be forthcoming. So Beria was always a “friend” and others against whom Beria gave “evidence” were always “enemies”. In this way, vast evil and destruction were set in motion, from which tens of millions of people suffered and died.

It seems clear that when the mind is in a noisy state of mistaking glamor for beauty, then it cannot be sensitive to the subtle inferences in the questions, with the aid of which it is establishing the fact. For if there is already the roar of 100 decibels, the tiny “sounds” that indicate that “something is wrong” cannot be heard. So what is needed is a mind that is in a general state of harmony, which will be felt or sensed as “silence” or “emptiness”, in the depths. The slightest disharmony on the surface will then “stand out” like a sore thumb. In the presence of such perception, the mind will explore and move until a new harmony is established. This exploration will contain as a by-product the perception of the falsity of certain inferences, and (where appropriate) the proposal of new inferences. It is essential that this be a by-product of a deeper non-verbal movement of harmony. For it is meaningless to try to establish harmony merely by manipulating the superficial content of ideas and feelings, (by adapting one’s words and other symbolic structures).

In a way, we can compare the situation to the use of a galvanometer. In some circuits, this instrument is used “positively” to measure the electric current. But in a Wheatstone bridge, it is used “negatively” to indicate “imbalance”, which can be regarded as a kind of “disharmony”. Here, its function is to enable the operator to change the circuit until it ceases to function. So its function is to bring about its non-function. Similarly, thought can function “positively” as a “measure” of the world in which we live. But its deeper function is “negative”. That is to say, we think when there is a “problem” to be solved. A problem indicates that there is disharmony somewhere, either inwardly in the mind, or outwardly in the perceived world. This problem implies a contradiction somewhere, in our total structure of inferences and results of observations. This contradiction, if it is at all subtle, is first indicated by a feeling of disharmony, which “stands out” against the harmonious “silence” in the depths of the mind. Creative movement originating in these depths “alters the circuits” until the sense of contradiction ceases. So in its negative role, the function of thought is to indicate disharmony and thus help the deeper levels of the mind to bring about harmony, indicated by the ceasing of the function of thought. When this happens, our inferences will generally have a different structure. In the process, we also observe, inwardly and outwardly, to see the truth and falsity of our inferences.

You can perhaps see that the questions you raise go very deep. Indeed, if there were more opportunity, I would show that they lead to the questions of time, space, pleasure, the nature of thought, and the delusory character of the separation of the observer and the observed. However, this will have to be for the future.

Meanwhile, I am sending by surface mail some talks given by J. Krishnamurti, whom I once mentioned to you. Here, you will find all these questions discussed from a more psychological point of view. Perhaps this is what is called for, before going very much more deeply into all these issues.

Best regards

David Bohm

P.S. About von Neumann’s saying hidden variables imply that qu. mchs. is “objectively false” I had an idea. The trouble is an ambiguity in these words. Do they mean only that qu. mchs. is an approximation and not an absolute truth? Or do they mean that qu. mchs. is already “objectively true”, so that hidden variables are impossible? It was really v. Neumann’s reponsibility to clarify the issue here. To do this, he should have refrained from discussing truth or falsity in qu. mchs. Rather he should have said: “The axioms of qu. mchs. are logically incompatible with hidden variables”. Then it would have been clear that hidden variables are still possible, because the axioms of qu. mchs. have been verified factually only to limited approx., in limited domains, etc. By saying “objectively false”, he confused the issue, because, for all we know, they $\underline{\underline{\text {are}}}$ “objectively false” in more accurate measurements. We have in reality no way of knowing whether the axioms are “objectively true” or not. All we know is that inferences drawn from them have been confirmed in a certain domain up to a certain approximation (as is true also, for example, of Euclidean geometry). So the issue of “objective falsity” of any set of actions is a complete red herring. One uses these words in this context only because one has a tacit empiricist metaphysics, which identifies (and therefore confuses) truth with verification of factual inferences. No axioms are ever true or false. Only inferences drawn from them in the field of observation can be true or false.

This is a crucial point, which illustrates the structure

The Answer to Question_n establishes the truth or falsity of Inference_n but not of Assumptions_n (which may be tacit as well as explicit).

Example:

Assumptions_n include the axioms of Euclidean geometry plus a lot of tacit “background” assumptions.

Fact_n is that we have been able to make measurements of various kinds, which form, as far as has been seen, a coherent structure, within these axioms and assumptions.

Inference_n is that the indefinite extension of such measures (over long distances) will yield similar results.

Question_n is “Will they do so?” This leads us to design instruments to answer it, as well as new theories showing how Euclidean geometry might fail.

Fact_n+1 is either “They do” or “They don’t”.

Von Neumann would confuse the issue by saying that if the answer is “They don’t”, Euclidean geometry has to be “objectively false”. Our response to this is that nobody can know finally and for sure whether Euclidean geometry is “objectively false” or not. Indeed, this is just why we are testing such inferences, with the aid of questions drawn from alternative systems of axioms. By using the phrase “objectively false”, von Neumann tends to lure or trap us into the further inference that since Euclidean geometry is thus far “objectively true”, this whole line of reasoning is unnecessary and fruitless. Of course defenders of v. Neumann can always say “He didn’t really mean this”. But this is like the question of “How to Win at Games Without Actually Cheating.”²

It was really von Neumann’s (and everybody’s) responsibility to phrase his conclusions, so that a lazy or tired mind is not likely to dig itself a trap and fall into it. The fact that so many people were convinced by v. Neumann that “hidden variables are impossible” is a proof of the effectiveness of this “trap”, made by the ambiguous phrase “objectively false”.

Oct 11, 1967

Dear Jeff

This is a brief supplement to yesterday’s letter.

First, let me give a more detailed description of the hierarchy of fact and inference. The first step is that one is always aware of previous inferences and factual observations. Then one has all sorts of assumptions. These too are inferences, but of a more general nature, drawn from a vast body of knowledge. Our most general assumptions constitute the metaphysics of the moment. I stress the “momentary” character of metaphysics, because it is really always changing in subtle ways. However, we tend to try to formulate it as if it never changed. This is, of course, a contradiction. The resulting confusion can be avoided by recognising and realizing (being aware) that we have a metaphysics and that it is inevitably always changing.

So we have the process:

Fact_n + Assumptive inferences_n $\rightarrow$ conclusive inferences_n

Conclusive inferences may include logical deductions from fact_n + assumptions_n, but usually more general processes are involved. Thus, there may be associative inferences (fact_n + assumption_n remind one of analogous situations). Then there are the inferences which generalize previous influences, as well as inferences which restrict the generality of previous inferences (thus generalizing the process of restriction). Besides, there are probably other kinds of inferences (e.g., hunches, intuitions, etc.)

The next step is:

Conclusive inferences_n $\rightarrow$ Question_n $\rightarrow$ Observation_n $\rightarrow$ Fact_n+1

We have to be aware of the whole process. Thus, when we formulate Question_n, we may feel that “something is wrong”, and this feeling may indicate that we are overgeneralizing our inferences about the factuality of some of our deeper assumptive inferences.

Our axioms are part of our general assumptive inferences. We do not directly test these assumptive inferences. Rather, we draw further inferences from our axioms, and test these latter inferences by observation. So our axioms can neither be “objectively true” nor “objectively false”. They can be compared and criticized only through their ability to lead to lower order inferences that agree with what is observed.

Here is where von Neumann’s ambiguity introduced confusion. When he said that qu. mchs. would have to be “objectively false” to allow hidden variables, he was playing on the ambiguity between the three phrases:

Phrase (A) may be interpreted as meaning either (B) or (C). Evidently, von Neumann’s statement would be true, if one took it to mean “hidden variables imply that qu. mchs. is objectively falsifiable.” This statement is obviously correct, but it is so obvious that nobody would even bother to say it, if that is all he meant to say. Therefore, when one reads von Neumann’s statement, one is inevitably led to search for a “deeper meaning”. This search is, of course, a trap, because his statement has no deeper meaning. But as soon as one begins the search, one considers meaning B. “If hidden variables exist, qu. mchs. would already have to be objectively falsified. But it has not been thus falsified, so that hidden variables therefore do not exist.” This argument is of course a non-sequitur. Nevertheless, it is tacitly what any reader of v. Neumann would tend to do, unless he were extremely awake and alert. Because v. Neumann has so much status and authority, few physicists are ready to believe that von Neumann would say something obvious, in a form which suggests something deep. So even those who suspected that something is wrong were frightened out of really inquiring deeply into what is wrong.

Even in the relationship of statistical mchs. and thermodynamics, one can see that if statistical mechanical assumptions are true, then thermodynamics must be “objectively falsifiable” (e.g., by fluctuation phenomena). By using ambiguous phraseology (i.e., “objectively false”) v. Neumann obscured the real significance of his theorem, which is this:

Quantum mechanical axioms imply that hidden variable explanations which attempt to treat quantum ensembles as simple averages of deeper deterministic ensembles will not work. (So that we cannot use ideas analogous to Gibbs ensembles, to explain quantum mechanics).

As you have pointed out, this means that the motion of individual members of hidden variable ensembles must depend in an inseparable way on the large scale environmental parameters.

I think that this is what has to be stressed in any further papers on the subject. The notion of structural process is a natural way of bringing in such a mutual dependence, through the idea of reflective function. Each level reflects the others. Mechanistic theories regard the macro-level as a reflection of the micro-level. But they regard the micro-level as “existing in itself” because it is the “fundamental reality”. Its laws do not depend on macro-parameters. Only its contingent features (e.g., initial conditions) can reflect the macro-parameters. Our new idea is that laws of the macro-level reflect both necessary and contingent features of the micro-level, so the laws of the micro-level reflect both necessary and contingent features of the macro-level.

Best regards

David Bohm

Oct 16, 1967

Dear Jeff,

I have just given a talk on our paper in Birmingham, from which I have, I think, learned something about how to present our ideas more effectively.

I began by saying that the usual interpretation of qu. theory has as many shades and varieties as there are interpreters. But there are, broadly speaking, two poles, represented by Bohr and by von Neumann.

Von Neumann adopts the idea of mind-matter dualism as a basis. He defines subject and object, placing the former on the “classical” side of the “cut” and the latter on the “quantum” side. But by doing this, he effectively treats the “subject” as yet another object. Thus, he implicitly defines two totally different kinds of “objects”, one of which is called the “observed” while the other is called the “observer”. This is similar to Descartes definition of mind as “thinking substance” and matter as “extended substance”. In both cases the problem arises as to how two such totally different domains (classical and quantum, or mind and matter) can be related. Descartes proposed that in his own mysterious way, God saw to the relationship of mind and matter. But von Neumann never proposed a similar explanation of how classical and quantum “objects” were to be related in his theory. The whole question is left “up in the air”, in a very unsatisfactory way. (e.g., by appealing to a principle of psychophysical parallelism).

Bohr is unclear on this question. At times, he seems to accept the subject-object dualism, calling the observing instrument the “subject”. But more deeply he has an entirely different notion, i.e., that the essential question is linguistic, having to do with unambiguous communication. Clearly, language is both subjective and objective. It is subjective with regard to its meaning, which varies subtly from person to person, and from moment to moment, even in a given individual. It can, however, be objective with regard to that part of its content that is unambiguously communicable. For example, when we use the word “chair”, we can, generally speaking, always agree on which objects are chairs, and on what can be done with them.

In Bohr’s view, the refinement of unambiguous language leads inevitably to the conceptual structure of classical physics – i.e., to the analysis of the world into particles and fields, which are completely described, in principle, by suitable sets of canonical pairs of variables (like position and momentum).

Classical physics is characterized by two related structures

1. Descriptive

2. Inferential

Clearly, we can use position and momentum variables to describe the states and changes of states of classical systems. In addition, we can draw inferences, in this framework, with regard to the basic laws of motion (e.g., Hamilton’s equation).

When experiment disagrees with classical theory, the first step is to try to change inferential structure (e.g., change the Hamiltonian, introduce new fields and particles, etc.). But quantum mechanics went much further than this. It implied that the entire inferential structure breaks down altogether and even the descriptive structure underwent a decisive change. For quantum mechanics implies the denial of the classical notion that position and momentum variables can be described together, at least beyond the limits of precision set by the uncertainty principle.

Bohr interpreted this situation in terms of the principle of complementarity. That is, the arrangements of matter needed to measure complementary variables precisely are incompatible, because of the quantum of action. Therefore, there is an inherent ambiguity in these variables. When this ambiguity is relevant, there will be a corresponding ambiguity in the separation of subject and object (observer and observed). For it will no longer be possible to describe a physical interaction in enough detail to see whether it originates in the instrument or in the electron. This ambiguity is reflected in our language, which uses the same terms to describe the conditions of the experiment (momentum or position measurement) along with the results of the experiment, the two being interwoven in an inextricable way. So Bohr does not really assume mind-matter dualism (though at times he may appear to do so, because he doesn’t make himself clear). Nor is he a positivist, since he puts language first, before the question of empirical observation. To Bohr, what can be observed depends on our language, which defines our terms and limits what we can recognize and communicate. In a way, this is like what Feyerabend said, in his article in Beyond the Edge of Certainty. But the key difference is that Bohr restricts the language of description of fact at the perceptual level to that of classical physics (limited by the principle of complementarity), while Feyerabend would allow all sorts of languages in this domain.

In Bohr’s view, the algorithms of quantum theory play the role of a metalanguage, which makes statements about the language, that is used to describe the perceivable phenomena. Because there are probability statements, the metalanguage is ambiguously related to the language.

Now, in Bohr’s view, the inferential statements are determined by the metalanguage, while the original (classical) language is being used only descriptively. Because of the ambiguous relationship of the metalanguage to the language, it follows that inferential statements have just the right degree of ambiguity to be compatible with the principle of complementarity. For example, the ambiguity of p (when q is fairly well defined) is reflected faithfully in the algorithm, which gives a corresponding probability distribution of p.

Now, my criticism of Bohr is that I feel he is wrong to project these linguistic questions into the quantum domain. In the psychological domain, one must consider the dual role of language, which is both subjective and objective. Some kind of metalanguage is needed to discuss this question. And eventually, it carries us to direct awareness, beyond what can be described in language, where subject and object are one. But in my view, there is no reason to regard the quantum algorithm as a metalanguage. This becomes necessary only when we assume that classical physics gives us the unique unambiguous language. I would rather explore the possibility of new languages (and conceptual structures) for describing the domain of directly observable large scale phenomena. These new languages would have room for a direct phenomenal interpretation of some of the terms in the algorithm of quantum mechanics.

One can further criticize Bohr’s point of view in that it makes the details of the individual process undescribable. Is Bohr not here confusing an inference with a confirmed fact? Certainly if we accept Bohr’s assumptions, the details will be indescribable, almost by definition. But the deeper question is that of whether Bohr’s assumptions are inevitable transcriptions of already observed facts, or whether they are merely proposals that seemed plausible to Bohr and his followers. It is like the boy who is stealing. To some people, the question “Is he a thief?” is a relevant one, because the notion of human nature as divided between “honest” and “thieving” is a well established fact to them. They can point to the empirical possibility of dividing people between those who steal and those who do not as confirmation of this view. The critics of this view must question the entire set of structural assumptions about human nature. Similarly, to criticize Bohr adequately, one must question his whole set of assumptions about language. Are they really adequately confirmed or not? Our own papers help in my view, to make such criticism possible.

Here, I emphasized the need for alternative sets of metaphysics, as a strategy to help keep science from being frozen in a closed circle of concepts. As Feyerabend points out, what we are ready to observe depends on our general theoretical structures, which provide a basic language for description and communication of observed fact. For this very reason, we need alternative theories, leading to alternative languages, if we are even to see what it means to refute a general theoretical point of view. Theories are not merely being accumulated, to give predictions and explanations. In addition, they confront each other, to permit a dynamic process of fundamental change and development.

One can illustrate the strategic role of our ideas by reconsidering the von Neumann “proof” that there are no hidden variables. Of course, he starts from certain axioms, which are probably not unreasonable “models” of the kind of quantum mechanics that physicists have been using fairly “successfully” over a period of years. But he says that according to his “proof”, quantum mechanics would have to be “objectively false” if hidden variable explanations of it are to be possible.

Here, we must ask “Isn’t von Neumann confusing axioms, which are always inferential in nature, with what he would call ‘objective facts’?” Axioms can never be true or false. This we can see to be a “higher order fact”, by looking carefully at the real structure-function of axioms in the total process of theorizing. In other words, the “philosopher of science” is also trying to establish facts – That is, he must confirm or falsify his higher level “meta-inferences” about the “inferences” used by physicists in their theorizing. I suggest that any competent person will see, if he examines the questions, that axioms are neither true nor false. Rather, we draw inferences from them, which are either confirmed or falsified. Ultimately, as we go to “meta-inferences” and “meta-meta-inferences” we must always appeal to what a competent person will agree to be the perceived truth. (The same is true about the question of whether or not the boy can have a “thieving” nature). There are no “incorrigible facts” at a bottom level. But I do find, factually, that all those with whom I discuss the question seriously do actually agree that axioms can neither be true nor false.

Very probably, in von Neumann’s day, this question was not too clearly understood. The old idea that a theory is either true or false still held sway. So von Neumann confused the inference that axioms can be true or false with a fact. We are now able to see that this is only an inference, and moreover, an inference that is not confirmed by observing the way physics can actually be done. We now see that empirical observation can falsify a set of axioms, without proving them to be “objectively false”. So the real content of von Neumann’s theorem can only be one of the two statements.

A. Hidden variables imply that quantum mechanics is potentially objectively falsifiable.

B.          "          "          "          "          "          "          is already objectively falsified.

Everybody will accept the validity of (A) immediately. But then it seems so obvious that one does not understand why an outstanding physicist like von Neumann made such a fuss about this trivially obvious point.³ So almost inevitably, one assumes that von Neumann must have had interpretation (B) in mind. But since quantum mechanics is thus far “objectively verified”, it follows that hidden variables are really impossible. Thus, von Neumann’s “proof” has led to confusion over the years.

Having cleared this point up, I now asked: “Is von Neumann’s proof really only a triviality, or didn’t he show something interesting, whose significance was hidden by the confusion between “potentially falsifiable” and “actually falsified”? What von Neumann showed was that if there are hidden variables underlying the quantum theory, these cannot be similar to the atomic “hidden variables” that underly classical statistical mechanics. For these latter satisfy the linearity condition

$$\begin{aligned} \langle aA+bB \rangle _{av} = a\langle A \rangle + b \langle B \rangle = a \int \rho (\lambda ) A(\lambda ) d\lambda + b \int \rho (\lambda ) B(\lambda ) d\lambda \end{aligned}$$

Von Neuman showed that if such conditions were satisfied in quantum mechanics, then dispersionless ensembles are impossible. Or alternatively he showed that dispersionless ensembles imply the breakdown of the linearity condition. But the linearity condition is closely related to the assumption that whereas the macro-level is determined by the micro-laws, these latter laws are completely independent of the situation at the macro-level. (In other words, that observables like A and B depend only on the “hidden” micro-variables, $\lambda$, and not on statistical large scale parameters, like T and P, which appear only in the distribution function, $\rho (\lambda )$.) Therefore, we are led, by von Neumann’s analysis, to consider the notion of a set of hidden variables that do not satisfy the linearity condition, and therefore do not satisfy the assumption that the micro-laws are independent of macro-conditions. This leads us to a new view of the structure of the world, which amounts to a new metaphysics. To some extent, Mach foreshadowed such a possibility by proposing that the inertial frame (even for the basic constituents of matter, whatever they may be) depends on the distribution of matter over the whole universe. But here, we are led to go much further, to propose that large-scale structures of many kinds may enter into the form of the micro-laws.

At this point, one sees the need for a more detailed model of such a new metaphysical point of view. Our proposals of hidden variables provide such a model. We see that, as von Neumann showed, our model implies the breakdown of the linearity assumption. In order to recover the linearity property, we propose a new process that randomizes the hidden variables in some unspecified period of time, $\tau$. So by constructing a hidden variable “model”, we actually see the deeper meaning of von Neumann’s analysis. The two steps (von Neumann’s analysis and our model) complement each other. If von Neumann had properly expressed the results of his analysis (without bringing in the confusion of objectively false – falsifiable – and falsified) his analysis would have been a powerful impetus and guide for the inquiry into hidden variable theories.

Our model then brings out several further significant points.

(1) It brings in the possibility of new orders at the level of the phenomenon. For example, if the hidden variables are defined precisely, the order of the results of a series of measurements of “non-commuting observables” is completely defined. And more generally, a non random distribution of hidden variables implies a partial definition of orders of this kind.

Such new orders, if observed, would contradict Bohr’s assumption that all unambiguously communicable orders have to be stated in the language of classical physics (e.g., positions and momenta of various objects). Here, what we can observe and communicate is a certain order of results, that is related unambiguously to the extended algorithms of quantum theory (which now includes hidden variables). So we refute the assertion that the algorithm is only a metalanguage, not directly and unambiguously related to the observable phenomena. In effect, we have brought in a new kind of language, both for describing the phenomena, and for drawing inferences about them.

One can compare the situation here to that prevailing in the beginnings of classical physics. By bringing in the calculus, Newton was able to talk about velocity, acceleration, etc., whereas before, there had been no way to do this. One could only vaguely describe a curve like this:

But now, one could be directed to observe and measure the various derivatives at different positions. Without this new mathematical language, this would have been impossible. Similarly, hidden variables enable us to define, talk about, and observe new properties, such as the order of a discrete set of results, as below:

(2) By defining new “observables”, hidden variables indicate what is needed to test the basic axiomatic structure of quantum mechanics as a whole. No experiment, unguided by new theories, can possibly provide such a test. For no matter what the results of such an experiment may be, one can always imagine a new Hamiltonian, a new particle or group, etc., that might explain the results of a disagreement away, without altering the basic general principles (e.g., new epicycles in the Ptolemaic theory to fit any possible observational facts). A test consists in a confrontation of one theoretical framework with another, to see if the latter does not lead to a valid new descriptive language opening up a wide range of new kinds of observations, permitting new kinds of inferences, that could be further tested, etc., etc.

(3) Our model raises the question of the propagation of the effects of hidden variables in a relativistically covariant way. Our present model does not achieve this result, but it points to two possible modifications:

(a) Quantum-mechanical correlations may actually be propagated at the speed of light or less. This could be tested experimentally in principle. Also, one could try to make an extended model, embodying this feature.

(b) In any case, the question of the maximum speed of propagation is inseparably connected with the basic physical assumptions. Thus, in general relativity, it is given by $g_{\mu \nu } dX^\mu dX^\nu = 0$, so that it depends on the metrical tensor, possibly also the quantum fluctuations of the latter. If there are hidden variables, some new criterion for maximum propagation speed may arise, which involves the hidden variables. The exploration of such a theory is one of the possible lines of further inquiry into the subject.

(4) Our model embodies a dependence of micro-quantities on conditions at the macro-level. Thus, it leads, as I indicated before, to a basically new metaphysics.

Here, we can bring in the notion of a total undivided and indivisible structural process as a natural embodiment of this kind of metaphysics. Each aspect is an abstraction, which reflects the whole in its own way. It is this reflection that makes observation possible. Von Neumann’s notion of a separate observer is thus a “red herring” that serves only to create confusion. Every level, including the human mind, contains a reflection of the whole, and therefore, a reflection of all the other reflections, as well. There is no need to introduce a separate “observer” because the “observer” interpenetrates the whole structure of all existence.

On the other hand, we also do not accept Bohr’s notion of the indescribability of the processes by which we come to learn about things in the universe. For we have given up altogether the attempt to describe phenomena solely in terms of the language of classical physics (just as Newton no longer tried to use epicycles as his main basic description of the phenomena of astronomy).

One important point is to distinguish between the indivisibility of phenomena and their individuality. We account for indivisibility directly by assuming that phenomena are merely aspects, abstracted from a total structural process (e.g., doorways and walls are indivisible and inseparable). Individuality is a more subtle motion. Usually it implies:

(a) Particularity – i.e., distinct difference and separation from other similar phenomena (but of course, not from the structural process as a whole).

(b) There must be an order in the level of the phenomena themselves, both as different individual phenomena are related, and in the inner structure of each phenomena. This order is in essence what is observable in the phenomena. When this order is random, we obtain statistical inferences.

(c) Each phenomenon (as an individual) has a quality of wholeness – i.e., it is itself, on its own level, a limited sort of totality (as well as being inseparable from its deeper levels). This constitutes a kind of closure, that marks off one phenomenon from another. (E.g., a series of musical notes constituting melody A would be one phenomenon, while a series constituting melody B would, in certain contexts, be another phenomenon.) One is seeking a physical theory where the phenomena would have a similar aspect of wholeness, or relative completion. This means that while they can be divided, they will not in general produce the same kind of phenomenon on division. (E.g., the notes of a melody are a different kind of phenomenon from the melody.)

All these qualities have a very different meaning from that given to them by Bohr. This has to be emphasized very strongly. Bohr takes wholeness to be an undefinable linguistic term. That is, the language of classical physics, limited by the principle of complementarity, does not permit us to do other than to discuss the phenomenon in its not further specifiable wholeness.

With best regards

David Bohm

Oct 19, 1967

Dear Jeff

I have received your letter of Oct 9, which I found very interesting. I think it becomes clear from this letter where we are failing to communicate.

The key issue is one that I have already discussed in a letter to you about a week ago, i.e., the tacit inferences that are in the questions, which shape and determine the observations with which we establish the fact. Let me first summarize and extend my views on this subject, and then I will discuss your letter in the light of this summary and extension.

As I indicated, the question “Is this boy a thief or not?” already contains the tacit inference that human nature is a mechanical sort of thing (like an electric motor) that can either have a thieving structure or not have it (or at most, one that could be “altered” mechanically from one structure to another, for example, by beatings, punishments, and rewards). Only a mind that is insensitive because it is full of “noise” (i.e., the confusion of glamor with beauty) could accept this question, without feeling that there is something wrong with this whole way of putting it. Part of this insensitivity comes from wrong but generally accepted notions (i.e., inferences) about the nature of our inferences, which allows us to assume that the fact is always obtainable by answering whatever question the mind happens to pose, at any given moment.

I am now proposing another set of inferences about the nature of the process of “facting”. As a competent person, you will have to observe whether my inferences are true or false (or whether the whole framework of my inferences is irrelevant, because based on wrong assumptions).

What we have been discussing this far is what I call “horizontal” facting. That is to say Fact₁ $\rightarrow$ Inference₁ $\rightarrow$ Question₁ $\rightarrow$ Fact₂ $\rightarrow$ Inference₂, etc. This is merely a time sequence, in which each set of inferences is regarded as independently structured, but having reference or relevance to the others (E.g., inference₂ is formed outside the structure of inference₁, but may imply the falsity of inference₁). But now, we must consider the vertical structure of facting, which is not a time sequence, but rather, the order of orders of the fact at any given moment. Thus, the question “Is the boy a thief?” contains, at that very moment, a certain set of tacit inferences about human nature.

Let us use letters (a,b,c & etc.) to indicate the vertical order of the fact, and numbers (1,2,3, etc.) to indicate the horizontal order. So we consider the question ‘Is the boy a thief?” to be Q¹_a. We now consider inferences I¹_b to be the inferences underlying this question. These inferences (e.g., that human beings can have a thieving mental structure) lead to a certain set of more subtle questions (Q¹_b, $\mathrm{Q}^{1}_{\,\,\,\mathrm{b}^{'}}$, $\mathrm{Q}^{1}_{\,\,\,\mathrm{b}^{''}}$, etc. ) $\equiv$ Q¹_B, which could ascertain the fact: “Is I¹_a true or false?” These questions are answered by careful and sensitive observation of all that is at our disposal.

In answering these questions, we will be tacitly assuming yet further influences, I¹_C $\equiv$ (I¹_c, $\mathrm{I}^{1}_{\,\,\,\mathrm{c}^{'}}$, $\mathrm{I}^{1}_{\,\,\,\mathrm{c}^{''}}$, etc. ) of a lower and more primitive level. For example, we will assume that a given object (O) belongs to someone else and not to the boy. Where necessary, these influences can be tested, by answering further questions Q¹_C $\equiv$ (Q¹_c, $\mathrm{Q}^{1}_{\,\,\,\mathrm{c}^{'}}$, $\mathrm{Q}^{1}_{\,\,\,\mathrm{c}^{''}}$, etc. ) by means of yet further observations.

As we go on with this process, we eventually reach inferences, I¹_p, I¹_q, I¹_r, etc., which are in the domain of ordinary perception (E.g. that object before my eyes is a boy). As a rule, we do not question these, because we regard them as adequately confirmed. But sometimes, we are led to question them, when sensitive and intelligent perception indicates that there is something wrong, that may have its origin in this sphere.

In principle, this process goes deeper and deeper, into the non-verbal movement that is the foundation of awareness, where it “fades away” to merge and amalgamate with the vast and eternally changing unknown totality, in which all that exists, including ourselves, has its foundation, and its being. Ultimately, it is in these depths that truth and falsity are perceived. For what is false leads, as you say, to conflict, i.e., disharmony. As it begins to do so, the still deeper levels of movement begin to end whatever is responsible for this disharmony, and to create a new perception, that is harmonious, because it does not conflict with itself or with further observations of the fact, at various levels.

Some people might object that all this is too vague and perhaps even “mystical”. Yet, there is one evident fact, which reveals an iron-bound necessity: i.e., that whoever is insensitive to beauty and ugliness cannot do other than delude himself on issues that are at all subtle. People who imagine that they can obtain a “solid” foundation for life, outside the need to be sensitive to beauty and ugliness, will inevitably get lost in the quagmire of confusion and delusion.

Nevertheless, it is important to probe more deeply into the question of why people are so anxious to find a “solid and secure foundation” for these deep processes. It is not adequate merely to condemn such efforts, without understanding how each human being, including oneself, has a structure of thought, which makes such an effort inevitable, until its deep roots are understood.

It is not hard to find the answer to this question. For a vast domain of past experience reveals all too clearly that human beings are prone to delude themselves, to believe what gives them pleasure, and to accept romantic and glamorous beliefs, ideals, and notions, even in the face of strong evidence that these are false or confused. Such thinking has led to disaster in a tremendous number of ways, as well as to the rigidification of pleasing metaphysical notions, in such a way as to freeze traditions and beliefs, and interfere with real learning and the genuine progress of science.

As a reaction to all this, people have tried to base everything on the directly perceived fact, at the level of “common sense”, or something of a similar nature (e.g., general consensus, or common usage of language, etc., etc.). No more of these romantic and exciting metaphysical fantasies for them! Rather they base themselves solely on the “solid” ground of empirical observation.

Of course, the difficulty with this position that it too is based solely on “romantic” delusions, adopted because it happens to give a pleasing sense of security to those who can believe in it. Indeed, as we have been seeing in all our correspondence, the “solid” ground of empirical fact begins to dissolve and turn into quicksand, as soon as one begins to scrutinize it with some care. So the empiricist position can be held only if one is not too sensitive, i.e., if the mind suppresses or otherwise evades awareness of the difficulties in this position. Therefore, you are right to say that the empiricist position is “evil”. Yet, you must be careful to note that it is merely a reaction to the opposite position, which is, at bottom, also “evil” and for precisely the same reason, i.e., it accepts pleasing delusions by suppressing sensitivity to disharmony and contradiction. Indeed, in a deep sense, all evil may be said to be just this – to suppress awareness of the whole fact, in order to impose a glamorously pleasing or exciting structure in the perceptions of the mind.

I feel therefore that you are wrong to identify empiricism as the sole root (or even the main root) of all evil. Rather, the truth is that “romantic speculation” and “hard-headed empiricism” are merely superficial variations on the same theme. If one simply condemns empiricism, one is tacitly starting back on the road of “romantic speculation”, on which any idea that the theorist cares to formulate can be considered to be the truth. Of course, the fact is that one’s theoretical ideas provide a descriptive language that shapes our observations. Here, Feyerabend is right to emphasize that the very terms of expression of the observed fact can be altered radically, by a new theory. Yet, this does not remove the need to formulate questions Qⁿ_p, leading to observations, Oⁿ_p, etc., which can confirm or falsify the new theory. And here, the key issue is not in the formulation of facts and inferences. Rather, it is in the sensitivity of the mind, which can detect disharmony anywhere in the vast horizontal and vertical structure of the process of “facting”. Both “romantic speculation” and “hard-headed empiricism” can be criticized as leading to insensitivity – i.e., to a kind of violent imposition of preconceived notions within the structure of perception. That the “romantic” does this is self-evident. But as Feyerabend showed so well, the empiricist does the same thing tacitly, by supposing that at a certain level, the general language for expressing the fact is incorrigible or unalterable.

So in answer to the first page of your letter, I would call your attention to the thousands of years of history, showing how people are ready to accept almost any idea as true, if it is pleasing, exciting, glamorous, convenient, apparently conducive to security, etc. It was in opposition to this tendency to delusion that people tried to insist on getting “incorrigible” facts. But as I have indicated, the notion of an “incorrigible” fact is just another such delusion, adopted because it gives a pleasing sense of security. So the real difficulty is that there was only a superficial change in the human tendency to accept pleasing delusions, a change which continued the same tendency at a new level, thus making the situation more confused than ever.

This is an example of what has gone wrong with human thinking, quite generally. That is, people see that something is wrong. But instead of getting to its roots, they make a superficial change, which continues the old evil at a new level, thus “confounding the confusion”.

Consider, for example, the question of violence. Basically, violence is the imposition of arbitrary ideas or demands on a process, which are in conflict with the real nature of that process. The very structure of society is rooted in violence. Thus, people are trained to conform to certain standards of what they are (not merely as to how they behave). For example, there is the injunction “Be a good boy! Be brave, noble, gentle, obedient, considerate, etc., etc.” Because, the order of the mind is infinitely subtle, it cannot conform to any such injunction. Thus, there is started a deep rooted conflict.

When people inevitably fail to conform to such standards, society sets up a compulsory order. That is, it tries violently to impose these standards. Such violence incites further violence, and counter-violence, etc., etc.

Eventually, people begin to see that violence is destructive. But instead of getting to its root, they form the ideal of “non-violence”. They then try to impose this ideal on themselves and on other people (e.g., by persuasion or propaganda). But they do not perceive that such an imposition merely continues the reality of violence, at a new level, where its effects are even more confusing than they were before. And indeed, all who proclaimed themselves as “non-violent” (e.g., Gandhi, the American Civil Rights Movement, etc.) actually initiated a great deal of open and manifest violence (e.g., riots, etc.) This is no accident. For one who proclaims non-violence is by that very act being violent, in the root of his thought.

Indeed, all our thought tends to be violent in structure. For because of glamor and the pleasure-pain principle, it tends to suppress all thoughts and perceptions that are in conflict with it, and to stimulate others that appear to back it up. When these efforts at pleasure are frustrated, then violence is continued in an outward and manifest reaction, aimed at suppressing or destroying the frustrating factor, and encouraging factors that work in a contrary direction. So violence in action is merely a continuation of violence in thought (or as von Treitschke put it: “War is the continuation of politics by other means.”)⁴

Therefore, to end violence, we need a deep and radical change in the entire violent structure of thought. Even when thought appears to be “peaceful” it is usually potentially violent. Either one is “worn out” from violence, so that the mind is momentarily “quiet” because it is deadened and insensitive. Or else, one is deluding himself into believing that all is beautiful and right. But let a disturbing factor enter our perceptions. We immediately experience a violent reaction. So at present, our “peace” is like the interlude between wars – not real harmony, but rather, a period of rest and recuperation, so that we can fight again.

And here, there is vast confusion. Many people glamorize violence as bravery, beautiful nobility, etc. Others say that it is not violent to react in self-defence, or when one is “in the right”. Yet one cannot defend anything, even what is “right”, except by some sort of violent attack on the other person. This in turn calls for a violent “defense” on his part, leading to a counter attack. Thus, defense leads inevitably to “escalation”.

Then there are those who say “Don’t defend. Just submit.” This is just another form of violence. For in order to avoid insecurity and unpleasant disturbances, one violently imposes the other person’s demands on himself. Actually, this cannot be done. The mind is too far beyond measure. Another part will resist, thus initiating deeper and subtler forms of violence.

I hope you see how deep and vast are the implications of the whole question of violence. Violence is rooted all pervasively in the over-all structure of our thought. Merely to change this structure superficially will be to continue violence at new levels that are more confused than ever. The ending of violence requires a total harmony, peace and silence, starting from the deepest levels of movement of the mind. And only this can lead to true perception on subtle issues.

The whole question is discussed in another way by Krishnamurti, some of whose writings you should soon receive. Meanwhile, I would like to suggest that when we respond to the ideas of another person, and find them wrong because they are arbitrary and violent impositions on the structure of the fact, it is necessary not to continue a similar violence at a new level, by defending oneself against the other person’s ideas. For each defense is an attack, which violently imposes one’s own ideas, in one way or another. Rather, one has to realize that as soon as one is aware of someone else’s ideas, these ideas are his own. So when I attack your ideas, I am really attacking my own thought, which is “displaying” what I have “made” of your ideas. This is a meaningless and destructive kind of conflict. I need another response, of a very different kind. I have to understand all ideas, regardless of whether they are “mine” or “yours”. Those that are false have to be dropped by the creative process that I have described in a previous letter. If I attack “your” ideas, in order to defend “my” ideas, I really continue both, in perpetual conflict. But if I understand all ideas, in a “neutral” way, then I may get free of what is false about both “my” ideas and “your” ideas.

Consider for example one’s reaction to empiricism. Evidently, one may not like empiricism, because it contains arbitrary, violent elements. In order to defend oneself against these, one makes an attack on empiricism. But this is just what the empiricist is doing. He finds the “romantic metaphysician” imposing arbitrary notions, in a violent way. So he sets up a “defense” against metaphysics.

Thus, on P.2 of your letter, you say that you can understand my saying that a proposition, P, is not a fact but an inference, only within the framework of the radical empiricist’s notion that there exist “incorrigible” facts. I wish, however, to ask you whether this assumption is not just an effort to “defend” yourself against radical empiricism. As such, it inevitably leads you to continue the basic structure of radical empiricism at a new level.

The essence of the radical empiricist position is that there is some deepest level of observational and linguistic terms that are “incorrigible”. Probably he would say that the meaning of the word “fact” belongs to this deepest level. Similarly, you say that I must also be accepting the “factual background” as part of the incorrigible “soil” of observational results on which the “tree of knowledge” can grow. If I cannot tell when something is at least not a fact, I cannot talk about confusing inference with fact. But is not this just another example of radical empiricism, which implies that all fundamental terms can be defined explicitly (either positively or negatively) and thus observed?

I am saying, in contrast to the radical empiricist, that fundamental terms are never explicitly definable. Rather, their meanings are indicated by the order and structure of relationships, revealed in how one uses these terms. This holds true especially for words like “fact” and “inference”. The first step is to establish a distinction between what is observed and what is inferred. This is merely a step in establishing the structure of our thought, not of its content. This distinction is not to be hard and fast, but rather, taken as two poles in a process (like moving Northward and moving Southward). We cannot escape that each element in our thought appears in its observational aspect and in its inferential aspect. Indeed, it is just because of this that inference can so easily be confused with a description of what has been observed.

I am not trying to say that what is observed is “incorrigible”. Far from it, I say that inference can often show the need to “correct” previous observations. But what I emphasize is the structural feature that a given term is ambiguous, in the sense that unless further specified, it may be a description of observation or an inference. Not to be fully aware of this ambiguity is to become a slave of arbitrary factors, which will lead a person to treat the inferential applications of a given term as if they were observational applications. This gives rise to delusion.

Now the whole process by which observation and inference interweave maybe called “facting”. In addition, common usage has it that an inference that is well confirmed by observation on many levels is to be called a “fact”. (E.g., the existence of Australia is a “fact” of this kind). If you, in your defense against empiricism, try to get people to cease to take such “facts” seriously, you will be only ceding the field of battle to the empiricists. For then people will rightly worry that without “facts”, they have no way to avoid becoming “romantic metaphysicians.”

I emphasize that my thesis about the “fact” deals with the structure and not the content of thought. For example, consider the word “not”. Can you define this word? Can you at least say what it is not? Evidently, you can’t do it, without using the word “not” or some equivalent term (e.g., limitation, negation, etc., etc.). The word “not” has no content at all. It is a basic term that helps establish the structure of every language. And whatever “positive” meaning is given to this word, you can always add: “It is not just this, but something more besides.” For this reason, no computer can ever really do reasoning. At best, it can imitate some aspects of reasoning, in a framework determined by how the word “not” is interpreted in terms of the operations of the computer (E.g., as an operation in a Boolean algebra).

Any notion that a fundamental word can be defined is in essence part of the radical empiricist position. Somehow, one supposes that one can directly perceive the unique meaning of words like “fact” or “inference”. But if we give up the empiricist position in physics, we must, ipso facto, give it up in epistemology and in theories of language. The only thing we ask is that, given a structure in which words are used, can we discover in observation something that corresponds to whatever this structure may be discovered to mean to us?

When I say someone is confusing inference with fact, I mean that any competent person should be able, in his own thought, to see that the inferential aspect of a given term is being treated in an order that would be appropriate only for the factual aspect of that term. Thus, if someone asks: “Is the boy a thief”, he is treating as already settled by observation the implicit question “Can human being’s mind be categorised in such mechanical terms?”

Falsity or evil “slips in” through the insensitivity, which allows this kind of confusion to take place. This insensitivity is a manifestation of a violent general structure in the operation of the mind. This violent structure is mistaken for harmony, because it is glamorized as beautiful and pleasing. Such a mistake between glamor and beauty is based finally on confusion about the question of the observer and the observed, along with confusion about the meaning of psychological time. This is indeed the ultimate source of evil. Perhaps after you have read Krishnamurti’s writings a bit, we can go further into these questions.

No single aspect of the whole process can really be identified as “the” source of all the trouble. Even the process of collecting “useful facts” at a given level of theory is in itself not wrong. Indeed, under certain conditions, it is really absolutely necessary. But what is wrong is that when sensitive perception could indicate the need for a new level of theory, the mind, drugged by its “glamorized” vision of the “successes” of the old theory, becomes insensitive, and goes on violently imposing the older ideas, suppressing the point in sub-liminal intimations of the need for change. So the real trouble is always violence. But violence is infinitely subtle, and can neither be identified as always being in a certain kind of process, nor as always being absent from such a process. Thus it would be quite easy for the “critical” views of Popper and Feyerabend to become a “mask”, behind which one could go on with his own peculiar form of violence.

In this connection, you say that a certain amount of facting is necessary, i.e., a certain amount of destructive activity is necessary for creation. You rightly call this a “paradoxical” conclusion. But isn’t the paradox due to the violence with which you defend your own views, by attacking those of the empiricist? If you did not wish totally to destroy (i.e., demolish) the position of the empiricist, you wouldn’t feel impelled to say that “collecting useful information” is always destructive. After all, if I ask you how to get to a certain place, what violence am I doing to you, to me, or to anyone else’s ideas? But if I suddenly did this in the middle of a deep discussion that was becoming embarrassing to my position, this would be violence, therefore destructive. The radical empiricist is more or less in the latter position, because whenever deep issues are raised, he insists on collecting more “useful information”, when in fact, no more information is really needed, at this particular juncture. His “ploy” is a form of violence, similar to “How to Win at Games Without Actually Cheating.”⁵

With best regards

              David Bohm

Oct 20, 1967

Dear Jeff,

This is a supplement to yesterday’s letter, answering yours of Oct 9.

Firstly, I want to emphasize the vast extent of the problem of violence. Violence is rooted in all thought, yours and mine, that of the empiricists, Rosenfeld⁶, Johnson, Kosygin, etc., etc. Very probably, it originates in the survival of some of the old animal instincts, built into the structure of the thalamus. Sometimes, one experiences what may be called “unalloyed violence” or “pure hatred”, which literally “fills the mind”, causing the adrenaline to flow, the blood pressure to go up, etc. In such a state, all the higher functions of thought and awareness are overcome and overwhelmed. A person is in a kind of “ecstasy of violence” in which he really does not know what he is doing. Primitive people and young children experience this state frequently. In a moment, they are overcome by murderous rage and do great harm to others. A minute later, it is all over, almost like a summer thunderstorm, and the air is clear again. Violence is, in this case, a monetary phenomenon, that “just happens” without any tendency to continue over long periods of time.

As civilization developed, men began to realize that such violence is dangerous and destructive. It was only natural to think of controlling this violence. After all, man had begun to “exercise control” in some of the more mechanical aspects of his life. So he conceived the notion of “controlling” violence. To do this, he condemned violence as “evil” and set up the opposite, i.e., the ideal of the “good” man, who is not violent, but rather cooperative, kind, loving, and understanding, being thus “wise” instead of “foolish”. Men were in this way enjoined to be “good” and not “evil”. The “evil” was “what is” and the “good” was “what should be”. So men made an effort to be “good”.

The difficulty with this effort is that the mind is infinitely subtle and complex, so that such a mechanical division of “good” and “evil” has little or no meaning. Therefore, the effort to impose it led to confusion. Indeed, what happened is that men experienced violent impulses and then, in this state of violence, they remembered the injunctions to “be good”. So, while moving in the order and structure of violence, (which is always imposing patterns, defending them, and destroying what gets in the way), it set up the imaginary ideal of “goodness” and tried to impose this, to defend it, and to destroy whatever was in the way. Of course the first thing in the way of such “goodness” is the whole set of spontaneous movements of the mind, which can never follow any pattern at all. So the mind began to be “good” by destroying its own creative impulses. Thus, it sets up a violent conflict within itself. In other words, the division of “good” and “evil” had the effect of continuing violence at a new level, where it was more confusing than ever. Indeed, the confusion became such that once started, violence could never end. For because men began to be frustrated in the creative depths of the mind, there arose a new violent demand for “expression”. After long periods of suppression of his violent urges, he felt the need for orgies, binges, fights, “free for alls”, wars, etc., etc., which he conceived to be “outlets” for his urges. These in turn led to new frustrations, which called for new violence at yet higher levels. Thus, unlike animal violence, (and primitive and childish human violence), civilized human adult violence tends to continue indefinitely, and to build up and spread out, to cover vast areas.

One of the factors that “confound the confusion” in this whole issue is the intimate relationship of violence and pleasure. As I have said before, there is real enjoyment or joy, which is not violent at all. It does not impose anything, nor does it defend itself, nor does it destroy what gets in its way. But there is a kind of imitation of enjoyment, which we may call “mechanical pleasure”, and which is always “glamorized” to imitate a really beautiful state of feeling. This kind of pleasure tries to maintain itself by imposing various patterns felt to be pleasant and secure. When these are threatened, it initiates defenses, aimed at destroying what appears to be in the way of its continuity. Therefore, this kind of pleasure is a manifestation of violence.

Now, pleasure and pain are abstractions, like the North and South Poles of a magnet. Every sensation and every state of feeling contains both. Because violence is an intense state of feeling that is full of energy, one can by thinking about it abstract intense pleasure and intense pain. Thus, many people really enjoy a fight, while others find a fight very unpleasant indeed. It all depends on which end of the “pleasure-pain dipole” the mind happens to focus. And of course, to focus on either side is a distortion. Only a mind that can be aware “impartially” of the pleasure and pain in violence can see that both are irrelevant trivialities. Unfortunately, however, our tradition and background of thought lead us to treat pleasure and pain as two separate (or at least separable) “opposites”. So either one focuses on the pleasure, and the brain wants more, or one focuses on the pain, and the brain wants less. Indeed, the man who enjoys fighting has totally suppressed or deadened awareness of the pain in it, while the man who hates fighting has suppressed awareness of the pleasure in it (perhaps because he wants the “higher” pleasure of thinking that he is a “good” man).

The essential point here is that the very movement of suppressing or deadening sensitivity to one side of the “dipole” or the other is itself a new order of violence, a process in which the mind tries to violate its own natural order of operation. Very often, one becomes so deadened to the pain in an urge or craving that one is not even aware of this process of violation. Likewise, one becomes deadened to the pleasure in violence, so that one ceases to realize that violation is going on. Consider for example, the “good” citizen, who is a “right-thinking” person. He sees how the children of the poor are wrongly treated, but he says “This cannot be helped. After all, their parents are shiftless. It is only a fact.” What this man doesn’t see is the deep pleasure in his feeling that his own children are, by comparison, secure and happy, along with the suppression of all the pain that is part of this thought (e.g., one sees that it really isn’t right, that it is even dangerous for his own children if others become corrupt, etc., etc.). What is going on in his own mind is a vast process of violence, aimed at producing a pleasing sense of security of oneself (and of the inevitability of the situation of the others, implying no responsibility to do anything). Such a mind is literally deadened and made dull, so that it cannot be original and creative in any field at all.

From such inward violence, there proceeds outward violence. Thus, when others disturb one’s pleasing sense of security, there arises a defensive movement, aimed at destroying whatever is conceived to be the source of the trouble. In this destructive movement is delusion. For the mind is always deadening itself to any idea that threatens its sense of security and opening itself up to any idea that does the opposite. So in its very thought, the mind is violating its own natural order, the more effectively to violate the order of whatever it is that outwardly seems to be in the way of the continuity of security and pleasure.

Almost all that is done in society proceeds from a violent structure of thought, aimed at imposing patterns, defending them, and destroying opposition. For example, all ambition is evidently violence of this kind. Much of what is called “love” is actually violence. A man who prides himself on being “good” is the most violent of all. For the “evil” movements of violence in himself are attributed to others. Thus, he is unaware of his violence, which can act without any check or hindrance at all. This point is significant even in scientific or philosophical controversies. One sees the “evil” views of the other person, which are “violating” the “right” order of things. So one condemns this “evil” and tries to “combat” it with one’s own “right” and “true” ideas. In doing this, one does not notice that the “evil” in the other person is precisely his violence. That is, he is excited by his own idea, which gives him great pleasure. To defend this pleasure, his brain suppresses awareness of what is wrong with it, and tries to “demolish” the arguments of others, who wish to call the deficiencies of his idea to his attention. This whole behavior is violence, which has its dynamic source of energy in the sense of excitement, glamorized to appear “beautiful”. But as soon as one calls this “evil”, one is doing the same thing. One is excited, and this gives pleasure (as well as pain, of course). To defend the pleasure (and get rid of the pain) the brain suppresses awareness of what is wrong with one’s own arguments, and tries to “demolish” the arguments one’s “evil” opponent. He in turn feels that his opponent is “evil” and has to be “demolished”. But evidently, this whole game is meaningless, and in fact, itself the very essence of evil.

Very much tied up with all this is the question of influencing and being influenced. Of course, a certain kind of influence is natural and inevitable. Thus, when I learn physics, the whole course of my life has been influenced in a great many ways. But there is another kind of influence that is actually a form of violence. I may try to impose my views by persuasion, trickery, pressure, or propaganda. Vice-versa, I may allow myself to be influenced by you. This too is violence. For I am now imposing your views on my mind, thus violating its natural order of operation. I may do this to please you, and thus to gain something from you, or to feel more secure. Perhaps I regard you as an authority, so that you know all about it. If I impose your views on my mind, then I too will “know”. So to follow the authority of another is violence, when this following implies imposition or acceptance of his ideas without real understanding.

Seeing something of this immeasurably vast structure of violence, inward and outward, what are we to do? Evidently, to try to be non-violent is an absurdity. As I explained in the previous letter, this is merely the violent imposition the ideal of non-violence, a meaningless, destructive, confused and delusory process, if ever there was one.

What is called for is not merely superficial change in the forms of violence. Rather, the mind has to be aware, (directly and non-verbally) of its deep roots. Then perhaps, the violent structure-function of the mind will come to an end.

In this connection, an important source of confusion is the set of words: “I am violent”. This implies that violence is a contingent feature of the “self”, or in other words, that the “self” can sometimes be violent, while at other times, it can be peaceful and harmonious. But the real state of affairs is that wherever there is a “self” there is an inherently and inescapably violent structure of thought. For the “self” is always aiming at pleasure of the mechanical sort, which depends on imposing patterns of thought, defending them, and destroying (or deadening awareness of) whatever is in their way. So instead of “I am violent”, one should better say: “I am violence”. That is to say, violence is the very essence of structure-function of the “self”. The ending of violence is the ending of the primary role of the structure-function of the “self”.

To this end, we have to be careful about how we are using words. One can use words descriptively and inferentially. Also, there is the ostensive use of words, which “points to” an example of what they mean. But in addition, there is something entirely different. Words may be used to set up a mirror in the mind.

Now, a mirror doesn’t tell you anything. It is neither descriptive, inferential, nor ostensive, in its function. Rather, with the aid of a mirror, you can see what you couldn’t see before. For example, suppose you ask me to tell you about yourself. You are not satisfied with my descriptions, inferences and ostensive “pointing” to things that have qualities similar to yours. Imagine now that there are no mirrors actually available. So I give you directions as to how to make a mirror. When you have done this, you can see, not only your face, but something more; i.e. the relationship between your outward expressions and the inward moods, urges, wishes, drives, thoughts, etc. You may have thought: “This is a nice feeling.” But now you see that it expresses itself as a silly or ugly appearance of the face. You wonder whether it is really so nice after all. Sensitive awareness discloses that there is a violent suppression or deadening of perception of the uglier aspects of this “nice feeling”. Without this deadening, one sees the reality – i.e., a meaningless process of violating the natural order of operation of the mind.

All our outward actions can thus serve as mirrors. But sometimes, words may indicate how the mind can set up an inward mirror that is even more sensitive and comprehensive in what it reveals. Verbal communications having to do with the structure of thought rather than its content, can have this effect. Thus, what I say about the structure of violence will have a certain meaning to you. It is possible that through these meanings, you will cease to regard the particular content of violence as basic (e.g., I am violent because I was frustrated). Instead, I now see that the entire structure of my violence as it is at this moment is a mirror to the dynamic source of action that is really meant by the word “I”. In other words, all my violent patterns of feeling and thought are mirroring the movements of the deep source of action which is meant by the words: “I – in my innermost essence.” If awareness ceases to be seriously concerned about changing these patterns into “better” ones (because the futility of this is clear), then it can simply pay attention to “what is”, without choice, neutrally, impartially, and factually. In doing this, one is aware of the relationship between these mirrored manifestations of violence and very subtle sub-liminal perceptual cues as to the activity of the central source of energy that is meant by the word “I”. In this way, the mind learns the real source of violence, while it also learns about the real ugliness and destructiveness of the effects of this violence. Spontaneously and naturally, without choice, effort, or act of will, the violent structure of the central source of energy begins to come to an end. This latter becomes harmonious, peaceful, silent and empty, not merely on the surface, but down to very great depths.

In this peaceful harmonious state of mind, there is not stagnation, but rather, action of great intensity and subtlety, far beyond that of ordinary feeling and thought. Let us compare the depths of mind to a central fire or “sun”, so “hot” that most of its “radiation” is X-rays and ultraviolet light, which are invisible. Therefore, it appears to be just “emptiness”. As this energy moves outward, toward the “surface” of the mind, it can “cool down” to become “visible”. Anything that is false or disharmonious on the “surface” works toward the depths, where its falsity is first “felt” and then revealed in more detail, in “thought”, which comes “outward” toward the “surface”. So the ability to be sensitive to beauty and ugliness underlies the ability to perceive truth and falsity.

In all this, it is crucial to understand the irrelevance and triviality of the content of our violent thoughts, feelings and actions. Rather, the whole of this content is now able to reveal the deeper structural process of the mind, which it is mirroring.

One interesting point. If you work in this way, day by day, moment by moment, being sensitive to violence as a mirror to the central source of dynamic action of the mind, you begin to get glimpses of that “pure” animal violence that we once experienced as children. However, we are now very afraid of it. So the mind tries to suppress it at one level, while it is continued at another. For as I have said, violence has a very enjoyable and pleasing side. The mind tries to retain this, thus continuing violence, while it tries to get rid of its dangerous and destructive side, by “controls” of the kind I have described earlier. But as has been indicated, these “controls” are delusory. Actually, they only proliferate violence into new fields.

The very effort to limit and control violence is what keeps it going indefinitely. The primitive man and the young child do not try to control violence, so that they are quickly “finished” with it. But is this not too dangerous?

This question implies a lack of understanding of the real nature of action. Generally, we assume that the essence of action is outward physical action, – i.e., to move, to speak, to write, etc. What we do not realize is that awareness is action. There is a vast and immeasurable action going on in the “central sun” of awareness. This action is creative. In it are formed the deeper aspects of our perceptions, feelings, thoughts, wishes, urges, etc., etc. The “overflow” of this action leads to the superficial activities of the mind, and to our outward physical actions.

So if we should experience the state of “pure” animal violence, this does not mean that this must act outwardly and physically. Rather, the mere “flowering” of this state in awareness is its basic action. Then, if awareness declares that this is appropriate, the action can proceed outward. Otherwise it is “switched off”, and there is no doubt that if one is deeply aware of the full meaning of this “animal” violence, it will not proceed toward outward action.

Of course the child or primitive man could not realize that this is the case. So they can do nothing but allow animal violence to express itself outwardly, after which it “switches itself off”. Civilized adults cannot do this, so that violence never “switches itself off”, but continues indefinitely in ever changing forms. However, when an adult understands how the mind can use its own inward manifestation of violence as a mirror to the central source of mental energy, then it is no longer compelled to express violence outwardly. Rather, the “mirror” is already an adequate expression of this violence, which allows it to “flower” and “switch itself off”. Thus, the mind is freed of the very root of violence.

Oct 23, 1967

Dear Jeff

This is to continue the letters of a few days ago, concerning questions of fact, inference, beauty, glamor, violence, etc.

Now, you have objected to my using terms like fact, observation, inference, etc., because these have become very confused (essentially as a result of their being used as “weapons” in the struggle between the “romantic” metaphysicians and the “hard-headed” empiricists). At the same time, you tacitly adopt the empiricist position by saying that in using the term “fact”, I should at least be able to say what it is not (i.e., how it differs from inference). Thus you imply that certain “fundamental” terms should be capable of clear and unambiguous definition, by being referred to other words, so that these definitions would constitute an “incorrigible” basis for our further discussions.

In response to this, I find the the word “love” springs to my mind. This word has been misused and confused all through the ages. And now, it is worse than ever. Every movie advertised is full of the word “love”, which is equated with sex, violent excitement, mutual dependence and emotional exploitation, cheap sentiment, etc., etc. Likewise in books and in common usage. Following the spirit of your suggestion, I would ask “Why do we use a word that has become so confused? Can we at least say what love is not? Since we can’t, wouldn’t it be better to drop this word all together?”

The difficulty with this approach is that there is a vast domain of life to which people have always referred by using the word “love”. If we stop using it, we simply become unable to refer to this domain. We could introduce another word, but this would only confuse the issue, because deep down, this word is imprinted into the structure of the mind. Besides, any other word would soon become involved in the same confusion, because the latter pervades all of our lives, and is not merely the result of certain words.

Basically, the confusion originates because of the struggle between the “romantic” metaphysical view of love and the “hard headed” empiricist view. The “romantic” is able to delude himself easily by accepting any glamorous or exciting notion that comes into his head, saying “This is such a beautiful feeling! Surely it must be real love!” Then along comes the “hard-headed” empiricist, who says “Away with such silly sentimentality! Let us get clear and well defined ‘incorrigible’ facts! Don’t mistake superficial motions for a ‘solid’ and ‘well-founded’ mutually advantageous relationship, that is the only ‘reliable’ basis for real love”. But of course, the “hard headed” realist is just as deluded as the “sentimental romantic”. Love is actually too subtle and immeasurable a thing to be defined “solidly” and “incorrigibly”. It can be perceived only with tremendous sensitivity. And because both the “romantic” and the “realist” are violent, neither can know love. Deep down, neither of them is very different. What is called for, therefore, is to end the meaningless struggle between “romantic” and “realist” that goes on within every human being. Neither point of view makes sense. We have actually to transcend the whole framework of violence, which imposes a point of view, and defends it, by attacking opposing points of view. Rather, the mind has to work from a basis of deep harmony and real peace (not the interlude between struggles and wars).

Now the words, “observation”, “fact”, “inference” refer also to a vast domain of real life. It is also no use to change them. Rather, as with the word “love”, we have to learn to use these words properly.

How can we learn this? Firstly, we have to be free of the narrow minded utilitarian approach, which says that our whole activity is being directed towards some already defined end (such as clarity, freedom from confusion, etc.). Actually, the “romantic” shares this approach with the “realist”. For tacitly, his end is glamorous excitement and pleasure. The “realist” also wants pleasure , but he emphasizes its “solid” security. But now, we perceive that any attempt to shape the mind according to preconceived notions is based on delusion, and is therefore doomed to inevitable frustration and disappointment. Rather, all that makes sense is to discover how the mind actually works. In the presence of the resulting awareness, the mind will spontaneously and naturally “find” its “way” to a new harmony, of a kind that is ever fresh and different. The mind is too infinitely complex and subtle to be described, predicted, or even “pointed to” in an “ostensive” fashion. But it is possible to use words to help the mind set up mirrors, in which it can “see” itself.

It is in this spirit that I am using words like “fact”, “observation”, “inference”, etc. In a previous letter, I referred to the “pleasure-pain dipole”, meaning by this that every sensation contains pleasure and pain, in the way that every magnet has a North and a South pole. Likewise, I want to refer to the “observation-inference dipole”. By this, I mean that every statement and every perceptual abstraction, from which statements are drawn, must have an observational aspect and an inferential aspect. By analyzing the statement, we can make what was observational into something inferential, but then, we introduce a new observational aspect at another level (as by breaking a magnet, we introduce new North and South poles).

So I am suggesting that we regard terms like “fact”, “observation”, “inference”, etc., as potential distinctions, which we intend to articulate as we go along. We do not and cannot begin by defining these distinctions “incorrigibly”, with certain “utilitarian” ends in mind. Rather, we need a kind of artistic spirit here. When the artist draws a line or makes a colored spot on the canvas, he does not begin by defining “incorrigibly” what this action “means” or what it “does not mean”. Rather its meaning is almost entirely potential. As he goes along, the meaning of the “whole picture” gradually takes shape, in a creative way. He himself didn’t know fully what he “meant” until he finished. And even then, he sees new meanings all the time in what he did. If we regard opposing pairs of verbal terms as “dipoles”, this gives us “room” to engage in such an “artistic” approach. For we realize that the meanings of the terms will emerge creatively as the “whole picture” begins to “take shape”.

So now, we are approaching our subject in an artistic spirit rather than in a utilitarian spirit (which latter always moves towards predetermined ends, rather than allowing the ends to form together with the means, as the “whole picture” slowly takes shape). It is a bit like a child, who learns in order to play, while he plays in order to learn. So let us see if we cannot be a little less grim and utilitarian in spirit, not always trying to “do good” and “be in the right”, and not always trying to “overcome the evil views of other people”.

Now, as I suggested in earlier letters, because “observation-inference” is a dipole, it is easy to confuse one with the other. Thus, when you break a magnet, North and South Poles appear at what is, in the beginning, the same place. You have to be sensitive to understand that in the beginning the distinction of North and South is potential. Later, it will become manifest. But if you don’t watch carefully, you may mix them up in the earlier stages.

What we often fail to notice is that an explicit question leading to an observation is based on a vast structure of tacit inferences. Either we may not know this, so that we regard the form of our question as inevitable. Or else, we may imagine that the deeper inferences are adequately confirmed, when in reality, they are not. Or else, we may misinterpret actual observational data, misconstruing it as a confirmation of these deeper inferences.

All of these mistakes can, at times, be “simple” and “honest”. But usually, they are a result of insensitivity. This insensitivity is seldom due to fatigue or brain damage or some such “simple” cause. Almost always, it is the result of inward violence, which “defends” certain ideas, by “attacking” those that get in the way. The most common form of attack is by “deadening” awareness of them, “blanking them out”, etc.

Of course, to end this kind of “evil”, we must end the deep rooted violence in the mind, so that the latter is harmonious and peaceful. But this requires that we understand every link in the “chain” of violence, from its deep roots to its various outward manifestations. One of the key links is just the confusion between the observational (or factual) role of a term and its inferential role.

Here, I would like to call attention to the fact that violence always operates behind an apparently natural “mask” or “cover-up”. For if it were seen to be the meaningless and empty movement that it actually is, the mind would immediately recoil from it altogether, as from poison. So there is a natural “cover-up”, and an emotional “cover-up”. Rationally, one “covers-up” by shifting the terms of our reasoning, so that what we are doing seems both logical and inevitable (E.g., we were only defending ourselves peacefully when the enemy violently attacked us). The emotional “cover-up” works by “glamorizing” our violent sensations, making them appear to be beautiful and enjoyable. Both “cover-ups” depend on “deadening”, “dulling”, “numbing”, “anaesthetising” processes, going on in the mind. These enable one to be insensitive enough to confuse observational fact and inference, so as to “naturalize” our violence, while the ugliness and painfulness of this process is also blocked from perception, leaving only the “glamorized” imitation of beauty and enjoyment.

As I indicated earlier, the only way out of this is the deep and total ending of the structure of violence. This requires a “mirror” which reveals the “outward” manifestations of violence, enabling sensitivity to find the deep and subtle roots, that can never be described in words.

Here, I am reminded of a new invention to help amputees, with electrically powered artificial limbs. These are directed by nervous impulses in the stump of the limb, picked up electrically, amplified, and fed into the artificial limb in a suitable way. Such a person can learn to use these limbs almost as well as his own. What he does is to “try” to move his arm, and watch what the artificial limb actually does. By being sensitive both to the pattern of inward action and outward manifestation, he can learn how these are related. Nobody can describe verbally the inward pattern of nervous impulses that moves the arm in a certain way. He has to be aware of these, at a deep non-verbal level. The key point is then this: The outward movements of the arm are like a “mirror” of the inward pattern of dynamic nervous energy, that is the “source” of one’s physical action.

Similarly, a certain structure of thought can lead the mind to regard the total manifest pattern of violence (in thought, feeling and action) as the outward “mirror” of the inward pattern of dynamic nervous energy, that is the deep source of all one’s action. In this way, one discovers where violence really originates. And since one doesn’t really want it, the mind just “turns off the switch” and brings it to an end. When the mind is thus peaceful and harmonious, it is not stagnant. Rather, there is a vast inward creative energy that manifests itself outwardly. At present, this energy is dissipated in friction and turbulence, which are due to violence. But when violence ends, there is tremendous creative energy.

Perhaps when you read Krishnamurti’s writings on the subject, you will get another view of it.

Best regards

               David Bohm

P.S. You can probably see that the utilitarian approach is basically violent. A certain aim is imposed, then defended, by attacking or destroying all opposition to this aim. Consider, for example, a modern city. Each person, each corporation, each group formulates its own utilitarian aim and imposes this aim, defends it, and attacks those who get in the way. The result is conflict all around. That is why I feel that a modern city is mostly ugly. Some of its functions do have a certain beauty. But basically, they all impose themselves violently on the individual who lives in the city. They leave little room for sensitive adaptation to the needs of the individual, even when they are not positively ugly.

This utilitarian point of view shows up in science, as the imposition of a certain view and its defense, by attacking other views. This in fact seems to be what von Neumann did. After working out what appeared to be a beautiful and comprehensive theory, von Neumann probably felt the need to make it “secure”, i.e., to defend it by attacking views that would seem to threaten it (i.e. hidden variables). Being in a “violent” frame of mind, he was insensitive enough confuse the notions of false – falsifiable – falsified, as I explained in a previous letter. Thus, he could make quantum theory appear to be “invulnerable”. People are always trying to make their achievements “secure” in similar ways, in many fields.

What we need is the artistic approach to science, rather than the utilitarian approach (i.e., the approach which sets up a certain aim, and defends the result that is thus achieved, by attacking or otherwise discouraging contrary approaches). But here, we need the genuinely artistic attitude. Most of modern art is in fact violent. Very often, indeed, the artist imposes his own emotional conflicts on the medium, and defends this imposition tacitly in various ways. Thus, he does art violently. Indeed some modern “artists” say openly that their creative work is to engage in destruction of various objects, such as furniture, thus “discharging” their violent impulses. (Even in a therapy, this is, of course false. For such “outlets” usually only serve to encourage a person in the habit of violence.)

Then there are those who impose rigid mathematical patterns on their work, and “defend” these by deadening their minds to the mechanicalness of what they are doing. Really, any artist who works from violence must be deadening his mind to the ugliness of what he is doing.

A really creative artist does not impose his arbitrary notions on the medium. On the other hand, he does not just allow his impulses to flow at random. For this too is violence. Creation is possible only when there is great sensitivity to the relationship between inward intentions and outward manifestations.

A fellow like Picasso seems to work by imposing his arbitrary ideas and feelings on the medium. He can do this because he tremendously overvalues all the arbitrary little things that come into his mind, mistaking them for something beautiful and creative. He talks of creation, by “destroying” the creations of nature. Is this merely an “outlet” for his deeply violent psyche?

Real creation is at the same time love for one’s medium. Nobody with love will try to destroy what he loves. He destroys only what gets in the way of what he wants to impose and defend.

Oct 25, 1967

Dear Jeff

This is still in answer to your letter on fact, inference, etc.

In an earlier letter, I referred to a hierarchical “vertical” structure of fact and inference, as distinguished from its “horizontal” structure of time development. It is this “vertical” structure that allows fact and inference to be confused, when the mind is an a state of violence and consequent insensitivity. Thus, the inferences underlying a given question (E.g., “Is the boy a thief or not?”) imply a tacit structure to the whole situation. Because of insensitivity and deadness (dullness, etc.), we may not even notice that the question contains such inferences. If one does notice it, one may imagine that these inferences were already confirmed in vast detail in the past. Here, recall that memory is tricky. It is always reshaping itself to emphasize what is pleasing and satisfactory, while it “blanks out” and suppresses what is not. It is not generally realized that this process is a basic aspect of violence, in which the past is being glamorized and made to appear beautiful. Such glamorization destroys one’s ability to see what is true and false, because one can no longer sense disharmony, ugliness, etc. In particular, it can cause one to fail to see that there is little or no real confirmation of one’s tacit inferences, and that there is often actually a lot of evidence against them, which is being “blanked out”, suppressed, or otherwise overlooked. And even if one does try to go over this evidence, one’s evaluations are distorted. For the brain tends to accept those evaluations that give pleasure, and to reject those that do not. So when one is in a violent state of mind (which is almost always) one can easily confuse what is actually a very dubious inference with a fact that has been thoroughly confirmed.

Of course, the confusion of the inference I¹_a with a fact depends on the dullness and deadness of the brain as it deals with lower order inferences (I¹_b, I¹_c, - -) and their questions (Q¹_b, Q¹_c - - -). This in turn is the result of glamorization of certain actually ugly sensations, so that these are confused with enjoyment, energy, liveliness and beauty, with the result that false answers can be accepted for these lower level questions. This process goes down deeper and deeper until it seems to reach our blood and our very pores. In other words, violence creates a similar state of confusion at all levels. And therefore the only way out is for the totality of violence to end, so that the mind is peaceful at very deep levels.

In your letter, you give an example of how the elaboration of a set of theories on a given level can be a “trap”, because none of them is really right. Each contradiction leads to a minor modification of the theory, which in turn is contradicted in the next step, etc., etc. What is called for is a creative change of order or level of the whole structure of the theory. And you identify as “evil” the tendency to work in the old framework.

Of course, this sort of thing often does happen, and is indeed happening today in physics. Nevertheless your blanket condemnation of working in a given framework goes too far, and is therefore itself a kind of violence. Indeed, very often a new set of ideas has a great creative potential. It is reasonable to work creatively, developing what is in these ideas, until the potential begins to show signs of exhaustion. It is then necessary to begin to experiment with new frameworks. Unfortunately, because of the violent imposition of demands for “success”, “results”, “precision”, etc., people keep on in the old framework long after it is exhausted. This is then the real source of “evil” – i.e., insensitivity to the signs that the framework is approaching exhaustion, as far as creative potential is concerned.

A similar situation arises in art. Mimesis had an enormous creative potential, over the centuries, which began to be exhausted in the 19th century. Most artists went on with it, anyway. A few experimented with something new, to change the framework, and open up new possibilities for creative work. These were first violently resisted. Then they were equally violently “accepted” by the vast majority of artists, who exploited these new developments as a means of “expression” of what was in the contents of the Ego. What these artists did was to “mix” the old security of mimesis with the new forms of art, thus leading to confusion.

Similarly, in science, a few scientists like Einstein opened up new creative possibilities, after those of Newtonian mechanics had been exhausted. These ideas were first violently resisted, then “accepted”. But generally, those who accepted such new ideas “mixed” them with the old, and thus tried to establish a new “secure” framework in which they could satisfy the Ego with “successes” and “results”. They did not see that (as in art) what is needed is now a continual creative development. This development must start with mastery of previous developments. It then goes on to see how these can be brought creatively to new levels, until they are exhausted. To leave a given framework before it is beginning to be exhausted is a false step. For then, you leave it for arbitrary reasons. Hence it is a violent imposition of your Ego on the physical situation. In seeing how the older structure is becoming exhausted, you get the “clues” for the natural and necessary changes toward a new order.

So creativity includes creative growth within a given framework, along with radical mutation of the framework from time to time. Each growth implies mutation in more limited aspects of the total structure. So it is all a question of the scope, extent, degree, and order of mutation, which must be appropriate at each stage of the process.

The conservative and the radical are both violent, hence uncreative. The conservative violently imposes continuation of the old order. The radical violently imposes a struggle against the old order. So deep down, both are similar in what counts most, i.e. the quality of violence. The truly creative man does not struggle either to continue the old order or to change it. Rather, from a state of peace and harmony at the deeper levels of the mind, he understands the old order, helps to bring it to full creative flower, and to die naturally when its time for ending has come. Meanwhile, he is always sowing the seeds of the new orders, and allowing them to come to maturity, in their own natural times. There is no violence in this process anywhere.

One more point – the modern city. As I explained in a previous letter, it is basically violent. It does not sensitively adapt to people’s needs, but forces people to adapt to its structure. Those who are poor evidently suffer, as they are forced to live in decaying, noisy slums where violence is everywhere. But even those who have a bit of money have to conform to the pattern of the city. Even those of its functions that may seem “beautifully” ordered on a superficial level are actually generally violent impositions on the pattern of life of the individual. Consider the road system, for example. Once you are caught in it, you are like a straw in a roaring torrent. Watching the moving lights of the cars from high up may give an impression of “beauty”. But to drive on the “freeways” day after day is to be violently assaulted in one’s nervous system. So basically the modern city is violent, hence ugly.

Some of the worst ugliness of a city is in the division between the “ugly” and “beautiful” parts. Thus, the middle class lives in quiet, “beautiful”, “peaceful” sections, while the poorer people and the Negroes live in noisy, decaying, ugly and “violent” sections. The middle class individual says: “My family is growing up in peaceful, harmonious, comfortable surroundings”. As he says this, his mind “blanks out” the nearby slums, or says that only “shiftless” people are living in them. Thus, his mind becomes dull and insensitive in vast areas of its functioning. He has done tremendous inward violence to his mind, and from this, outward violence tends to follow, both his own and that of his children. The latter, feeling the “dullness” of the “peaceful”, “harmonious” surroundings begin to take drugs, to become hippies, and to engage in other forms of violence. So the man was deluding himself when he thought of the “beautiful” surroundings in which his children were growing up.

Best regards

David Bohm

Nov 6, 1967

Dear Jeff

This is a brief letter, to register receipt of your two letters, and to invite you to the conference⁷ (symposium) described in the attached sheet? Can you come?

I am afraid I can’t come to America this Spring, as we are going to Israel again for a month. Could I have a rain-check for a year, on this visit?

I won’t comment on your very interesting letters in detail till later. I would only want to say something about the man who (like the hippies) finds pure beauty and joy in what you call the “micro”- aspects of the “music of life”. To me, this seems to be a disguised form of violence. After all, violence is refusal to look at the whole of the fact, insofar as this is accessible (or establishable) in the mind of man. Glamor becomes a substitute for beauty whenever the mind “anaesthetizes” or otherwise escapes awareness of $\underline{\underline{\text {any}}}$ aspect of reality. It is not really the object of awareness to produce a state of joy. Rather, this is properly a by-product of complete, total and harmonious movement. Real joy contains a perception of disharmony and sorrow, insofar as these exist. If these are not perceived, they will work anyway, to produce delusion, chaos, destruction, violence.

Let me put it another way. Chemical studies show that LSD cuts down the oxygen content of the blood (alcohol may do this also, to a lesser extent). This weakens the clarity of perception of contradiction and conflict. At present, we are conditioned to inhibition of certain perceptions, because these could lead to awareness of painful conflicts. With LSD (and alcohol) these inhibitions are themselves inhibited. So certain capacities, previously blocked, are now released. Thus, the hippie can feel: “It’s here, It’s here!” However true awareness requires all our faculties, including those usually called “critical”. These are knocked out by LSD. Hence, such a man is not really aware.

The key point is that to be “cured” of violence, man must be aware of the factors that are deadening, dulling him, making him insensitive. LSD can knock these out altogether. But if the inhibitions are “asleep”, then they cannot be perceived. So no true liberation can be produced by LSD. On the contrary, man becomes a slave of the drug, because he needs it to produce his desired experiences. This drug is a violent interference with the natural order of operation of the mind; – i.e., it violates this order by putting the critical intellectual faculties to sleep. These latter are just as integral to the mind as are emotions and beautiful perceptions.

In any case, any imposed pattern of perfection is violence. If I find the “macro” aspects of life confusing, and refuse to look at them, I am imposing the pattern, defending it, and attacking those functions of the brain that inform me about the macro aspects (e.g., by becoming “dead” to these functions). So the man you describe in your letter was actually practicing an extreme form of violence on his own mind.

Real harmony is incompatible with any fixed pattern, defined as “perfection”. Thus, if I say “Joy is only in the micro-aspects of perception”, this fixes and limits the patterns of the mind. We need to break each order of symmetry or harmony, to make possible the next level of richness of harmony. This break produces danger i.e., that it opens the door to violence, delusion, chaos, etc. But if, through fear or wish to be always secure in our joy, we refuse to open-this-door, we condemn ourselves to just the kind of violence, delusion, and chaos, that we want to escape. For real creativity calls for eternally breaking the old order and ordering these breaks, to form the next order of harmony. And the moment we cease to be creative, we do this through violence, so that we are destructive. Whoever is not creative is violent. Mediocrity is violence. Covering up mediocrity by taking a drug is a still deeper form of violence.

What is called for is the deep perception of the root of violence. Then you won’t need a drug. Nor will you need an “insane” concentration on the “micro” aspects of life. Real joy will then occur spontaneously, naturally, creatively. It will include all violence, and yet be utterly beyond this violence, in its deepest being. In a state of real joy, a man can feel all the sorrow of the whole world, and yet realize that there is a deeper level still, where this sorrow has no place, no meaning. No “ugly” fact is left out of this perception. But one sees that nevertheless, this “ugliness” is only a superficial “wave” on top of the “music of life”. Yet, it is important to perceive it. For otherwise, we do not and cannot function in a natural, coherent way. Rational, coherent orderly function is part of real joy. In addition, it is necessary for maintaining the kind of state of health, (physical and mental) in which real joy can take place. When social function is irrational and violent, the state of real joy can perceive this, without losing the harmony of the depths. From this deeper harmony can arise the action needed to correct the superficial disharmony, leading to a state of creation.

I am happy to hear of your expected child, and hope that all goes well with your wife (and the child).

Best regards

        David Bohm

Nov 7, 1967

Dear Jeff

This is a brief supplement to yesterday’s letter.

First, one of the previous series of letters on violence contained a request to send this letter on to Biederman. If you have not already done so, will you please find this letter, and send it on to Biederman.

About drugs, such as LSD. I have been interested in them since about 1958. I have talked with psychiatrists who use them (in treatments of patients) and I have read several books written by people who experimented with LSD. At first I was attracted by the idea of “liberating creativity” with LSD. But the more I looked into it, the less enthusiastic I became.

A typical account of LSD experiences was by a woman, being treated by a psychiatrist. (The theory was that it would help reveal the unconscious.) Before her first treatment, the woman happened to be reading Life magazine, all about evolution. So after taking the drug, she literally experienced the process of evolution. In boring detail, she described, step-by-step, how she was an amoeba, how it felt to be a worm, a dinosaur, etc., etc. Evidently this was all an elaboration of the article she had been reading. In subsequent treatments, one also saw, from her reports, that she was elaborating various kinds of conditioning, mostly fortuitous in nature. Other authors describe experiences which constitute similar elaborations. E.g., the psychiatrist told a woman to try to remember the sexual behaviour of her mother and father, and not surprisingly, she experienced herself as a baby watching the behaviour, (though both she and the psychiatrist agreed that this was probably fabricated by the mind).

It began to be clear to me that LSD inhibits the critical faculties of the mind. In this way, it allows every form of fantasy and delusion to be projected, as if it were being directly perceived. This is a spurious kind of “creativity”. For its basic source is evidently in the “memory tapes” of conditioning, which act as “programmes” that determine the images and feelings that are perceived in consciousness. Far from being creativity, this is a form of mediocrity, therefore, of violence. Real creativity comes from the unknown depths, not from the computer “programmes” stored up in memory.

Indeed, many takers of LSD report violent and disturbing hallucinations, which are fantastically terrifying. It is all a matter of fortuitous events, determining whether the pleasant or painful aspects of one’s conditioning will be “replayed” at a certain moment.

Even those who seem to see the “music of life” at deeper “micro” levels are suspect. After all, they have heard about this sort of thing from other “hippies” or from reading about “mystical” states of ecstasy, or from perusing books on Zen Buddhism, or in other ways. Perhaps they may have accidentally “seen” the “micro-music” once, and this was “tape-recorded” as a “programme”. In one way or another, they come to experience “the tape-recording” of the “micro-music” as if it were a real perception. After all, if a woman can literally “experience” the process of evolution or the sexual behaviour of her mother and father, it is equally possible to “experience” a good imitation of “mystical ecstasy”. Anything is possible, once the critical faculties of the intellect are inhibited.

As I indicated in yesterday’s letter, what is called for is that all faculties of the mind be at their highest level of activity, needed for true awareness. The critical faculties of the mind may well now be inhibiting the free movement of images, feelings, and “models” needed for creation. Nevertheless, the right answer to this is not to inhibit the critical faculties by a violent, therefore mechanical, interference with the chemistry of the nervous system, by means of a drug. To inhibit an inhibition is generally a wrong approach. What is needed is by clear and sensitive perception, to “feel out” the real root of the inhibition and bring it to an end. Then the mind will be harmonious and creative, from its depths. Even if the LSD taken really sees the “micro-music” (which is, as I have pointed out, very doubtful) this is a wrong thing to do. For if one inhibits the inhibiting factor, one will never perceive it. Rather, one will be compelled to end the inhibition mechanically by introducing further chemical inhibitions, that destroy the clarity of the intellect.

In real creativity, there is no attempt either to hold a fixed pattern or to change it. Rather, the mind is participating in the total process of reality, including external nature and the processes of the human being. It is learning what these processes are – what is their natural order. From this learning, is acting creatively to form new higher orders, that are in harmony with the whole natural order. This creation has neither beginning nor end. To live is to be creative. Each level of harmony begins to break spontaneously. The sensitivity of the mind sees here the possibility of new orders of creation. On the other hand, when the mind is dull, insensitive, dead, it either tries to hold the old order or to impose arbitrary new orders (which are really “unconscious” features of the old order, stored in the “programmes” of the memory “tapes”). All this is, of course, violence.

So to fail to be creative is, in the same step, to be violent. Mankind has been almost entirely mediocre and violent for thousands of years. Only a few have been somewhat creative. But their creative work was applied in a utilitarian way toward mediocre and violent ends. When someone is creative, what is called for is that other people respond equally creatively, first by learning from him, then by extending what he did to new fields, and then by going on to new levels of creation of their own, (extending to yet higher levels and newer orders). Whoever does not do this is doing violence to the creative work of others, and the inevitable result of such a violent state of mind is destruction.

Nevertheless, one must be clearly aware of the trap of violently reacting against the violence of others. The fact is that whoever is violent is thereby dull, insensitive, deadened, numbed, and anesthetized. Thereby, he does not know what he is doing, and cannot do other than he does. One must understand the absolute inevitability of the behavior of a man who is trapped in his own violence. When he does not understand, he inevitably deludes himself. And when he does understand, he equally inevitably stops deluding himself. Whether he understands or not depends on largely fortuitous characteristics of his conditioning, and how it responds to what is said (whether true or false). It may happen that A, who is deluded and deadened by violence, hears something from B that “penetrates” and ends the state of violence. But C, hearing the same thing, may violently resist what B says, because of his peculiar conditioning. Yet D (or even B) may later say something else that happens to “penetrate” C’s conditioning. So the question of breaking through violence is one requiring exploration in a provisional and tentative way. To lay down hard and fast rules on the subject is itself a form of violence.

To put it in other terms, choice is violence, when applied to the order of the depths of the mind. Choice makes sense in superficial and mechanical questions, where the evaluation of the factors that determine the choice has a meaningful basis. But the mind is too vast and immeasurable to be thus evaluated. Therefore, every such choice (e.g., to be a “better” person) is an attempt to impose an order violently on the mind. It is doomed to fail and to add to confusion, because it has no basis in the real order of movement of the mind. Rather, what is called for is a sensitive and uncommitted “choiceless” awareness, that sees the factual order of the mind and in the same step takes the right action, in an order that is far beyond the “micro-level” of thought.

We may sum up by extending the discussion of fact and inference a bit. With Korzybski⁸, let us consider the four broad levels of operation of the mind:

A. Event level

B. Object level

C. Descriptive level

D. Inferential level

Actually, these are interwoven and interpenetrate. Yet, in some rough sense, we may order them as above for analytic purposes.

The event level is what you have called the “micro level” of awareness. It contains the “music”, the vibrations of sensation, the depths of beauty and love, far beyond anything that can be put in words.

On top of the event level is a sort of “wave” of variation of quality and intensity, from which is abstracted the object level. The “wave” gives information about relatively stable forms, orders, structures, etc. It also refers to their separation, difference, similarity, relationship, etc., etc.

In the descriptive level, what we see of objects (and to a slight extent, of events) is abstracted further, into the structure of language and associated thoughts (which latter form a yet slower and more irregular “wave” at a yet more “macro” level).

In the inferential level, a yet higher level of distinction occurs, in which the orders and relationships, structures, of the described objects is abstracted and generalized.

Of course, the process moves both ways. The inferential level helps shape and structure the descriptive level. In other words, what is an inference in one context becomes a descriptive term in another context (i.e., we have a “description-inference dipole”). The descriptive structure influences our perception of objects. The structure of objects influences our perceptions at the event level.

Moreover there are higher order inferences about inferences. These constitute abstract thought.

Now, delusion is a wrong order of abstraction, resulting from violence, which dulls and deadens the mind, so that it lacks sensitivity for seeing the wrongness of this order.

The simplest delusion is to mistake the inferential level for the descriptive level (E.g., “this boy is a thief” is taken as a description of fact, observed directly at the object level).

The next order of delusion is to mistake the descriptive level for the object level. When this takes place with regard to external objects, it is called a “hallucination”.

The mechanism of hallucination is probably based on the fact that all perception contains “models” of the world. When the mind is dulled and deadened by drugs or by the intensity of violence that leads to insanity, these “models” can no longer be distinguished from the direct perceptions of objects that are, in normal perception, the observational, factual basis for eliciting each particular model and correcting or altering it when it is wrong.

But of course the “self” is an object that depends on the projection of descriptive and inferential levels of thought into the “models” of perception. Everyone has a “model” of his “self” and the “selves” of other people. But it is very hard to see that there is no real object, corresponding to this “model”. Only a mind deeply free of violence can see this. On the other hand, this delusion helps sustain the violence that deadens the mind, and thus further sustains the delusion. It is a sort of vicious circle.

The deepest order of delusion is to project thought onto the event level, without knowing that this is happening. In this way, the brain can create glamorized substitutes for beauty and harmony, while it substitutes violence for true passion, and self-indulgent pleasure for love. When a man is under the influence of drugs or insane, he can go further, and mistake the “noise” of his conditioning for the “micro-music” that is the essence of real joy.

One more point. You are right to emphasize the value of Piaget’s notions of assimilation and accommodation⁹ . Yet, you must also notice that what we assimilate is an abstraction from the whole of reality. We do not assimilate all of it, in its full concrete detail. So “accommodation in order to assimilate” creates an abstraction. And from this abstraction, we act, either creatively or violently and destructively. When our abstractions are in a right order, then from this assimilation flows a creative action. Otherwise there is a destructive action.

What Piaget fails to mention is creation (and what gets in the way of it, which is violence and destruction). This is a serious omission, as it leaves out the whole point of the assimilation process, its whole reason for being, as well as its crowning point and culmination.

Best regards

     David Bohm

P.S. Piaget implies that assimilation is basically for a utilitarian aim, either of the individual or of society. To put utilitarian values at the foundation of perception is a form of violence. Piaget shares this kind of violence with most scientists in the psychological, biological, and physical spheres of work. Either they imply that certain recognized “use values” are the basic determinant factors or they imply that “survival” is the basic value, or they cynically say that the basic value is to keep ahead in the “rat race” of society.

Of course, survival is, in a certain way, a necessary by-product of the basic order of life. But in the long run, this makes sense only in a context of overall harmony and creativity.

Nov 15, 1967

Dear Jeff

Received your letter of Nov 12 on my review of your article on DLP etc.¹⁰ As I am now rather busy, I can make only a few preliminary comments in this letter. More will follow later.

I will leave most of the minor technical details to your own discretion. If there is something I didn’t understand, this may perhaps indicate to you the need for a slight rephrasing, or for adding a sentence or a paragraph here and there. For example, (P.9 near top) a brief statement, telling that this is a consequence of quantum mechanical laws, would suffice. P.10 near top requires a brief summary of some of the conclusions of our article on this point. Don’t require the mainly philosophical reader to pore through technical details to get hold of your points essential to the philosophy. Same about P.10, middle.

About Bohr, Schumacher is claiming that he is seriously misunderstood. I am now discussing these issues with him, and will let you know what I manage to understand from these discussions, if and when I do understand something. Schumacher agrees that Bohr was wrong to assume that the only unambiguous language is that of classical physics (suitably refined). But he claims that Bohr’s linguistic thesis is not all that different from what I am trying to say about structural process. This I haven’t seen, as yet.

All this emphasizes the need for more care in discussing what Bohr says. Is it really possible to extract a few strands from Bohr without serious distortion? I would very much object to it if a “Bohrian” were to do this with structural process. Doesn’t one need to understand the essence of Bohr’s position, in order intelligently to compare it with a structural-process position? Is empiricism the essence of Bohr’s position? Schumacher claims, very seriously, that in many ways, Bohr is doing what Feyerabend advocates in his article in Beyond the Edge of Certainty.¹¹

Does Bohr accept “the myth of the given”? Schumacher denies that he does. Schumacher claims that Bohr also dismisses epistemology as well as ontology. To be precise, he wants to assert the lack of a sharp division between the two.

All agree however that Bohr is wrong to restrict language to a classical form (however it is refined or developed). It is not certain that he believes every observation has a certain “objective control” (which would be “incorrigible”, the “factual stuff” out of which knowledge is built, etc.).

You ask why Bohr could not describe the sequence of spin measurements? The answer is that to Bohr this description would have no phenomenal significance. According to the laws of qu. mchs, the sequence would be random (within q.m. probabilities). Therefore, it could tell you no more about the individual electrons than the sequence, heads, tails, etc., tells you about the individual coin.

Because we propose a theory in which this sequence is not random, but related to hidden variables, the sequence could, in our theory, have phenomenal significance. This is characteristic of the role of mathematical concepts in indicating what might be “observable”. Thus, before the invention of the calculus, nobody would have thought of acceleration of a body as an “observable phenomenon”. Likewise, given a theory of hidden variables, an ordered sequence of spin measurements becomes an “observable phenomenon” revealing an aspect of the state of the hidden variables (as acceleration reveals an aspect of the state of motion of a particle).

It is not merely the “radical empiricist” philosophy that is involved. There is more; i.e., the mistaking of an inference for a fact. The inference is that the probabilistic laws associated with qu. mchal algorithms are universally valid.

If the article is to be stimulating, it must be clear, at least, a lot more clear than it is now. The reader should not require an “immense effort” to see what the point at issue is. (However much effort he will need to “make up his mind”.)

Your clarification about the meaning of the top paragraph of P.3 seems O.K. to me.

It may be useful for you to explain all this in a letter to me, rather than in conversation. It will perhaps help you to write a better article.

Regards,

           David Bohm

P.S. I feel that we should begin right now to treat this whole issue without violence of any kind. Then perhaps we can help to clear it up. We don’t want, if we can avoid it, to stimulate proponents of Bohr to make a violent refutation of your article. Rather, it is best to cause them to think deeply about their whole position. Do you agree with this? It is not just the “neutral” reader that one has to have in mind. Can one help clear up this whole confused muddle of violence in basic physics? It might be a (very modest) beginning of a contribution to peace in the world.

NOV 24th 1967

[Date added – CT.]

Dear Jeff

I hope that by now your child is born and that your wife is well.

I shall try to give a more detailed answer to your letters.

Firstly, I am in sympathy with the general spirit and aim of your article.¹² But I believe that in its present form, it will not achieve its aim, because many points in it are not clear. In addition, its format is very inappropriate. I would suggest minimizing the math. formalism in the body of the article, leaving this for the Appendix. I picture the reader as a philosophically oriented person, interested in qu. mchs., and what it means. It seems to me that the sudden interjection of formulae into verbal text has a jarring effect on the brain, not conducive to understanding. If you must bring in formulae, explain them verbally, (especially the meaning of the symbols) in some detail. Otherwise, you tend to lose the reader very easily.

Now, about von Neumann, I am in general agreement with what you say. But I would emphasize the following:

(1) V. Neumann’s axioms assume, in effect, that observable averages are linear functions of the operator matrix elements, O_mn.

(2) He proves that this assumption is incompatible with dispersion free ensembles – hence with hidden variables of any kind.

(3) In our paper, we give up this assumption, by assuming a non-linear connection between observable averages and the matrix elements. Thus, in effect, we make use of one of v. Neumann’s results, to indicate the sort of change of theory that is needed, to bring in hidden variables.

(4) To recover v. Neumann’s theory in a suitable limit, we introduce a randomizing process.

But more generally, we are not obliged to recover v. Neumann’s theory, if we don’t want to. We could instead make a theory that related to perceivable, observable fact at a much lower order of abstraction. (E.g., it would have in it abstractions corresponding to interference fringes and energy levels, but might not get these out of the same formalism as that of current qu. mchs.)

In a way, J. and P. try, as you say, to show that hidden variables are excluded, as they would imply that qu. mchs. is already objectively falsified. Of course, they are confused about this point. You would be doing the whole subject a good service if you could get to the bottom of this confusion. Assuming that they are not “raving mad lunatics”, how could they come to believe that the present facts already exclude hidden variables? If you can discover this, you might also make it more clear just what, if anything, they are doing wrong.

        I’ll continue further later.

Best regards

        David Bohm

[The following single page of type-written text has been inserted here – CT.]

DEC 10th 1967

[Date added – CT.]

THE NEGATIVE APPROACH TO THE MEANING OF LANGUAGE

D. BOHM

Words and their meanings are never more than abstractions, which cannot substitute for that to which they refer (e.g. using the word for “dinner” and thinking about what it means to us cannot provide the kind of nourishment that comes from actually eating a meal). Moreover, words cannot abstract all that is to be known about any given thing. Indeed, they do not even abstract all that is essential to the function of that thing (e.g. the word “chair” abstracts what is essential for the function of supporting a person who sits on it, but not what is essential to its functioning at the atomic or nuclear level). So, it is necessary to recognise that all language has an essentially negative and partial relationship to that to which it refers. A. Korzybski has put this relationship very succintly in the assertion:

“Whatever we say it is, it isn’t.”

This statement is not a metaphysical assertion about the basic nature of “what is.” Rather, it is a very deep challenge to the entire structure of our communications, both external and internal (which latter are called “thought”).

To understand this challenge, let us begin with the fact: “We are always talking about it” (“It” refers to anything whatsoever). When we read Korzybski’s statement, our first response is to see that we have already begun to say something about “it” (whatever “it” may happen to be). And then, noticing that “it” is not what we say, and that what we say is at most incomplete abstraction even from what is to be known, we assume that “it” must be something else, as well as something more. But “something else” and “something more” are also what we say “it” is. As we do this for a while, we begin to be struck by the absurdity of the whole procedure. For whatever we say it is, it isn’t.

What is the appropriate response to such a situation? Evidently, one has to stop saying anything at all, not merely outwardly but also inwardly. It is suggested here that if all the “chatter” of thought can really stop, then something new can happen. But even to say this much may be going too far. For if this means that “it” will be “something new,” then the novelty that we say “it” is will be what “it” is not. The paradox with which the reader has to be left is “What is it when there is no saying at all, neither outwardly nor inwardly?”

Dec, 1967        

DEC 23rd 1967

[Date added – CT.]

Dear Jeff,

I am now writing to finish up my answer to your past three letters.

Firstly, congratulations on the birth of your son. I am glad to hear that your wife, Robin, did so well with the new method of childbirth.

Secondly, when you come to England, it would be best to see me before July 15, if possible, as I shall be away in August on vacation.

Thirdly, did you get the talks by Krishnamurti?

Now, to get down to your letters. I want to discuss von Neumann again. Recall what we said:

(a) No axioms are ever really true or false, but only confirmable and falsifiable, as well as (perhaps) actually confirmed and falsified in certain ways.

(b) By saying that hidden variables are ruled out unless qu. mchs. is objectively false, v. Neumann was being very ambiguous at a crucial point.

(c) The legitimate meaning of (b) is that qu. mchs. is objectively falsifiable. The illegitimate meaning is objectively falsified.

By in effect leading the reader to confuse these two meanings, v. Neumann made it impossible to clear this question up at all. Either people accepted his statement to mean objectively falsified and thus unjustifiably rejected hidden variables. Or else, they rejected this meaning, and tried to demonstrate that hidden variables are really possible. To do this is to try to batter down an open door. Thus, the very emphasis on the possibility of hidden variables further confused the issue. For it caused people to overlook the real fact, which is that v. Neumann never really “ruled out” hidden variables at all. The solution to the confusion is to point out that “objectively false” means “objectively falsifiable” in v. Neumann’s context.

But in some ways, the trouble had a deeper root. It has been an ancient Western tradition that what is true is eternally true (emphasized by Plato, among others). Thus if qu. mchs. is true, it is eternally true. Therefore it can never be falsified. And if it is falsifiable, it is not eternally true, therefore not really true, therefore $\underline{\underline{\underline{\text {already}}}}$ “objectively false”.

This is perhaps more deeply why v. Neumann used the word “false” instead of “falsifiable”. For in this tradition, “falsifiable” means “false”.

Meanwhile, the tradition has been changing. So today, words such as “falsifiable” have new meanings, relative to v. Neumann’s day. To overlook this is to add to the confusion.

What you say about violence is very interesting. It is certainly of crucial importance to watch out for violence in one’s work. Here, one must consider not only a violent content to what is said by oneself or others, but also a violent style. Thus. one can say very peaceful things in a violent style. One also has to watch for violent meanings in one’s thoughts and feelings. Violence is a total process, which destroys the proper order of functioning of the mind, in every respect.

The attempt to impose an order on the mind is violence. This is often very subtle. Thus, as I indicated in earlier letters, when a man takes drugs, he usually has some tacit end in view, which is to put his mind into a “better” or “more interesting” state. Even if he says he is only “exploring”, in reality, he will find that he means by this that he is “seeking” some tacitly defined kind of satisfaction. What the drug probably does is to inhibit the critical intelligence sufficiently, so that he will fail to realize that what he has “found” is only what his own thought has put together, so that it is not what is real. Something similar can be done by alcohol, by violent excitement (e.g., primitive dances by the whirling dervishes, etc.) and by other means. When the mind deteriorates to a state of insanity, a similar breakdown of critical intelligence can produce more or less the same kinds of results.

All of this is violence. Indeed, the man “who doesn’t want to know” is being inwardly very violent. For a part of him already knows (e.g., that his state of “ecstasy” is a delusion). Another part then blasts, bashes, anaesthetizes and deadens the part that “knows” so that awareness of this knowledge is destroyed. Is this not an extreme form of violence? The same happens everyday to each of us. Doesn’t the White man already know deeply that human nature is one? To get the pleasure of feeling superior, he bashes, blasts and suppresses this knowledge in his own brain. Similarly the physicist knows that the other fellow’s ideas may have merit, but to get a “good feeling”, he suppresses this knowledge, and presents the other fellow as a kind of idiot.

I think the whole question can be presented succinctly in Korzybski’s terms. What is deepest in the mind is the event level. This includes fleeting sensations of every kind, along with feelings of beauty and ugliness, anger, violence, love, hate, etc., etc.

Here, one must distinguish the above from the positivist view, which treats sensations almost as if they were identifiable things. One must instead emphasize the ineffability of the event level. There is no way to define it or deduce it analytically or to deduce something else from it analytically.

Then come the level of objects. These are perceived and felt as relatively stable entities, having qualities that are signified by the feelings and other movements at the event level. The objects form a total field, which includes a particular object that is called the subject. (That is, “me”, “I”, the “self”, etc.). This latter point has led to endless confusion, which will be discussed more, perhaps, in later letters.

Then comes the level of verbal description. As objects are defined through abstracting from the event level, so verbal descriptions are defined mainly through abstracting from the object level (though such descriptions also abstract directly from the event level to a small extent).

Then comes the level of inference, which abstracts mainly from description. This leads to inference about inference, or abstract thought, going on to an in principle unlimited series of orders of abstraction.

Now what is “the fact”. This is the name of the whole process described above, which is really “facting” or “establishing the fact”. But it is also the name of that part of the process which is in the first two levels. In this meaning, the third level is a description of fact. (But, of course, such a description always contains a vast totality of tacit inferences, accepted as true and as practically the equivalent of descriptions of fact.)

Abstraction is a two-way movement. For each level conditions and shapes the level below it. Thus, to see an object in the dark is already to stabilize and shape the sensations that help reveal it to us. To describe an object verbally is already to call attention perceptually to this object, and to help shape what we see about it. To make an inference is already to alter the terms of our descriptions and thus alter their “shapes”. Higher level inferences perform a similar function relative to lower level inferences.

So there is a two-way stream, abstracting upward and abstracting downward. In a way, as the words abstract descriptively from our perceptual images, so the perceptual images are abstracting from the words. Thus the observer is the observed. Each “element” in the series has both roles in addition to each other element. But one role is properly the dominant one. (The word “dominant” has the meaning it has in music – the dominant theme – and not that of domination of one man who imposes an order on another by force). So let us indicate “dominant” by a heavy arrow, “subordinate” by a lighter arrow. We then have

The above is the normal order of operation. To change the order of dominance in any part is to introduce delusion, confusion and violence. Thus, if an inference is confused with description of fact, we have the following:

For example, the inference “The boy can be either a thief or not a thief” may dominate the descriptive level. So one’s question becomes “Is the boy a thief or not?” The observed facts at the object and event level are then correlated and selected so as to provide an “answer” to this question, which would be either “Yes” or “No”. Thus we get delusion and confusion.

There is already violence in this procedure. For the delicate movements of perception at the lower levels are already being blocked by the effects of the “heavy arrow” from the inferential to the descriptive level. Those perceptions that would contradict the inference are either ignored or suppressed or reinterpreted, so as to fit what is in the inferential level.

One has here a kind of “vicious circle” similar to a “feed-back” loop in electronic circuits. Descriptions of fact shaped too strongly by the inferential level, come back up, to shape the very inference that is moving downwards. Thus, the whole process builds up to a fantastic intensity, which makes it very hard to alter.

This violence gets much worse when the “heavy arrow” goes down into object and event levels. In this way, the inference of anger leads to the feeling of anger, the inference of pleasure to the feeling of pleasure, the inference of security to the feeling of security, the inference of danger to the feeling of danger (i.e., fear). One gets “feedback loops” running from feeling and perceptions to thoughts and back down again, with tremendous intensity. No wonder then, that there is violence.

Here what is important is the role of inference of a “self”. One of the objects in the field of perception – i.e. the body – is already non-verbally understood to be a central source of perception and action. This object is described by the name “I” and various properties are attributed to it. When the order of emphasis in abstraction is reversed, this produces “heavy arrows” going down through descriptive, object and event levels. Because the word “I” calls attention to the whole field of what is felt and experienced, as well as to all action initiating in mind and body, these “heavy arrows” lead into key areas of brain function. “Feed-back” loops are set up, which mix up the entire process. The result is illusion, delusion, confusion, violence, destruction, suffering, sorrow, etc.

As long as there is “I” in this role, there cannot be anything other than violence and delusion. This has gone on for tens of thousands of years. Can it end? This is the key question, on which the survival of the race may well depend.

I’ll write more later.

Best regards

Dave