CEPA eprint 1729 (HVF-133)
Through the Eyes of the Other
Foerster H. von (1991) Through the Eyes of the Other. In: Steier F. (ed.) Reflexivity and Research. Sage Publications, London: 63–75. Available at http://cepa.info/1729
Table of Contents
Teach me my, not your language.Teach them their, not your language.Teach us our, not your or their language.(Brim, 1986)
‘Almost everything in metaphysics is controversial and it is therefore not surprising that there is little agreement among those who call themselves metaphysicians about what precisely it is that they are attempting.’ Thus begins W. H. Walsh’s article ‘Metaphysics, Nature of’ in the Encyclopaedia of Philosophy.
I shall support Walsh’s observation by not seeking agreement with others who call themselves metaphysicians about what precisely it is that they are attempting, for I am going to say precisely what I want us to see when we turn into metaphysicians: We turn into metaphysicians, whether or not we call ourselves such, whenever we decide upon questions that are in principle undecidable.
There are indeed among questions, problems, proposals, propositions, etc., those that are decidable and those that are in principle undecidable. The question ‘Is the number 372,153,102 divisible (without remainder) by 2?’ clearly is decidable. Note that this decision is not a bit more difficult to make, even if the number to be divided has not 9 as above, but a million, a billion or a trillion digits.
Of course, one can invent other questions that can be decided as easily as this one, or much more difficult ones, or extraordinarily difficult ones, whose decidability, however, is guaranteed by the acceptance of the rules of a formalism that permits one to reach from any node in a complex crystal-like framework of relations any other node by worming one’s way along connecting links, Syntax, arithmetic, Aristotelian syllogisms, etc., are such formalisms.
It may be argued that more than a half century ago the Viennese mathematician Kurt Gödel (1931) showed that even the formalism of Principia Mathematica’s awesome logical framework, so carefully constructed by Alfred North Whitehead and Bertrand Russell (1910–1913) had been invaded by undecidables.
It is, however, not necessary to turn to Geode] or to Whitehead or to Russell to draw attention to questions that are in principle undecidable. For instance, questions about the origin of the universe are in principle undecidable. This is apparent by the many different answers that are given to these questions. Some claim that the origin of the universe was a singular act of creation; others say that there was never a beginning: the universe is a perpetually self-regenerating system in an eternal dynamic equilibrium. Still others insist that what appears to us now as our universe are the remnants of a ‘Big Bang’ that occurred perhaps 10 or 20 billion years ago, whose faint echo one is supposed to ‘hear’ over large radio antennas. I shall leave it to the reader to find out what Eskimos, Arapesh, Hindus, Chinese, Ibos, Balinese and others would tell us about this event. In other words: Tell me about the origin of the universe, and I will tell you who you are.
The difference between decidable and in principle undecidable questions should now be sufficiently clear for me to introduce the following theorem (von Foerster, 1989):
Only those questions that are in principle undecidable, we can decide.
Simply because all decidable questions are already decided by the choice of the relational framework within which these questions have been posed, together with the rules that connect any proposition (say, ‘the question’), with any other proposition (say, ‘the answer’) within that framework. Sometimes it may go fast, at other times it may take an excruciatingly long time until, via compelling logical steps, the irrefutable ‘yea’ or ‘nay’ is obtained.
But we are under no compulsion, not even under that of logic, when we are making decisions about in principle undecidable questions. However, with this freedom gained, we have assumed the responsibility of our decision. This shows that the complement to necessity is not chance (Monod, 1972) but choice. The choice made on the following two pairs of in principle undecidable questions will be the content of the remaining two points of this chapter:
(1) Here is one pair of questions:
Am I apart from the universe?(That is, whenever I look, I’m looking as through a peephole upon an unfolding universe.)
OrAm I a part of the universe?(That is, whenever I act, I’m changing myself and the universe.)
(2) And here the other pair:
Is the world the primary cause?(That is, my experience is the consequence.)OrIs my experience the primary cause?(That is, the world is the consequence.)
Why are these questions undecidable in principle? Simply because if they were decidable, a framework must have been chosen within which they are decidable. But since choosing a framework is, in itself, deciding an undecidable question, we can take the decisions upon these questions as devices for generating the appropriate framework.
Those who decide for themselves to be observers of an independent universe and report to us the results of their observations have provided us with the vast body of orthodox wisdom. The power of this position is trust in the abilities of ultimately describing unequivocally the uniqueness of the universe – ‘Truth’ – and of describing this universe without the properties of the observer entering his descriptions – ‘Objectivity’. The notions of Truth and Objectivity guarantee the popularity of this position, the former promoting authority – ‘It is as I tell it’ – the latter removing responsibility – ‘I tell it as it is’. Moreover, by separating oneself from the universe, one separates oneself from others as well. Hence, without consequences for oneself one is now in a position to tell the others: ‘Thou shalt …’ or ‘Thou shalt not …’ The method here is to reject reflexivity.
When I ask myself ‘Am I a part of the universe?’ and answer ‘Yes, I am’, I decide here and now that, whenever I act, not only /change, but the universe changes as well. I have adopted this position not because of an antagonism to orthodoxy, or orthodoxy’s many fundamental flaws as, for instance, the impossibility of describing anything unequivocally (for it is not I, but the listener, who determines the meaning of my utterance); or the impossibility of achieving ultimate truth (for there is no way of comparing what is the case with what I think is the case, since I can compare this only with what I think I thought was the case); or the impossibility of making objective descriptions (for without the observer’s ability to observe and describe, there wouldn’t be any descriptions in the first place), etc.
I have adopted the position of being part of the universe because it
ties me with my actions inseparably to all others, and thus establishes a prerequisite for a foundation of ethics.
In what way?
I see it in the way that for any discourse we may have, say, in science, philosophy, epistemology, psychotherapy, even in politics, we are to master the use of our language so that ethics is implicit; that is, so that language does not degenerate into preaching morals.
We have from Wittgenstein’s Tractatus (1921) Proposition 6.421:
‘Es ist klar, dass sich die Ethik nicht aussprechen lasst’, which I translate thus: ‘It is clear that ethics cannot be articulated.’
What does he mean by that? He goes on (1961: 6.422) to put this into a larger context:
When an ethical law of the form ‘Thou shalt …’ is laid down, one’s first thought is ‘And what if I do not do it?’ It is clear, however, that ethics has nothing to do with punishment and reward. (However) there must indeed be some kind of ethical reward and ethical punishment, but they must reside in the action itself.
Here ethics becomes implicit; the method is reflexivity, and commandments are no longer ‘Thou shalt …’ or ‘Thou shalt not but ‘I shall …’, or shall not …’
Notions of reflexivity and self-reference that turn on themselves, that need themselves to come into being, that preserve the tie between observer and observed, speaker and speech, and partners in dialogue are now at the core of at least five branches of science and philosophy. They are in biology ‘autopoiesis’ (Varela et al., 1974); in mathematics ‘Eigen-values’, ‘Eigen-behavior’ (von Foerster, 1976) and ‘attractors’ (Abraham and Shaw, 1981); in logic a ‘calculus of self-reference’ (Varela, 1975); in linguistics ‘performance utterances’ (Austin, 1961); in epistemology ‘reality as (social) construct’ (Watzlawick, 1984).
The intrinsic nature of ‘self, turning upon itself, becoming visible only while turning, is immanent in language itself. First of all, language speaks about itself: there is a word for language, viz., ‘language’, a word for word, viz., ‘word’, etc. Then there are questions: ask ‘Why “why?” ?’ or ‘What is a question?’, or ‘What is language?’, and you may see the answers turning back to the questions, or the questions carrying the answers on their back; or you may answer the questions with a question: ‘So what?’ And then there are the two antagonistic tracks on which language runs all the time: its appearance contradicted by its function. In its appearance language seems to be denotative, monologuing about things in a world ‘out there’; in its dialogic function, however, it is connotative, appealing to concepts in the other’s mind. In its appearance it seems that the speaker describes what he sees when looking as through a peephole upon an unfolding universe; in its function, however, language is a coordinating agent for actions among conversing human beings (cf. Winograd and Flores, 1987). We have from Martin Buber (1961):
Contemplate the human with the human, and you will see the dynamic duality, the human essence, together: here is the giving and the receiving, here the agressive and the defensive power, here the quality of searching and responding, always both in one, mutually complementing in alternating action, demonstrating together what it is: to be human. Now you can turn to the single one, and you recognize him as human for his potential of relating; then look at the whole and recognize the human for his richness of relating. We may come closer to answering the question: what is a human?, when we come to understand him as the being in whose dialogic, in his mutually present two-getherness, the encounter of the one with the other is realized at all times.
When 350 years ago Renê Descartes was tortured by doubts about his own existence – ‘Am I?’ or ‘Am I not?’ – he wants us to believe that he solved his problem by the self-referential monologue ‘Cogito ergo sum!’ —’I think, therefore, I am!’ This is language in its appearance as he knew very well, otherwise he would not have published his insight soon after in his Discours de la methode for the benefit of others. Hence, in all honesty, he should have exclaimed ‘Cogito ergo sumus!’ – ‘I think therefore we are!’
In its appearance, the self-referential nature of language generates consciousness about oneself: self-consciousness; but in its function, by embracing the other as dialogical participant, it is the origin of conscience.
Is the world the primary cause and my experience the consequence, or is my experience the primary cause and the world the consequence?
Those who decide for themselves that the world is the primary cause of their experience, and who report to us their experience, convinced they are talking about the world, have been seduced by the persuasive appearance of language: their speech is monologue. These are the same I mentioned before, namely, those who decide for themselves to be apart from a universe that they observe as it unfolds. Hence, what I said before about the power of the limits of this position applies here too.
When I ask myself: ‘Is my experience the primary cause and the world the consequence?’, and I answer ‘Yes, it is!’, then I decide here and now not only what my world is to become, but also who I’m going to be. I have adopted this position because it ties my actions inseparably to my responsibility.
Speaking about causes, their consequences and effects, the question arises: ‘What is the agent that transforms a cause into its effect; what is the operation that does this transformation?’ It shows the complete canon of causation to be triadic: ‘cause/operator/effect’.
The origin of this schema of an explanatory device can be traced back to Aristotle who saw its formal equivalence with logical syllogisms, particularly with that of deductive inference. The schema of syllogism ‘minor premise/major premise/conclusion’ he projected onto causation, with cause, operator, effect taking the corresponding positions. With ‘All men are mortal’ as major premise (operator), ‘Socrates is a man’ as minor premise (cause), one can draw the irrefutable conclusion (effect) from these premises: ‘Socrates is mortal’.
The sense of certainty, reliability, and infallibility that this schema emanates propelled it to become the mainstay of Western thought, dislodging in this process many other explanatory devices as, for instance, metaphor, analogy, hyperbole, parable, etc. Depending upon the field of study, the members of that triad got different names. In physics we have, of course, cause/Laws of Nature/effect; in ethology stimulus/organism/response; in (some branches of) psychology motivation/personality/behavior; in mathematics we have x (the independent variable), f (a function), and y (the dependent variable), that is, x / f / y; and, finally, in computer science the triad is input/computation/output.
Ever since Alan Turing, the inventor of the ‘Turing Machine’, called his logico-mathematical framework a ‘machine’ (Turing, 1936), this term is now being applied to other well defined functional properties of abstract entities, and not necessarily to an assembly of cogwheels, buttons and levers, or chips, discs, and connectors, although such assemblies may represent these abstract functional units.
One distinguishes two kinds of such machines, the trivial and the non-trivial machine (von Foerster, 1970, 1972). A trivial machine is characterized by a one-to-one relationship between its input (stimulus, cause, etc.) and its output (response, effect, etc.). This invariable relationship is the ‘machine’, and since this relationship is, by our choice, determined once and for all, this is a deterministic system; and since an output once observed for a given input will be the same for the same input given later, this is also a predictable system.
Another feature of trivial machines is that they are analytically determinable. Anyone who does not know the input/output relation ‘f’ of a trivial machine – say, the one depicted in Figure 4.1 – will be able, after a few trials, to establish its input/output relation: a translation of the first four letters of the Roman alphabet into those of the Greek alphabet. It is not difficult to see that for much larger, or even very large, numbers of input states the analytical problem is trivial: the number of trials is precisely the number of distinguishable input states.
Non-trivial machines, however, are quite different creatures. Their input/output relationship is not invariant, but is determined by the machine’s previous operation. In other words, its preceding steps determine its present behavior. While these machines are again deterministic systems, there are some that are in principle, and others for all practical purposes, unanalyzable, hence unpredictable (Gill, 1962): an output once observed for a given input will most likely be not the same for the same input given later.
In order to grasp the profound difference between these two kinds of machine, it may be helpful to envision ‘internal states’, ‘z’, in these machines. While in the trivial machine only one internal state participates always in its operation, in the non-trivial machine it is the shift from one internal state to others that makes it so elusive.
Figure 4: 2 is the simplest version of a non-trivial machine, a machine with only two internal states: ‘1’ or ‘2’. The two tables in Figure 4.2 list under the corresponding headings the performance of this machine when in either one of the two internal states. In the third column it also lists the next state, z’, into which it will switch after completion of its present operation. For example, this machine, initially in state ‘1’ given the input B, will produce output
Upload missing image:
and go into state ‘2’; given B again, it will produce
Upload missing image:
and return to ‘1’; given now C, it will again produce
Upload missing image:
, and so on and so forth.
It may be worth noting that the number, N, of different non-trivial machines that can be synthesized, limited to precisely 4 inputs and 4 outputs as above, is not trivial indeed:
N = 24^4 x 24^4 = 2512 or about 10155
If the analyst who wishes to solve the ‘machine identification’ problem of the machine in Figure 4.2 has a computer at his or her disposal that computes an x,y pair every nanosecond (10-12 sec.), he or she has to let it run an average of about 10136 years, or approximately ten trillion to the tenth power ages of our universe, to have a chance to catch the secrets of this machine; that is non-trivial machines are transcomputational.
It is also clear that for systems with more internal states the transcomputability of their workings is even more pronounced. Moreover, it is not difficult to synthesize non-trivial machines for which the machine identification is in principle unsolvable. In other words, the rules of transformation, the functions of the operator, the Laws of Nature, the major premises, etc., the links between cause and effect are in non-trivial systems analytically indeterminable. To put it even stronger: the notion of causality in analytical studies has lost its meaning, hence is inapplicable; or in Wittgenstein’s version (1961: Proposition 5.1361): ‘We cannot infer the events of the future from the events of the present. Belief in the causal nexus is superstition.’
Causation as explanatory principle for observed systems has joined Gregory Bateson’s (1972) no man’s land he invented for another such explanatory principle, ‘instinct’, which explains: ‘Anything—almost anything at all. Anything you want it to explain.’ For synthesized and trivial systems, however, causation remains to be an operative conceptual device. Why? Because in synthesizing trivial or non-trivial machines we have chosen the framework within which all relational questions ‘Why this when that?’ are decidable. And when analyzing a system, we deem it to be trivial, it is we who make that decision. But also the most trivial machine money can buy, say, a Rolls-Royce, may show its true, history dependent, non-trivial nature by, perhaps, on one occasion refusing to go on in the middle of the road. A professional trivialisateur, upon being called, finds it needs petrol to resume again its expected status of triviality.
When asked, all my friends consider themselves to be like non-trivial machines, and some of them think likewise of others. These friends and all the others who populate the world create the most fundamental epistemological problem, because the world, seen as a large non-trivial machine, is thus history dependent, analytically indeterminable, and unpredictable. How shall we go about it?
I can see three strategies that are currently applied to alleviate this situation: ignore the problem; trivialize the world; develop an epistemology of non-triviality.
The most popular version of attacking this problem is of course to ignore it, but the method of universal trivialization follows not too far behind. One may call it the ‘Laplacian solution’, for it was he who eliminated from his considerations all elements that could cause trouble for his theory, himself, his contemporaries, and other nontrivial annoyances, and then pronounced the universe to be a trivial machine (LaPlace, 1814): if for a superhuman intelligence the present condition of all particles in the universe were known ‘nothing would be uncertain and the future and the past would be present to his eyes. The human mind offers, in the perfection which it has been able to give to astronomy, a feeble idea of this intelligence.’
The tremendous attraction of having to deal with something analyzable, reliable, and predictable persuades one to pay for guarantees that our watches, lawnmowers, airplanes, etc., maintain their no-choice quality. The danger begins when we extend this demand to others, to our children, our families, and other larger social bodies by trying to trivialize them; that is, by reducing their number of choices, instead of enlarging it (von Foerster, 1973).
When about a half century ago the first signs of intrinsic uncertainties in observational procedures on the level of elementary particles were recognized by Heisenberg’s uncertainty principle, and that recognition was later extended to the unanalyzability of complex systems with their large repertory of internal states, no strategies for approaching these difficulties existed.
Only about 25 years ago, the insight that these systems do not operate in isolation, but are embedded in a larger context, that they are acted upon by something, maybe by another non-trivial system upon which they act in turn, and that inner-action, rather than action, should be attended to, brought forth an avalanche of theoretical, experi-mental, and clinical work. Based on this participatory position, the many schools of thought extend from formal logic (Lofgren, 1983), to mathematics (Abraham and Shaw, 1981; Peitgen and Richter, 1986), astronomy (Buehler and Eichhorn, 1987), to managerial and social sciences (Ulrich and Probst, 1984), to theoretical biology (Glass and Mackay, 1988), to biological dynamics (Koslow et al., 1987), to family and systemic therapy (Malagoli Togliatti and Telfener, 1983; Hargens, 1989; Segal, 1986), and to popularizations (Gleick, 1987), to mention a few.
The foundation for the vast expansion of this interest and activities is the demonstration of the operational equivalence of an arbitrary large number of interacting non-trivial machines with a single, nontrivial machine recursively operating on itself (see Figure 4.3), and the demonstration that under this condition these systems approach dynamic equilibria that go today under various names: fixed points, Eigen-values, Eigen-behaviors, attractors, strange attractors, and so on, which account for the stability of things observed or created, be they objects, concepts, languages, customs, rituals, cultures or whatever. These arise when reflexivity, recursivity, circularity,
reproduces the entity it operates on, as in Figure 4.4 (Kauffman, 1987). For instance √1 = 1, will assume this identity whatever the number was from which one started recursively square-rooting (try with your own calculator); or when a sentence produces its own truth:
THIS SENTENCE HAS THIRTYONE LETTERS
in which ‘THIRTYONE’ is of course one Eigen-value of this sentence (find another one), or when a utterance says what it does, such as ‘I apologize’, ‘I promise’. But to whom does one apologize, to whom do we promise? It is the other, through whose eyes we may see ourselves.
I met the Viennese psychiatrist Victor Frankl after the Second World War. He had come out from the hell of the death camps, the only surviving member of his family. In postwar Vienna, conquered by the Russians and then occupied by the four Allied powers, his presence and practice as a therapist, as a healer of traumatic experiences, was of lifegiving importance. One day a man in a deep depression was brought to him. He and his wife had been in different extermination camps, miraculously both had survived and had been reunited in Vienna. They had been together only a few months when she died of a disease she had contracted in the camp. The man became despondent. He stopped eating, he ceased to participate in life around him, Friends brought him to Frankl, and he and Frankl talked for a long time. Finally, Frank! asked him: ‘Assume that God would give me the power to create a woman identical to your wife. You would not be able to see or sense any difference; appearance, taste, conversations, memories, all would be the same as your wife’s. Would you ask me to produce such a woman?’ There was a long silence. Then the man said: ‘No’. Frank] said: ‘Thank you’; and the man walked home and began to participate in life again.
When I heard this I asked Dr Frankl: ‘What happened? What did you do?’; and he replied: ‘All his life, in the union of these two humans, the man had seen himself through the eyes of his wife. When she died, he was blind. But when he saw that he was blind, he could see! So it is with us: we see ourselves through the eyes of the other.’
Abraham, R. and Shaw, C. (1981) Dynamics – The geometry of behavior. Santa Cruz: Aerial Press,
Austin, J.L. (1961) ‘Performative utterances’, in J. Urmson and G. Warnock (eds), J.L. Austin: Philosophical Papers.
Bateson, G. (1972) ‘Metalogue: What is an instinct’, in G. Bateson, Steps to an ecology of mind. New York: Ballantine Books. pp. 38-60.
Brün, H. (1986) ‘#35’ in My Words and Where I Want Them. London: Princelet Editions.
Buber, M. (1961) Das Problem des Menschen. Heidelberg: Lambert Schneider.
Buehler, G.R. and Eichhorn, H. (eds) (1987) Chaotic Phenomena in Astrophysics. New York: An. NY Acad. Sci., 497.
Gill, A. (1962) Introduction to the Theory of Finite-State Machines. New York: McGraw-Hill,
Glass, L. and Mackay, M.C. (1988) From Clocks to Chaos. Princeton: Princeton University Press.
Gleick, J. (1987) Chaos: Making a New Science. New York: Viking.
Gödel, K. (1931) ‘Üeber formal unentscheidbare Saetze der Principia Mathematica and verwandter Systeme’. I Monatsh. Mat. Phys, 38: 173-98.
Hargens, J. (ed.) (1989) Systemic Therapy: A European Perspective. Dortmund: Borgmann.
Kauffman, L.H. (1987) ‘Self-reference and recursive forms’, J Soc. and Biol. Struct., 10: 53-72.
Koslow, S.H., Mandell, A.J. and Shlesinger, M.F. (eds) (1987) Perspectives in Biological Dynamics and Theoretical Medicine. New York: An. NY Acad Sci., 504.
LaPlace, P.S. de (1814) Essai philosophique sur les probabilites. Paris.
Lofgren, L. (1983) ‘Autology for parts and wholes’, in Parts and Wholes: An inventory of present thinking about parts and wholes. Vol. II: Commentary. Stockholm: Forskningsradsnamnden, 4-23.
Malagoli Togliatti, M. and Telfener, U. (eds) (1983) La terapia sistemica. Rome: Astrolabio.
Monod, J. (1972) Chance and Necessity. London: Collins.
Peitgen, H.0, and Richter, P.H. (1986) The Beauty of Fractals. Berlin: Springer Verlag.
Segal, L. (1986) The Dream of Reality: Heinz von Foerster’s Constructivism. New York: Norton.
Turing, A.M. (1936) ‘On computable numbers, with an application to the Entscheidungsproblem’, Proc. London Math. Soc. 2(42): 230-65.
Ulrich, H. and Probst, G.J.B. (eds) (1984) Self-Organization and Management of Social Systems. Berlin: Springer Verlag.
Varela, F. (1975) ‘A calculus for self-reference’, Mt. Jnl General Systems, 2: 5-24.
Varela, F., Maturana, H. and Uribe, R. (1974) ‘Autopoiesis: The organization of living systems, its characterization and a model’, Biosystems, 5: 187-96.
von Foerster, H. (1970) ‘Molecular ethology, an immodest proposal for semantic clarification’, in G. Ungar (ed.), Molecular Mechanisms in Memory and Learning. New York: Plenum. (Also in H. von Foerster (1981), Observing Systems. Salinas: Intersystems. pp. 149-90.)
von Foerster, H. (1972) ‘Perception of the future and the future of perception’, Instructional Science, 1: 31-43. (Also in H. von Foerster (1981), Observing Systems. Salinas: Intersystems. pp. 191-204.
von Foerster, H. (1973) ‘On constructing a reality’, in W. Preiser (ed.), Environmental Design Research 2. Stroudberg. Dowden, Hutchison and Ross. pp. 35-46. (Also in H. von Foerster (1981) Observing Systems. Salinas: Intersystems. pp. 41-61.)
von Foerster, H. (1976) ‘Objects: Tokens for (eigen-) behaviors’, Cybernetics Forum, 8: 91-6. (Also in H. von Foerster (1981), Observing Systems. Salinas: Intersystems. pp. 273-86.)
von Foerster, H. (1989) ‘Wahrnehmen wahrnehmen’, in Philosophien der neuen Technologie. Berlin: Merve Verlag. pp. 27-40.
Watzlawick, P. (1984) The Invented Reality. New York: Norton.
Whitehead, A.N. and Russell, B. (1910-1913) Principia Mathematica. Cambridge, England: Cambridge University Press.
Winograd, T. and Flores, F. (1987) Understanding Computers and Cognition. Norwood, NJ: Ablex.
Wittgenstein, L. (1921) ‘Logisch-philosophische Abhandlung’, in W. Ostwald (ed.) Ann. d. Naturphil. Unesma Leipzig.
Wittgenstein, L. (1961) Tractatus Logico-Philosophicus (edited by A.J. Ayer). London: Routledge & Kegan Paul.
Found a mistake? Contact corrections/at/cepa.infoDownloaded from http://cepa.info/1729 on 2016-05-13 · Publication curated by Alexander Riegler