mercredi 26 juillet 2017

Trajective chains in mesology / Augustin Berque

The Merchant Navy: The chain-locker (Henry Carr: 1942)
The International Society for the Philosophy of Chemistry (ISPC)
Colloque international de philosophie de la chimie, Paris, 3-6 juillet 2017

Trajective chains in mesology

von Neumann chains in physics, etc. – and in chemistry ?

by Augustin BERQUE

Abstract – The leading idea of mesology (the study of milieux) originates in Plato's Timaeus, with the paradoxical relationship of chôra (milieu) and genesis (relative being), which are posed as both an imprint and a matrix of each other. Though  foreboded, the idea of milieu was locked out by Plato's rationalism, because it infriges the principle of the excluded middle : A (an imprint) cannot be non-A (a matrix), and there is no third term, both A and non-A. After Uexküll, who proved experimentally that an animal and its proprer milieu (Umwelt, not to be confused with the general data of the environment, Umgebung) are precisely in such a relationship, and after Watsuji, who, as for the human, named this relationship fûdosei (mediance) and defined it as the structural moment of human existence, mesology has logically and ontologically formulated empirical reality r (that of concrete milieux)  in the following way : r = S/P, which reads "reality r is the subject S as the predicate P". Reality is neither S (the Real in itself) nor P (a subjective representation), but emerges in a process called trajective chain by dint of which, indefinitely, S is assumed as P, producing S/P, which in its turn is hypostasized into S' by P', and so on in the following way : (((S/P)/P')/P'')/P'''... etc.  Homologous chains have been observed also in physics, and named "Neumann chains". Then what about chemistry?
Plan – § 1. Starting up mesology ; § 2. The establishment of mesology ; § 3. Mediance and trajective chains ; § 4. Trajective chains, semiologic chains, von Neumann chains ; § 5. Chemistry in the concrete milieux of the Ecumene.

§ 1. Starting up mesology
You will not find “mesology” in the Shorter Oxford English Dictionary; nor will you find mésologie in most dictionaries of the French language, although this word was listed in the first edition (1906) of the Petit Larousse, with the following definition: “Partie de la biologie qui traite des rapports des milieux et des organismes”.[1] Here, the word milieu is understood in the present sense of “environment”, i.e. the external conditions affecting the life of a plant or an animal. The word mésologie was coined by a physician, Charles Robin, who proposed it as one of the first tasks of biology at the inaugural session of the Société de biologie in Paris on June 7th, 1848. Robin, a direct disciple of Auguste Comte, professed positivism, that turn of mind which the Oxford Shorter defines as “recognizing only positive facts and observable phenomena and rejecting metaphysics and theism”. That same mesology was also advocated by Robin’s colleague, Louis-Adolphe Bertillon, who gave it an uncontestable lustre in the second half of the XIXth century, as testifies its presence in the Petit Larousse of 1906. Although Ökologie, introduced by Haeckel in 1866 in the second volume of his Generelle Morphologie der Organismen, had penetrated in French as écologie as soon as 1874, it was still not listed in that first edition of the Petit Larousse. Yet, écologie was to supersede mésologie in the XXth century, and the latter faded away from dictionaries after the Second World War.  
            One may wonder why the word mésologie was replaced by that of écologie, since their initial definitions were almost exactly the same.[2]  I surmise that this may be due to two reasons. First, the scope of mésologie, which encompassed also the social sciences in Bertillon’s  mind, was too vast for a single positivistic science. Second, the influence of the English language, in which only “ecology” existed. Indeed, although “milieu” can be found also in English, and, as in French, can be understood as a synonym of “environment”, English adopted directly “ecology” as the science of environment, without the stage of a “science of milieux” (i.e. mesology).

§ 2. The establishment of mesology
Now, while mésologie was sinking into oblivion in France, it was born again in Germany under the name of Umweltlehre in the works of Jakob von Uexküll (1864-1944), and then in Japan, in Watsuji Tetsurô’s[3] (1889-1960) essay Fûdo (Milieu, 1935), under the name of fûdogaku 風土学. Uexküll was a biologist – famous as one of the founders of ethology, and the forerunner of biosemiotics –, whereas Watsuji was a philosopher, but both were influenced by phenomenology. This led them to pose a founding difference between environment (Umgebung, kankyô 環境) and milieu (Umwelt, fûdo 風土) : whereas environment is a universal objective raw datum, milieu is the peculiar way in which a certain living being, singular or plural (a species, an organism, a person, a society…), interprets the environment and elaborates its proper relationship with it.
            Accordingly, milieu supposes the subjecthood of the concerned being, whereas environment does’nt. This led Watsuji to conceive of mesology as a phenomenological hermeneutics (kaishaku genshôgaku 解釈現象学). On the other hand, environment is a typical modern object in the tradition of Descartes’ dualism and mechanicism, while milieu is something else.   
            It is not impossible that Watsuji, who was much younger than Uexküll, had heard of him during a stay he made in Germany in 1927-1928, possibly through his reading of Heidegger, who was deeply influenced in those years by Uexküll’s views as testifies his seminar of 1929-1930[4]; but this is only a conjecture. Be it as it may, the two authors published almost simultaneously, each of them on his own side, respectively one of the two founding classics of the new mesology: Streifzüge durch die Umwelten von Tieren und Menschen (Raids into the milieux of animals and humans, 1934) and Fûdo. Ningengakuteki kôsatsu (Milieux. A study of the human interlink, 1935)[5]. Their points of view are homologous, except that Uexküll considers the living in general (in fact mainly animals), and Watsuji the human in particular.
            Of course, Uexküll being a naturalist, and Watsuji a philosopher, they do not have the same methods.  That of Watsuji is historical, whereas that of Uexküll is experimental, and as such exemplarily scientific. This is what makes his theses a true revolution, since it is by using the very methods of modern science that he rejects its mechanicism. For him, animals are not machines reacting to stimuli, they are machinists (Maschiniste) interpreting signs; and correlatively, they are not simply determined by the environment, they create their own milieu in a mutual relationship with it.  
            Accordingly, in the same environment (Umgebung), the milieu (Umwelt) of a certain species differs from that of another species; but in any case, the milieu is always the best suited for the concerned species, as the species is for that milieu. It follows that, though pessimal the environment may seem, the milieu is always optimal. Subsequent discoveries of extremophile species have ceaselesssly corroborated this rule.
            Speaking of “objective reality” in such conditions is at best only an abstraction, and in fact expresses a naive anthropocentrism. As Uexküll’s experiments have proven, “that an animal may ever enter into relation with an object, this tacit hypothesis [that of behaviourism] is false”. [6] That with which it enters into relation, and which, for it, is concrete reality, are the things proper to its milieu, not the universal objects of the environment, as they may be considered by ecology with the look from nowhere of modern science (which is no more than the scientist’s Umwelt).
            Let us give an example. An electromagnetic wave of λ = 700 nm is a universal physical datum. In our species, Homo sapiens, this wavelength is perceived (interpreted) as red. In the species Bos taurus (the cow), this same wavelength is not perceived as a colour. It is outside the range of visible colours. For a bull, red does not exist. A bull cannot enter into relation with red, as Uexküll would have put it. In a corrida, what excites the bull is not the colour of the muleta, it is the gesticulations of the toreador. Likewise, the human eye does not perceive infrared, which a snake’s eye perceives. It does not perceive ultraviolet, which a butterfly’s eye perceives. In addition to this specification of things proper to the human species, crops up a further specification, proper to each human culture. Red, for example, is traditionally not the colour of a wedding dress in Europe, whereas it is in Japan. A red light is perceived as “stop!” by an average driver, but for the red guards of the Cultural Revolution, it meant on the contrary “forward!”. Etc.
            This is to say that concrete things, as such and not abstracted into objects, are never universal, nor neutral; starting from the physiological level, they are always, in relation with a certain subjecthood, fraught with a specific meaning and value. This is what Uexküll calls Ton (tone), which I shall render with as[7]. He declines these different kinds of reality of one and the same object into diverse categories : Esston (as food), Schutzton (as a shelter), Wohnton (as a habitat), Hinderniston (as an obstacle), etc. The same grass, for example, will exist as food for a cow, as an obstacle for an ant, etc.
            One should note that these different Töner (or different as), in Uexküll, are often composed with a verb : essen (eat), schutzen (defend), wohnen (inhabit), hindern (prevent), etc. There is here obviously an ongoing action. Uexküll even accentuates this active aspect by speaking also of Tönung (tonation) ; for instance Esstönung, tonation as food. That action, in short, is a certain process, which transforms some of the abstract data of the environment into the concrete reality of a milieu.
            In mesology, this process is called trajection[8]. Inbetween the two theoretical poles of the subjective and the objective, it “trajects” the Umgebung into an Umwelt, the abstract objects of the environment into the trajective things or affordances (one should of course remind here Gibson’s concept) which constitute a concrete milieu.
            This radically transgresses the ontological frame of the modern Western classic paradigm (MWCP). The concrete things of a milieu are not substantial objects, subsisting out there in the identity of their in-itselfness (Ansichheit); they are always in the making in their interaction with a subject, and reciprocally, the subject is always in the making, in interaction with these things; in other words, in a dynamic coupling with a milieu.
           Uexküll, however, did not make such an ontological inference; this is because he was a scientist rather than a philosopher.[9] Watsuji on the other hand, who was a philosopher, posed in the first line of Fûdo the ontological concept of fûdosei風土性, which he defined in the same line as “the structural moment of human existence (ningen sonzai no kôzô keiki 人間存在の構造契機)”. I translated fûdosei[10] with médiance because of that definition. Mediance is indeed the dynamic coupling, or Strukturmoment (kôzô keiki 構造契機), of two “halves” (medietates in Latin, hence “mediance”), one which is – to borrow Plato’s vocabulary – a certain being’s physical topos, and one which is that being’s chôra, or milieu.  

§ 3. Mediance and trajective chains
Though Uexküll did not propose such an ontological concept as mediance, he uses an image which amounts to the same : Gegengefüge, that is the counter-fitting, or mutual adequacy, of the animal and its milieu; and  though the definition of mediance was formulated for the first time in 1935 by a Japanese philosopher, the corresponding idea has profound roots in the history of Western ontology, since its first manifestation can be traced as far back as Plato’s Timaeus, with the non-concept of chôra. I write “non-concept” because, although Plato suggests a trajective relationship between relative Being (genesis) and its milieu (chôra), chôra being both the imprint (ekmageion, 50 c 1) and the matrix (mother : mêtêr, 50 d 2, or nurse : tithênê, 52 d 4) of genesis[11], he renounces to define chôra. He only circumambulates around it by dint of such metaphors, which moreover are contradictory, and eventually dismisses it as a dream (oneiropoloumen blepontes, 52 b 3) or “bastard reasoning” (logismô tini nothô, 52 b 2).
            Indeed, being both an imprint (A) and a matrix (non-A), chôra, i.e. the idea of milieu, challenges the three principles of identity (A is A), of contradiction (A is not non-A) and of the excluded middle (which would be both A and non-A, or neither A nor non-A)., which have been the pillars of Western rationalism until quantum mechanics has proved that a same particle can be at the same time both a corpuscle (A) and a wave (non-A).
            Twentieth-century physics, thus, has revived the problem which Western rationalism had foreclosed for more than two millenia. Now, this was not the case East of Aden, where tetralemmas have been of common use from India to Japan, especially so in Buddhism. I shall understand here tetralemma in Yamauchi Tokuryû’s order, that is 1. A, 2. non-A, 3. neither A nor non-A, and 4. both A and non-A. As he argued convincingly in Logos and lemma[12], putting binegation (neither A nor non-A) in fourth position, as is usually done, leads literally to nothing, whereas putting there biassertion (both A and non-A) opens up all sorts of possible worlds. Now, this seems to be the case in mesology, since a single and same object A in the environment (e.g. a tuft of grass) can exist as many kinds of realities (e.g. as food for a cow, as a shelter for a beetle, etc.) according to the respective mediances of these diverse beings.  
            Trajection is precisely the process through which a same object becomes (gignomai : genesis) different kinds of realities. This process is analogous to a predication in logic, as expressed by the formula S is P, in which S is the logical subject (what the matter is about) and P the predicate (what is said about S).
            One needs here to bear two points in mind : that 1. the logician’s “subject” is the physicist’s “object” ; and  that 2. in the history of Western thought since Aristotle, the relation of subject to predicate in logic is homologous to the relation of substance to accident in metaphysics. This entails that in the relation S/P (S as P), S is substantial and objective, whereas P is unsubstantial and subjective; and this is indeed why, in the MWCP, S amounts to Descartes’ res extensa (the object) and P to what the res cogitans (the subject) has in mind about S.
            According to this dualism, reality is S (the object), and P a mere subjective representation of S. Now, this does not hold for mesology. S is an abstraction, whereas concrete reality is the way S exists as something for a certain being I (the interpreter of S as P). The binary relation S-P is an abstraction; concretely, we have a ternary relation S-I-P, meaning that S exists as P for I. For instance, one and the same tuft of grass S exists as food (P) for the cow, but as a shelter (P’) for the beetle, as an obstacle (P’’) for the ant, and so on. Concrete reality is not S, but the relation S-I-P, which implicates at the same time substance, interpretation and non-substance in the process of trajection, which is a process of concrescence, i.e. becoming concrete or “growing together” (cum crescere) in the dynamic coupling of a certain being with its milieu.
            Now, saying that the trajection of reality is a process amounts to saying that reality is historical. It is not an objective R (the Real, or S in itself), but a trajective r (reality in the making, or S as P for I). In this historical process, new beings I, I’, I’’, I’’’ etc. indefinitely re-interpret S as P, P’, P’’, P’’’ and so on. Keeping I implicit for graphic simplicity’s sake, this can be formulated as : r (reality in the making) =  (((S/P)/P’)/P’’)/P’’’… and so on, indefinitely. This is called in mesology a trajective chain.

§ 4. Trajective chains, semiologic chains, von Neumann chains
The above formulation shows that P, in the succession of S/P, (S/P)’, (S/P)’’ etc., is indefinitely placed in the position of S’, S’’, S’’’ etc. relatively to P’, P’’, P’’’ etc. In other words, it does not only become the subject of a new predicate, it is substantialized, or hypostatized.
            This, at first glance, seems irrational: how can nothingness (non-substance) become something substantial?
            Now, this process is analogous to what Roland Barthes, in his Mythologies[13], called “semiologic chains” (chaînes sémiologiques), a process in which historical realities become myths, and in which signs, understood as the relation between the signifier and the signified (signifiant/signifié, or Sã/Sé), become the signifiers of further signifieds. This can be represented exactly in the same way as trajective chains, viz (((Sã/Sé)/ Sé’)/Sé’’)/Sé’’’… and so on.
            This is not only an effect of formal figuration. First, a new signified is nothing else than a new predicate. Accordingly, the transformation of history into myth is homologous to the assumption of S into P, or of substance into non-substance. Second,  the course of history shows plenty of examples in which unsubstantial myths, i.e. mere words, like in the hypostasis of P into S’,  are transformed into quite substantial things, e.g. pyramids in ancient Egyptian religion. An even more striking example is how the Arcadian myth, after three thousand years of progressive substantialization, eventually provoked the present warming of planet Earth, via the multiplication of detached houses associated with the massive use of individual cars[14].  
            In any case, these chains concretely suppose the agency of an interpreter I in the ternary relation S-I-P, or Sã-I-Sé. The binary Saussurian (and Barthesian) relation Sã/Sé is an abstraction. In this respect, the mesological conception of trajective or semiologic chains is akin to the Peircean triadic conception of semiosis, though what Peirce called “the interpretant” does not exactly coincide with what I call here the interpreter, that is a living being, and especially a human being.
            Yet, the interpreter I is never limited to the topos of an individual person or organism; it is always and necessarily medial, i.e. composing such a topos with its corresponding chôra, or milieu. In the case of the human, this milieu is eco-techno-symbolical. In addition to ecosystems, it comprises all sorts of institutions, both technical (e.g. buildings) and symbolic (e.g. languages or religions), indefinitely intertwining into concrete and concrescent reality S/P, and always conditioning an individual person’s thought and supposedly “free” will. In this sense, the milieu (chôra) amounts to what Marx called Produktionverhältnisse, Foucault dispositif, etc.[15]
            The Christian tendency to insist on the individual’s free will (since the absolute dwells in each person’s conscience, as Saint Paul and Saint Augustine said: manes in memoria mea, Domine) made that the West was late in realizing this conditioning, but East of Aden, especially so in Buddhism, it had been known for centuries as prajñapti in sanskrit, sesetsu 施説 in Japanese, etc.[16]    
            From the point of view of mesology, the agency of this conditioning is nothing else than the trajection of reality S/P. It is what is at work in the “/” between S and P, i.e. the as in “S as P”, at any ontological level, from matter to mind via life, and from the res extensa to the res cogitans.
            Such a view is not modern, since it is not dualist. Neither is it premodern, or “prelogical”, as Lévy-Bruhl would have put it. It is transmodern: beyond modern dualism and beyond the principle of the excluded middle (called in French tiers exclu, i.e. “excluded third [term]”), yet basing precisely on what modern science’s dualism has experimentally proven both in biology (as we have seen with Uexküll) and in physics, as we shall see below.
            It is now almost commonly assumed that, in quantum mechanics, measurement affects the system which is measured, as Niels Bohr and Werner Heisenberg, nearly one century ago, have argued in the so-called Copenhagen interpretation. “Interpretation” here is not only what the modern cogito deems reality to be, i.e. theoretically S, but empirically S/P (S as P). Moreover, it is what the physical device of the experiment itself produces, i.e. causing the set of probabilities to collapse into only one reality (in other words, causing Schrödinger’s cat to be either dead or alive, not both). Human interpretation will then follow this first “interpretation”, which makes one and the same particle exist as either a corpuscle or a wave (not both); an “interpretation” which is purely physical, independent from the observer’s personal interpretation, and supposedly objective since it is experimentally proven.
            From a mesological point of view, the experimental device here is nothing else than a dispositif or a sesetsu[17], performing as such a trajection which makes the wave function “collapse into”, or exist as (ek-sist out of the experiment as) either a wave or a corpuscle, i.e. either S/P or S/P’, in the ternary structural moment S-I-P or S-I’-P’ (I or I’ being here the device). This entails that, ontologically, the result of the experiment is not abstractly “objective” (S), but concretely trajective (S/P or S/P’). And for the same reason, successive measurements with successive devices will produce a chain of results which are homologous to a trajective chain (((S/P)/P’)/P’’)/P’’’… and so on. Such chains, in physics, have been called “von Neumann chains”. [18]
§ 5. Chemistry in the concrete milieux of the Ecumene
I am not a chemist, nor an alchemist or even a philosopher of science, but a geographer wondering about the reality of the ecumene, that is the relationship of Humankind with the Earth, or the whole of human milieux. This has led me to advocating mesology in the sense of Uexküll’s Umweltlehre and Watsuji’s fûdoron. In such a view, chemical reactions are what happens between the ontological level of the Planet and that of the Biosphere, i.e. between matter and life, in the form of trajective chains.
            It would first come to mind that these chains, at the ontological level of chemical reactions, may correspond to the trophic chains of the Biosphere. In such a logic, the chemical constituents of a tuft of grass (S), which are assumed as food (P) by a cow, will then be digested so as to become, in the following hypostasis of this trajection, the cow itself (S’).
            In the present state of my reflexion, this equation of metabolism to a trajective chain is hardly more than a metaphor, and it should need much more relevant investigation than I can do as a geographer. Yet, knowing that chemical systems are the very language of nature[19], this hypothesis should make sense. For example, the chemical mimicry of a myrmecophile insect (living among ants in their nest), can be understood both logically and ontologically if we consider it as a trajection in which this insect, because it imitates an ant’s smell, is confused with a true ant and thus is not killed by them; it means indeed that, though abstractly it is not an ant, concretely, that is for these ants, this insect exists as an ant.
            The trajection here is a chemical process in which the smell in question corresponds to different cuticular hydrocarbons, i.e. alkanes and alkenes on the insect’s cuticle. The non-ant insects in this respect seem to follow two strategies : “either they emit a smell similar to that of the ants (chemical mimicry), or they emit little smell, but rapidly cover themselves with the smell of the ants or that of the interior of the nest (chemical camouflage)”.[20] And by doing so, these truly non-ants (S) exist as concrete – if not authentic – ants (S/P). Which, by the way, beyond the principle of the excluded middle, corresponds to the fourth lemma of the tetralemma : both non-A and A, assuming that A stands for Ant and non-A for non-Ant.
            Metabolism for its part has been compared with an “assembly line, in order to describe the enzymatic systems which are responsible for the production of metabolites (…) These machineries (…) are complexes of highly specialized enzymes with complementary functions. Each of the enzymes of that complex is responsible for the chemical transformation of a substratum, the result of which is transmitted to the next enzyme in order to realize the next transformation”.[21]  
            From a mesological point of view, this “assembly line” is nothing else than a trajective chain, in which what in fact is a “transformation” (S/P) is hypostatized into the “substratum”, i.e. the hupokeimenon : subjectum: suppositum S’ of the next trajection (S/P)/P’, and so on, like in the chain (((S/P)/P’)/P’’)/P’’’… etc.
            As can be seen, resulting from a transformation (the trajection of S as P), the reality S/P, though placed in subject position S’ vis-à-vis the next predicate P’, is not a “true” S, but an ersatz of S. It is S’, a hypostasis of S/P, not the original S. Now, this is precisely the process of life, and that of the evolution of life on the Earth, which made possible the trajection of the initial Planet into the Biosphere, then of the latter into the Ecumene.[22]
            In physics, this transformation of an original suppositum corresponds to what d’Espagnat calls “le réel voilé”, which is not a theoretically pure S, i.e. “le Réel” in itself (an sich, as Kant would have put it, and as I like to represent it with a capital R); it is no more than empirical reality, which I represent with a small r in the formula r = S/P.
            By the way, it should be noted that, unfortunately, “le réel voilé” has been rendered in English with “veiled reality”, which – that’s the case to say – veils this distinction between the Real (S) and reality (S/P). Some kind of trajection seems to have occurred during this translation over the not-so-English Channel…       

Palaiseau, 27 June 2017.

Augustin BERQUE (1942 -), a French geographer and orientalist, is a retired director of studies at the École des Hautes Études en Sciences Sociales (EHESS), Paris, where he teached mesology. A member of the Academia europaea, he was in 2009 the first Westerner to receive the Fukuoka Grand Prize for Asian cultures. Among his books, Thinking through landscape, London : Routledge, 2013. E-mail : <>

[1] “Part of biology which studies the relations of milieux and organisms”.  For an overview of mesology in the present sense, i.e. the study of milieux, see my La Mésologie, pourquoi et pour quoi faire?, Nanterre : Presses Universitaires de Paris Ouest, 2014, and  Là, sur les bords de l’Yvette. Dialogues mésologiques, Bastia : éditions Éoliennes, 2017.
[2] Haeckel’s definition of Ökologie, in 1866, was the following : “Wissenschaft von den Beziehungen den Organismen zur umgebenden Aussenwelt (Science of the relations of organisms with the surrounding external world)”.
[3] In the normal East Asian order, family name (Watsuji) first.
[4] Recorded posthumously in his Die Grundbegriffe der Metaphysik (The Founding Concepts of Metaphysics), Frankfurt am Main : Klostermann, 1983.
[5] Both have been translated into English, but the translator of Fûdo completely missed Watsuji’s point. One should rather read the French translation (Fûdo, le milieu humain, Paris : CNRS, 2011).
[6] « (…)  die stillschweigende Voraussetzung, ein Tier könne jemals mit einem Gegenstand in Beziehung treten, falsch ist ». Streifzüge…, op. cit., p. 105 in the Rowohlt edition, Hamburg, 1956. My translation.
[7] En tant que in French.
[8] I introduced this concept in Le Sauvage et l’artifice. Les Japonais devant la nature, Paris : Gallimard, 1986 (English transl. Japan. Nature, artifice and Japanese culture, London : Pilkington Press, 1997).
[9] Commenting Uexküll in his 1929-1930 seminar, Heidegger showed that things do not exist in themselves, but as something (als etwas).
[10] In Le Sauvage…, op. cit.
[11] More details on this in my Écoumène: introduction à l’étude des milieux humains (Ecumene: introduction to the study of human milieux), Paris: Belin, 2000, 2008, chapter I.
[12] YAMAUCHI Tokuryû, Rogosu to renma, Tokyo : Iwanami, 1974. I discuss it from the point of view of mesology in my Poétique de la Terre. Histoire naturelle et histoire humaine, essai de mésologie (Poetics of the Earth. Natural history and human history, an essay in mesology), Paris : Belin, 2014. My translation of Rogosu to renma into French is forthcoming at CNRS éditions, Paris.
[13] Paris : Seuil, 1957.
[14] I have detailed this case in Histoire de l’habitat ideal (History of the ideal abode), Paris : Le Félin, 2010.
[15] On this theme, see Giorgio AGAMBEN, Qu’est-ce qu’un dispositif ?, Paris : Payot & Rivage, 2007 (Che cos’è un dispositivo?, 2006).
[16] On this theme, see Yamauchi, op. cit.
[17] Or rather shisetsu in the ordinary pronunciation of the same施説, which in modern Japanese means equipment, facilities, institution, establishment.
[18] For more details, see Bernard d’ESPAGNAT, Traité de physique et de philosophie (Treatise of physics and philosophy), Paris, Fayard, 2002, p. 128 : « Supposons, par exemple, que l’on ait affaire à un système composé constitué d’un microsystème S (une particule, un atome…), d’un instrument de mesure I réglé de telle sorte que la position L de son aiguille indicatrice soit corrélée avec la valeur G d’une grandeur appartenant à S, d’un deuxième instrument I’ réglé de façon que son aiguille L’ prenne une position corrélée avec celle de L, d’un troisième instrument I’’ programmé de même à l’égard de L’, etc., jusqu’à, finalement, un instrument If dont on observe l’aiguille indicatrice. En physique classique l’observation en question nous informerait, indirectement, de la valeur que la grandeur G possédait, avant toute mesure, sur S. En physique quantique le problème est de calculer avec quelle probabilité on aura l’impression d’indirectement percevoir, gâce à If , telle ou telle valeur de G. Le formalisme montre que, pour ce faire, on peut, en principe, procéder de diverses manières équivalentes. On peut, par exemple, incorporer les paramètres de I et de I’ dans la fonction d’onde et traiter le grand système macroscopique composé de I’’, … If comme s’il était le ‘sujet percevant’. Ou on peut faire de même en laissant I du ‘côté quantique’ – du côté de la fonction d’onde – et en mettant I’ du côté du ‘sujet percevant’. Ou enfin, et c’est le plus simple, on peut ne mettre que S du côté quantique, ce qui revient à ne pas du tout faire intervenir dans la fonction d’onde initiale les paramètres des instruments. Tout se passe alors comme si le grand système macroscopique composé des I, I’, I’’ … If constituait le ‘sujet percevant’. Cette équivalence de principe entre divers procédés de calcul a été pour la première fois signalée par von Neumann et la suite des I, I’, I’’ … If se nomme pour cette raison ‘chaîne de von Neumann’ ».
[19] As is clearly shown by Martine HOSSAERT-McKEY et al., Écologie chimique. Le langage de la nature (Ecological chemistry. The language of nature), Paris: Le Cherche Midi, 2012.
[20] Op. cit., p. 69.
[21] Op. cit., p. 144.
[22] I have detailed this interpretation in Poétique de la Terre, op. cit.