“Signs Becoming Signs”
PREFACE
1. While rather critical of Peirce, Richard Rorty (1979) categorizes leading contemporary thinkers such as Dewey, Heidegger, and Wittgenstein as “antifoundationalists.” Granted. The search for “foundations” in terms of understanding the world sub specie aeternitatis has been at long last abandoned. However, regarding, for example, Wittgenstein’s case, the general verdict has it that groundings are the riverbed of language within “forms of life” rather than the bedrock of ultimate “foundations” in the classical sense. Language is a contextualized human activity constantly shifting with the ebbs and flows of the entire community of speakers. Add to this Peirce’s notions of evolution and self-correction, and, though they evoke a nineteenth-century image of cumulative progress—after all, Peirce could not but exist within his own community, albeit at the margins—we have the idea that knowledge does not entail grasping some inner core but knowing how to think and act in a certain way. Peirce, it hardly needs stating, deserves a serious look in relation to Wittgenstein and other leading twentieth-century philosophers.
1. ASYMPTOTICALLY GETTING THERE
1. The “slash” between the “semiotically real” and the “actually real” will be highlighted throughout this inquiry. Although Peirce never used this pair of terms in the manner I do, the distinction between them, I would suggest, is in general commensurate with his doctrine of signs. They also fall in line with the thought of a large body of contemporary scientists and philosophers of science, as Einstein’s (1950:98) words bear witness: “The sense experiences are the given subject matter. But the theory that shall interpret them is man-made. It is the result of an extremely laborious process of adaptation: hypothetical, never completely final, always subject to question and doubt.” I will allude repeatedly to the “semiotically real” when referring to a particular world picture in common with a given individual and his community of knowers. The “actually real,” in contrast, is the focus of the final or ultimate interpretant, which must remain inaccessible, as Peirce argued regarding his rather controversial “convergence theory” of truth, to a given individual or community.
2. This might appear to be a “coherence theory” of truth, but it is not. Rather, the relations between sentences determining their truth or falsity depend principally on conventions and contexts, as we shall note below.
3. I do not consider metaphysical statements and paradoxes to be meaningless or absurd as they are in the logical positivist framework. Rather, what is regarded as meaningless or absurd depends upon a particular time and place. Within another context, an apparent absurdity could well become the most profound of truths.
4. Such fictive constructs might imply, upon functioning as scientific metaphors or models (Black 1962; Hesse 1966), a potential for attaining increasing determinateness when successive “cuts” come into existence in the book of assertions, since the false propositions—inadequate aspects of the model—may in the future be successively canceled and the true ones retained (Roberts 1973:31-40).
5. For problems inherent in attempting to conceive of a world of higher dimensions than one’s own, see Abbott’s (1952) work on “Flatland,” Burger’s (1968) on “Sphereland,” and Rucker’s (1977) insightful comments on both.
6. See Rescher (1978) for more detail. It is worthy of note that Peirce’s notion of “cumulative knowledge,” via his “convergence theory of truth,” is relevant to the concept of “evolutionary epistemology,” which in recent years has been subject to criticism (Radnitsky and Bartley 1987).
7. Admittedly, Peirce’s use of the term consciousness is rather debatable, especially in light of recent work by Derrida (1973, 1974), among others. However, see Merrell (1985a) for a partial vindication of Peirce in this regard.
8. Spencer-Brown’s mark is comparable to Niels Bohr’s theoretical distinction, a partly arbitrary “cut,” between the object and the physicist’s instrument used to detect that object. Murdoch (1987:97-103) provides an enlightening discussion of the contradiction this “cut” creates between the classical (macroscopic) treatment of the instrument and the quantum-mechanical (microscopic) treatment of the object.
9. This “block” universe interpretation of relativity theory is admittedly controversial. See, for example, Capek’s (1981) counterargument in the same volume.
10. Time-dependent considerations of truth and meaning variance go against the grain of most modem logic, an obvious exception being modal logic. Modern logic, especially since the early Wittgenstein, Russell, and Frege, is that of an omniscient and timeless being, the logician, for whom the only possible world is the actual one. Peirce, on the other hand, considered logic, like Aristotelian and medieval time-dependent logic, to be a normative science. He believed in real possibilities and necessities of a particular person’s knowledge at a particular time and place (CP:5.435ff., 6.367). Subsequently, he made a strong plea, though he himself never carried out the project, for a formal logic of time dimensions (Prior 1957).
11. In the 1930s, the terms chronon and hodon were invented for designating the atoms of time and space respectively. The value of the chronon was computed at between 10-21 and 10-24 seconds, and that of the hodon at 10-13 cm. Concerning these values, Capek (1961:231) suggests that
for all practical purposes, and considered macroscopically, space and time are continuous: the duration of chronons is so insignificant that they may safely be equated with durationless instants; similarly, the difference between mathematical points and spatial regions of the radius 10-13 cm is entirely negligible on our macroscopic scale.
This, however, neither clears up the muddle between Hartshorne and Peirce nor makes the difference between the classical continuous space and time and its modem atomistic counterparts less radical. The last word on this, as on other debatable issues concerning quantum theory and relativity, is still up for grabs.
12. Regarding mathematics, interestingly enough, Einstein often mused that as far as propositions refer to “reality,” they are not certain, and as far as they are certain, they do not refer to “reality.” Eugene Wigner’s (1969:138) words on the unintelligibility of such “reference” are especially revealing:
The miracle of the appropriateness of the language of mathematics for the formulation of the laws of physics is a wonderful gift which we neither understand nor deserve. We should be grateful for it and hope that it will remain valid in future research and that it will extend, for better or for worse, to our pleasure even though perhaps also to our bafflement, to wide branches of learning.
13. W. V. O. Quine’s (1960, 1969; also Gibson 1982, 1988; Romanos 1984) radical intranslatability and inscrutability of reference theses are germane to the issue at hand, though time and space do not allow for expatiation on this theme here.
14. For a sympathetic or mildly critical view of the Sapir-Whorf hypothesis, to which I shall return in chapter 7, see Black (1962:244-58), Penn (1972), Carroll (1964), and an interesting study by Heynick (1983). For a critical view, see Fishman (1980, 1982). If the Sapir-Whorf hypothesis has its day, language and culture do all the work; the individual has hardly any say-so on the matter. “Reality,” in the Sapir-Whorf conception, lends itself to a potential infinity of different classifications, and one learns to divide it up according to the implicit mandates of one’s culture, but more particularly of one’s language. Eleanor Rosch (1974, 1975a, 1975b, 1977, 1978), among others, places in question the existence of such linguistically fixed categories. They tend to become fuzzy and melt into one another, she contends, thus allowing for variations. And rather than arbitrary, they are by and large motivated.
Peirce would likely defer slightly toward Rosch’s notion in his admission that “reality,” as Secondness, does motivate: it forces itself upon the observer, bringing about interpretations which are at times contrary to her wishes. However, to reiterate, when considering the entire linguistic community, there will never be absolute agreement on the nature of “reality,” hence a degree of arbitrariness, and therefore linguistic determinism, will inevitably prevail in any and all classificatory systems. Categories are not simply mental constructs in the head, though a lively give-and-take between the internally “real” and the externally “real” can and does lead to a victory—however ephemeral—of the latter over the former. In short, categories both produce and are the product of one’s “semiotically real world,” which sets the limits of the modus operandi one can employ in a continued attempt to cope with one’s surroundings.
15. As Born (in Pais 1986:256) once put it: “It is necessary to drop completely the physical pictures of Schrödinger which aim at a revitalization of the classical continuum theory, to return only the formalism and to fill that with a new physical content.”
16. One of the chief problems is, as suggested above, linguistic—and here perhaps Sapir-Whorf take center stage once again. In brief, the language of relativity portrays a continuous view of matter—the field concept—while the language of quantum theory propagates the atomistic view. The two languages have thus far proved to be by and large formally contradictory yet complementary—even though embracing this principle still tends to produce a profound cognitive dissonance.
17. In other words, the statement “Lemons are sour” can easily be put to the test by subjecting a lemon to the taste buds. But determinate meaning of the proposition is not to be found so easily. The range of all conceivable ramifications of this initial test must be exhausted, which include the chemical composition of citric acid, the pH factor of lemon juice, the ratio of H+ ions to OH- ions, the action of acidity on nerve endings on the tongue, impulses traveling from the taste buds to the brain, human cultural values regarding acidity in contrast to bitterness and sweetness, and so on. Each of these actual tests “out there” in the physical world implies, or directly involves, a subsequent proposition regarding what would be the case if certain conditions were to exist. The meaning of “Lemons are sour” cannot be exhausted short of an infinity of tests and generation of their respective propositions within pragmatic contexts.
Such proposition formulation and test procedures, however, are not always the product of intentional sign activity. For example, suppose a drug, x, reputed to help prevent cholesterol buildup, is released to across-the-counter pharmaceutical retailers. First, the “meaning” (i.e., use, which pragmatically endows the drug-as-sign with an interpretant, be it legitimate or not) of the drug is put to the test on the competitive market, and true to predictions, heart problems are lowered. Over time, consequently, habit is gradually developed on the part of the consumers. The drug’s “meaning” becomes embedded; it is now purchased and used rather mindlessly (like a red traffic light, which, if disregarded, might subject one’s car to a collision; therefore one stops the car by habit). For Peirce, what a sign “means” is ultimately determined by what habit is involved, which can be a bane and a boon. Without habit, community agreement and hence convention would be well-nigh impossible, but habit can also stultify, compelling one to mindless activity such that one is no longer conscious of one’s motives and the consequences of one’s actions (see Nesher 1983; Langer 1989).
In this sense, to speak of “semiotic reality” is to speak of relationships between a language—or other sign systems—and the “real.” In the sentence “‘Photon’ refers to a discrete packet of energy,” there is discrimination between the sign being discussed and the signs used to discuss it. “Photon” signifies something presumably in the “real” world called a “discrete packet of energy.” But notice that reference consists of a relation between the sign “photon” and the signs “packets of energy,” not directly between signs and things “out there” that are nonsigns. In other words, the sign “photon” refers to its “semiotic object” in the “semiotically real.” I use this particular example to stress the point that the signs a physicist uses to qualify his observations and the signs of his theoretical language, in concert, are not some pristine mirror of the world. They are symbolic, and they take on meaning and value in the interconnected system of interpretation. No observation has any meaning at all until it is interpreted by theory, convention, or habit. “Photons” are “semiotically real” not because they establish some mystical linkage between signs and the “real” but because physicists can speak about them, and their expectations regarding them can be fulfilled by means of their shared assumptions, conventions, and understandings: “photons” are the joint opinion of the community in Peirce’s conception.
18. See Heelan (1970a, 1970b, 1971, 1983) for the context-dependency of “quantum logic”; also Merrell (1985a, 1990), where I develop this theme with respect to the semiotic perspective.
19. I should point out at this juncture that I by no means wish to imply that Peirce is everything to everybody. He was a helpless child of his times in certain respects, as we all are. Besides, modern thought can be traced back to a plethora of “sources.” Einstein can be found in Spinoza, in Parmenides, in the Kabbala; Bohr can be found in Oriental thought, as can Schopenhauer from one angle, Whitehead from another, and Peirce from yet another. The list is inexhaustible. Holton’s (1973) “themata” and Jung’s “archetypes”—alluded to by physicist J. M. Jauch (1973)—are relevant here. If contemporary physics has picked up two contradictory “themata,” one in quantum theory and another in relativity, I find it not at all surprising that Peirce’s thought contains both, a contradiction he labored to resolve up to the end of his life.
20. The term representation, as it is used in logical positivist jargon, is rapidly becoming outmoded. I prefer to use signification but have remained with Peirce’s term when it becomes necessary within the context of his writing.
21. Einstein’s words are apropos in this respect:
A human being is part of the whole, called by us “Universe”; a point limited in time and space. He experiences himself, his thoughts and feelings as something separated from the rest—a kind of optical delusion of his consciousness. This delusion is a kind of prison for us, restricting us to our personal desires and to affection for a few persons nearest us. Our task must be to free ourselves from this prison by widening our circle of compassion to embrace all living creatures and the whole of nature in its beauty. Nobody is able to achieve this completely but the striving for such achievement is, in itself, a part of the liberation and a foundation for inner secuity. (In Herbert 1985:250)
2. BOHM’S TROPOLOGY AND PEIRCE’S TYPOLOGY
1. For popularizations of the “new physics,” see Capra (1975), Talbot (1980), and Zukav (1979)—dubbed vulgarizations by some (Bernstein 1982)—and for two captivating syntheses of relativity and the quantum world for laypersons, see Sachs (1988) and Davies (1988).
2. This phenomenon is related to the double-slit experiment, mentioned above, and especially the EPR experiment—so named after Einstein, Podolsky, and Rosendesigned in an effort to dislodge the Bohr-Copenhagen interpretation of quantum mechanics. Bohm is reported to have received his inspiration for “hidden variables” while musing over the nonlocalization implied by the EPR experiment (Hiley and Peat 1987:12-13).
3. Although Wheeler (1980a, 1982, 1984), I must add, is adamantly against “faster than light travel” insofar as it is used to account for ESP and other comparable phenomena. I raise this point here, since I will briefly discuss Wheeler’s view in chapters 5 and 6.
4. Bohm’s rather Spinozist language of the implicate (enfolded within itself) and the explicate (unfolded) is, interestingly enough, also found in some of the early writings of Gilles Deleuze (especially 1968, 1969)—whose view, along with that of his collaborator, Felix Guattari, will be discussed in chapters 3 and 4. Deleuze often uses the terms to distinguish between inanimate and animate matter. The first is stable, the second metastable, involving a process of individuation in which difference differentiates itself, a virtuality constantly becoming actuality without ever becoming actualized, stable being.
5. However, Capek (1961, 1981), as well as Prigogine (1983:185-219)—to be discussed in chapter 5—argues against the artificiality of this static view, contending that time is not simply another dimension of space.
6. Bohm, though Hartshorne does not mention his name, is the chief proponent of the “hidden variables” thesis, which is grounded in the argument that the Copenhagen interpretation is incomplete, especially in light of its uncertainty principle (see Herbert 1985).
7. For Bohm, at the lowermost level of the holomovement, both the particle and the wave are implied, which is at once more primitive and more profound, Bohm proponents argue, than the Copenhagen interpretation (see Hiley and Peat 1987).
8. This example bears on Hilbert’s “Hotel Paradox.” The hotel, capable of accommodating an infinity of guests in its equal number of rooms, was filled to capacity on a particular day. The arrival of a new guest presented no problem for the manager. He simply had the occupant of room one move to room two, the occupant of that room move to room three, and so on. Then the new guest was placed in room one. By the same operation, an unlimited stream of new arrivals could be given a room. The hotel could never be filled! But, quite paradoxically, it was filled, since it already had an infinity of occupants.
9. I will further treat the topic of consciousness in light of speculations in contemporary physics in chapter 7.
10. The move during the 1970s was toward the idea that memories are distributed over the cortex. This concept has occasionally surfaced over the years. Karl Lashley (1951) speculated on the possibility of standing wave patterns set up by the interference of volleys of neural firing, much in the sense of Fourier transforms. Lashley did not specify, however, how such wave patterns could distribute a retrievable memory. D. O. Hebb (1949) agrees with Lashley, though he believes Lashley went too far in his denial that the same cells need to be excited to arouse the same perception. Pribram (1971, 1980, 1981, 1982), like Lashley, postulates that memories are laid down as interference patterns. Unlike Lashley, however, he has developed a specific model, the holograph, to illustrate his thesis. In brief, Pribram places the locus of neural interaction at junctions (synapses) between neurons. Neural junctions are activated from a “wavefront” which can simultaneously interfere with other such “wavefronts” in the same ensemble but originating in other parts of the nervous system. This interference constitutes the basis of his notion of information processing, or holography.
11. In this respect, and following Bohm, F. David Peat (1987:174) writes that
our thoughts are the explicit forms thrown up by the underlying movements of the explicate order of mind. Like the vortex of a river, . . . thoughts have no absolute, independent existence of their own but are constantly being supported by the underlying process of their ground. Ultimately this movement of mind merges into that of matter so that the two should not be considered as dual aspects of nature but as arising out of the same underlying ground. . . . In this way it appears that individual minds have a common or collective origin that has something in common with that of matter. . . . [T]herefore, mind is able to act upon mind, and mind and matter exert an influence one on the other. But this should be thought of not as some form of causal interaction since individual minds, and mind and matter, are not fundamentally separate but are simply the explicate forms that emerge out of a common, generative order. (See also Bohm 1987a:72-99)
In a comparable vein, Heisenberg (1958a:81) tells us, “Natural science does not simply describe and explain nature; it is a part of the interplay between nature and ourselves; it describes nature as exposed to our method of questioning.”
12. I must mention that Pribram is now somewhat out of fashion, especially since his hypothesis has not yet been effectively validated (see Briggs and Peat 1989)—though one can only speculate on what he will offer in his soon-to-appear book. Besides, Bohm (1987b) has expressed reservations regarding certain limitations of the hologram model.
13. Interestingly enough, in this light, Peirce points out that in many cases of brain damage, recovery leads to a condition in which “other parts of the brain are made to do the work, after a fashion, with perhaps other parts of the body” (CP:7.376), which implies that the mind itself has not been impaired.
14. Regarding such “memory,” see the thesis developed by Ilya Prigogine (Nicolis and Prigogine 1989), which will be the brief focus of attention in chapter 7.
15. Unlike Bohm, and the majority of physicists, a small group of avant-garde mavericks, including Musés (1977), Toben (1975), Walker (1970), Walker and Herbert (1977), Wigner (1967), and Wolf (1984), propose that “mind” or “consciousness” should be injected into quantum equations. They have produced elaborate mathematical explanations in their attempt to demonstrate the crucial role of consciousness at the microphysical level, which, above all else, has enticed a few “pop” physicists into mysticism and parapsychology. Other physicists, after putting forth wild conjectures, then inject a note of sobriety into their discourse, somewhat typical of which are the remarks of Geoffrey Chew (1968), author of the “bootstrap” hypothesis: “Carried to its logical extreme, the bootstrap conjecture implies that the existence of consciousness, along with all other aspects of nature, is necessary for self-consistency of the whole. Such a notion, although not obviously nonsensical, is patently unscientific.”
Wheeler, as I mentioned in note 3 of this chapter, has been stridently opposed to such idealistic interpretations of quantum physics. Nonetheless, Skolimowski (1987:75), an ardent follower of Wheeler, observes that the philosophy of Karl Popper “appears to offer less and less for the newly emerging epistemological problems of quantum physics, especially for the conception of reality as co-extensive with the mind. Popper is an epistemological realist of the traditional kind, as he believes in a firm reality out there, a view the New Physics does not support” (see also Skolimowski 1986).
16. Of interest in this regard is the work of Kolers (1972), and the use of his empirical studies in a theory of aesthetic perception by Goodman (1978).
17. In like fashion, Peirce (CP:5.475) observes that the performance of a piece of music is a composite sign, a sort of holosign.
18. The river metaphor is a commonplace. It is also evoked by Bohm (1987b:34), and by physicist James Jeans, whose words bear citing:
We may picture the world of reality as a deep-flowing stream; the world of appearance is its surface, below which we cannot see. Events deep down in the stream throw up bubbles and eddies on to the surface of the stream. These are the transfers of energy and radiation of our common life, which affect our sense and so activate our minds; below these lie deep waters which we can only know by inference. These bubbles and eddies show atomicity, but we know of no corresponding atomicity in the currents below. (In Commins and Linscott 1947:381)
3. THE TENUOUS “REALITY” OF SIGNS
1. Peirce’s habit bears certain similarity to Rupert Sheldrake’s (1988a) “morphic fields,” though the analogy is easily stretched to its breaking point (in Sheldrake [1988b] he relates his “morphic field” idea directly to Peirce’s habit). Sheldrake’s controversial theory proposes that fields of information influence structures not only of living organisms but of inorganic matter as well. All matter has an associated field of memory whose function is to guide the formation of structures. Molecules, crystals, plants, and animals all develop according to their proper form. If there were no fields of information and memory, there would simply be too many alternatives and contingencies for nature to exhibit the type of unity in diversity that is seen in the structures of matter and living things. The first time a new molecule is generated or a crystal grows, it follows a blind path down the valleys and hollows of its energy landscape, determined by the various local forces that operate on it. But this process also gives rise to a “morphic field,” which is a kind of memory of the material processes involved. The next time this process occurs, it has the advantage, Sheldrake contends, of being guided by information from the morphic field. As repetitions accumulate, the field builds in strength and becomes more active in guiding the process. The effect is similar to Waddington’s (1957) “chreodic paths,” which are like skiers going down a slope. The first skier tentatively finds a path between trees, rocks, and broken terrain, which facilitates successive skiers, and finally a well-worn path develops, compelling them to follow certain pathways of least resistance. In a word, habit has exercised its force.
2. This was also the case during the structuralist euphoria of the mid-1960s with the appearance of relatively formal, parsimonious models for folklore and literary texts from Barthes (1966), Bremond (1966), Greimas (1966), Todorov (1966), et al.
3. See Gleick (1987) and Briggs and Peat (1989) for a layperson’s account. I introduce the “physics of chaos” here for the sake of the following discussion. I turn attention specifically to nonlinearity and order from chaos in more detail in chapter 5 with respect to the holistic implications of Peirce’s semiotics.
4. For the infatuation among scientists with harmony, simplicity, and elegant formulations, see Zee (1986) and Augros and Stanciu (1984), Wechsler (1978); also Girard (1984), who suggests that perhaps there is an inherent need to interpret disorder as order in myths.
5. I must guard against a potential misunderstanding of the terms disorder, chaos, and randomness, which, in their vulgar interpretation, do not escape ambiguity and even inconsistency, and in their most adequate usage inevitably remain fuzzy. In the first place, “randomness” cannot be determined. If a series is computed to the nth digit and no order is discovered, we still cannot be certain that the series is random, for there is no knowing whether n + 1 will evince some sort of order. In the second place, any tenable concept of “disorder” and “chaos” must be, in my estimation, subjective—it is we who perceive a meaning or pattern in a particular arrangement of symbols, the map of a city, a work of art, or a set of playing cards.
To the question, “Is a tropical forest ordered or disordered?” the answer, in this view, must be that there is an ecological process best qualified as organized complexity, though for a lonely human, unversed in ecological systems and wandering about on the jungle floor, disorder is sure to prevail. There is organization, though it appears disorderly, yet entropy is on the decrease, given the perpetual swim of all living systems against the inorganic stream toward entropy increase stipulated by the Second Law of Thermodynamics. On the other hand, an exothermic reaction leading to the crystallization of ions out of their aqueous solution is entropic, though it would appear to the nonchemist that the regular repetition of growth of units in the symmetrical lattice is orderly. When a gas expands from one vessel to another when the valve between the two is opened, the system becomes more uniformly distributed and symmetrical, hence it might appear ordered, though entropy is on the upswing (Denbigh and Denbigh 1985). In short, entropy cannot be appropriately related to disorder, in the subjective use of the term, which bears witness to the inevitable rift between mathematical language and our experienced world, between the “real” and a given “semiotically real” world.
6. I should mention at this point that Hjelmslev (1961), elaborating on somewhat of a linguistic counterpart to Bohm’s quantum universe, distinguishes between unformed (implicate) matter and the substance shaped (explicated) by the forms of expression and content. Deleuze and Guattari (1987:chap. 2), in parallel fashion, appropriate the Spinozistic concepts of implication and explication, envelopment and development for their own argument.
7. Deleuze and Guattari’s “schizophrenic” must be taken literally as well as figuratively: literally insofar as this is the way society is, and figuratively insofar as the concept applies to all signs (minds as signs in contiguity with other signs).
8. See also Hofstadter (1979), who provides an extensive discussion along these lines regarding an ant colony, and Edward O Wilson (1971), for his work with ant colonies as complex organisms. This capacity of insect societies for organizing themselves toward a common goal is perhaps best illustrated in the work of Austrian Nobel laureate Karl von Frisch, who reports in Animal Architecture (1974) the case of a termite nest which had been covered by a plastic tent, resulting in diminished ventilation. Within forty-eight hours the insects had developed additional structures of a novel design which restored the old ventilation rate under the changed conditions. Obviously, information is transferred at the macrobranch of the evolution of life in terms of functions which may evolve independently of the exchange relations with the environment.
9. This does not contradict the loss of control to which I alluded in the last lines of the previous section. But especially in hierarchical living systems, as we shall observe below in a discussion of Prigogine, control brought about by dissipative structures can lead to order out of unpredictable and otherwise uncontrollable chaos.
10. In another way of putting it, Firstness as abduction is a possibility that presents itself for further speculation, inquiry, and thought. An abduction can then be linked to Secondness, induction, which is cumulative, becoming progressively more complex as the speaker’s memory bank increases in size. Finally, the abduction-induction process gives way to Thirdness, deduction, which leads to the conclusion in what appears to be this relatively simple case of “it rains.”
11. In his argument against the mirror of nature metaphor, Rorty might well have evoked Quine’s (1969:26-86) “myth of a museum” in his own effort to debunk the positivist’s faith in reference and univocal meaning. According to this “myth,” exhibits are meanings and words are labels, and to switch languages is merely to change labels, meanings remaining invariant. Quine offers his celebrated dilemma of a field linguist attempting to translate a native expression, gavagai, as “rabbit,” “undetached rabbit parts,” or “a particular temporal stage in the rabbit’s development.” If the linguist does not know a single word in the native tongue, pointing, gestures, or any other mode of communication will not suffice for tranferring gavagai into the English language. There is simply no infallible empirical clue allowing us to determine reference or meaning in this and all such cases.
Historically, extension has been supposedly up-front and stalwart, intension uncertain and anemic. Now, Quine claims, both are inscrutable. Of course, within the parochial confines of our own language, extension continues to appear more reliable than intension, but this is not necessarily so: the empirical determinacy of extension, through time, cannot even be guaranteed within our own language. Thus for Quine, and more recently from Derrida’s (1973) vantage, there is no experience that can provide data with which a statement can be absolutely verified. There are no data because there is no absolute immediacy or presence. This renders the meaning of the most basic of statements such as “That raven is black” simply undecidable, for they do not refer instantaneously to some independent empirical entity in the “real” world. In this view, we can do no more than gather up, mediately along the stream of time, an unruly collection of unfixed meanings and hope for the best. Reference is ultimately imaginary, or at most it is a linguistic act.
12. See Merrell (1982, 1983, 1985b, 1990) for further development of embedded and automatized sign processing and interpretation, which draws in large part from Polanyi’s (1958, 1966) focal and subsidiary awareness as a component of tacit knowing.
13. Although, to repeat, relativity and quantum theory have not been effectively united. Moreover, quantum theory remains mired in controversy; many of the problems have never been resolved to the general satisfaction of the scientific community.
14. For this distinction between intellectual and nonintellectual and conscious and nonconscious habit, see Nesher (1983) and Merrell (1990).
15. See Marty (1982) and Tursman (1987)—also Gorlée (1989)—for development of the topic of sign generacy as opposed to de-generacy, and use of the terms in this context.
4. SIGNS BECOMING MIND BECOMING SIGNS
1. See Peirce (CP:2.227-308), Tursman (1987:chap. 2), and Merrell (1990:chap. 2) for discussion of the sign types indicated by the integers 1, 2, and 3, which stand for Firstness, Secondness, and Thirdness. For the significance of o, which precedes all signness, see Baer (1988).
2. This occurs by logical multiplication to yield o, as in non-Boolean, nondistribu-tive, context-dependent logic. Hence the one-way arrow from o to 1. And hence the radical asymmetry it implies. See Heelan (1970a, 1970b, 1971, 1983), also Merrell (1990), where I have written more extensively on this nondistributive, non-Boolean model.
3. Regarding the following pages, see Roberts (1973) for further discussion; also Merrell (1982) for development of a comparable theme regarding what I label “bound-aried spaces.”
4. However, Cull and Frank (1979) argue against the “arbitrariness” of Spencer-Brownian marks, a critique that could also be leveled against existential graphs.
5. Rucker (1977:120) observes that Borges’s “refutation of time”—to which I shall turn in chapter 7—bears similarity to Gödel’s (1949) argument, via Einsteinian relativity, for the unreality of time and change.
6. This bears on the notion that a degree of conceptual sloppiness is a necessary feature of empirical concepts (see, in Peirce studies, Swineburne [1969], and in philosophy of science, Feyerabend [1975]). Significantly, in this light, E. H. Hutten (1956:33) observes that “the demand for consistency [nonvagueness] is made by the logical law of contradiction—not both a and non-a—and completeness [generality] by the law of excluded middle—either a or non-a.”
7. For example, Skagestad (1989; also 1983) effectively argues that a degree of fuzziness of concepts is necessary. Incommensurability is thus overcome by vagueness of all signs. If sign A and sign B were absolutely precise, then they would be absolutely different. Consequently, there could be no commensurability between them, and hence no possibility of moving from A to B (the Zeno problem once again).
8. Quite significantly, in this respect, Peirce (CP:6.172) tells us that “true generality is . . . nothing but a rudimentary form of true continuity. Continuity is nothing but perfect generality of a law of relationship.”
9. This conception of differences is not Saussurean. I repeat, similarities and differences are more akin to Bateson’s (1972) words: differences that make a difference. Peircean semiotic differences also bear comparison with Deleuze (1968) on “difference and repetition.” Deleuze’s Nietzschean-inspired concept of differences, rather than semiological signs as arbitrary entities referring to other signs, consists of differences based on physical forces (or Secondness). They are the product of a system of forces of which signs are symptoms; the study of signs is more properly a symptomatology than Saussurean semiology (see Bogue 1989:55-80).
10. Since Sheffer’s “stroke” (and Peirce’s formulation) implies incompatibility or inconsistency in the most fundamental sense, Whitehead (1938:52) tells us that it provides for “the whole movement of logic.” And E. H. Hutten (1962:178) remarks that “it is the very essence of rationality to abolish contradictions; but logic—being the most rational thing in the world—is generated by contradiction.”
11. I must point out that Rescher and Brandom (1979) give logically rigorous definitions to their terms, while I use them “rhetorically,” for the purpose of generating discourse in natural rather than formal language. See Murdoch (1987:54-58) for the relation between “superposition” as I use the term and its use in the quantum theoretical formulation.
12. According to Deleuze and Guattari, everything is a machine, a part coupled to a second part, which in turn is coupled to a third part, and so on, in a binary connective synthesis forming chains of machines through which flows or fluxes pass. Every machine “is related to a continual material flow . . . that it cuts into,” and “each associative flow must be seen as ideal, an endless flux” (Deleuze and Guattari 1983:43-44). These flows constitute a universal continuum of unceasing production. Regarding human machines, there is a flow of milk between a breast machine and a mouth machine, or a flow of words between a mouth machine and an ear machine. Regarding all machines, there are incessant flows of energy and information. Technical machines differ from human “desiring machines” in that they combine dependent parts into a whole which either functions efficiently or breaks down, while desiring machines are a heterogeneous concoction of chiefly independent parts which work only under conditions of chaos, dissipative structures, and bifurcations.
13. Further, in regard to figure 5, 1 is microscopic. This is the domain of consciousness, but it is not (yet) consciousness of such and such. It is feeling, quality. It presents the possibility of abductions as a result of perturbations in the experienced inner or outer “semiotically real” world. This experienced world, the world of immediate consciousness—without there (yet) being consciousness of—is analogous to Popper’s (1972, 1974) World 2. Category 2 in figure 5 is molecular, the realm of individuals. With respect to language, it consists of sounds in the air or marks on paper without their (yet) having any meaning for someone in some respect or capacity. Inductive inferences arising from abductions are made possible in the counterpart to Popper’s World 1 or physical “reality.” World 1 is the “reality” of books as black marks on paper, mere physical existence. Only when read, that is, put to use, do they become actualized as signs, as something for somebody in some respect or capacity. Otherwise, they are possible signs. As quality, they are the immediacy of signness (World 2) without yet having gained entrance into any specific consciousness. Category 3 is molar, incorporating aggregates of individuals from 2 made possible by 1. This is the realm of would be’s, of deductive inferences, of signs about other signs and signs countering other signs, which, in their highest form, are the equivalent of Popper’s counterarguments in World 3. In sum, 1 as microscopic is pure energy, 2 is matter, and 3 is meaning, which corresponds effectively to Peirce’s Firstness, Secondness, and Thirdness (see also Bohm 1987a:72-99).
14. Significantly, complex (imaginary) numbers, incorporated into figure 7, are employed in the description of superpositions, and the two dimensions of the complex number system are intimately related to the three dimensions of space. In this sense the complex plane becomes the fourth dimension of imaginary time in the space-time continuum (see Penrose 1977, 1987).
15. Recall also note 13 of this chapter. I must hasten to point out, though briefly, that Popper’s thought departs from that of Peirce insofar as (1) he propagates a variation of the Cartesian brain/mind split (Popper and Eccles 1977), (2) his “objective knowledge” (Popper 1972) is devoid of the subjective component in Peirce’s “objective idealism,” (3) his not-so-naive realism (Popper 1982), though predicated on a Peirce-like indeterminism, is closer to the empirical tenets of logical positivism—of course, Popper would deny this—than Peirce. The fact remains, however, that in many respects there are striking affinities between Popper and Peirce (Freeman and Skolimowski 1974; Freeman 1983).
16. This concept has surfaced repeatedly in recent times. To cite merely a few instances: “Man ultimately finds nothing more in things than he himself has laid in them—this process of finding again is science, the actual process of laying a meaning in things, is art, religion, love, pride” (Nietzsche 1913:103). Rorty (1982:xlii) observes that Sartre, Dewey, Foucault, and James, in addition to Nietzsche, collectively conclude that “there is nothing deep down inside us except what we have put there ourselves.” One of the most telling and straightforward comments on this theme is found in the work of Eddington (1978:200-201):
All through the physical world runs that unknown content, which must surely be the stuff of our consciousness. . . . Where science has progressed the farthest, the mind has but regained from nature that which the mind has put into nature.
We have found a strange foot-print on the shores of the unknown. We have devised profound theories, one after another, to account for its origin. At last, we have succeeded in reconstructing the creature that made the foot-print. And Lo! it is our own.
All of which reminds one of Borges’s (1964b:93) man who set out to construct a map of the world: “For many years he filled a space with images of provinces, kingdoms, mountains, shores, ships, islands, fish, rooms, instruments, stars, horses, and persons. Shortly before dying he discovered in this patient labyrinth on his map the outline of his own face.”
17. Interestingly enough, in this light, Popper’s arguments plus their respective counterarguments, as scientific hypotheses and refutations, must continue ad infinitum, for we can know only what is not the case.
5. MEDIATING FALLIBLY
1. With reason Peirce solemnly declared that there can be no final interpretant for the immanent sign user. Moreover, Peirce’s “mind of some vast consciousness” as all-sufficient and self-sufficient is also in-sufficient for finitistic thinking. It can be accounted for solely with infinitistic thinking of the Cantor set sort: Aleph + 1 = Aleph → Aleph + 2 = Aleph → Aleph + 3 = Aleph . . . n. The sign of infinity engulfs any and all integers in the series, and in so doing it always remains the same as it was.
2. To give an idea of the magnitude of all possible semiostates, cyberneticist Stafford Beer once calculated that the brain considered as a digital device would possess the capacity for 10109 possible brain states. For practical purposes this number can be described as countably infinite even if Beer erred to the tune of a few billion. By comparison, the number of electrons in the universe is estimated to be a mere 10120 (Fraser 1979:114).
3. This superposition of semiostates and their “collapse” into a state of awareness is comparable to Bohm’s backgrounded (the implicate) and foregrounded (the explicate), as described in chapers 1 and 2 (see also Merrell 1983, 1990).
4. The terms are used by Lacan (1966, 1977; also Wilden 1968). Although I do not follow Lacan’s definitions, I have appropriated his terms for the context of this discussion.
5. Yet I do not subscribe to Hofstadter’s algorithmic view of mind according to which it is sufficient to compute an algorithm (or perhaps the end product of an algorithm) in order to determine whether thoughts, or even consciousness, have been developed during the implementation of that algorithm. This is an operational or Turing test by means of questions and answers between the investigator and the object of investigation (considered as a mere computing device) which will eventually lead to a determination of the level of awareness or consciousness of the object (see Penrose 1987).
6. This “clash” is brought about when an unexpected item of the communitygenerated “semiotically real,” or the “real” itself, happens to pop up in the interaction between the individual self and its other in confrontation with the external Other.
7. It bears mentioning at this juncture that in spite of the differences between Wheeler, Prigogine, Bohm, and Peirce, their thesis on the evolution of laws draws them together (regarding the general twentieth-century scientific view in this respect, see Briggs and Peat 1989).
8. See also Bohm (1957) for such “swerving” of an actual series of events from the ideal dictated by physical laws.
9. It is conceded today that there is hardly any hope of actually carrying out a program of absolute certainty leading to the exact and true value of a continuous physical quantity on account of the subtlety and complexity of nature. Few investigators in the nineteenth century even dared speculate on the possibility that the classical quest for certainty was a deluded dream, that there are perpetual and spontaneous random fluctuations. Peirce was an exception in this respect. After citing Epicurus, he writes that “we now see clearly that the peculiar function of the molecular hypothesis in physics is to open an entry for the calculus or probabilities.” And later, he observes:
Try to verify any law of nature, and you will find that the more precise your observations, the more certain they will be to show irregular departure from the law. We are accustomed to ascribe these, and I do not say wrongly, to errors of observation; yet we cannot usually account for such errors in any antecedently probable way. Trace their causes back far enough, and you will be forced to admit they are always due to arbitrary determination, or chance. (CP:6.46)
The development of statistical mechanics and kinetic theory of heat ultimately provided the proper context for acceptance of Peirce’s reservations about classical determinacy, but unfortunately, rejection of the kinetic theory did not come about until the early years of this century. Consequently, Peirce has been given little credit for his path-breaking efforts (see Suppes 1984:83-84).
10. Interestingly enough, Peirce claims that synechism flatly denies Parmenides’ universe, declaring that being “is a matter of more or less, so as to merge insensibly into nothing” (CP:7.569).
11. For a discussion of the relevance of hypercycles to semiosis, see Merrell (1990; Merrell and Anderson 1989b).
12. Schrödinger invented what he called a “ridiculous example” with which to debunk the Copenhagen interpretation of quantum mechanics:
A cat is placed in a steel chamber, together with the following hellish contraption. . . . A Geiger counter contains a tiny amount of radioactive substance, so tiny that within an hour one of the atoms may decay, but it is equally probable that none will decay. If one decays the counter will . . . activate the little hammer which will break a container of cyanide. (Schrödinger, in Layzer 1990:106)
According to the Copenhagen interpretation, at the end of an hour, the cat-plus-hellish-contraption will be half in the state “cat alive, bottle intact” and half in the state “cat dead, bottle broken.” This apparently contradictory situation was Schrödinger’s method of demonstrating that Bohr and his cohorts were on the wrong track. According to quantum physics, a system can change in two completely different ways: while it is not being observed it evolves continuously and predictably—which was Schrödinger’s preference-but if a measurement is carried out to discover which of its discrete states the system is “actually” in, its state changes discontinuously and unpredictably—the Copenhagen view. At the end of the experiment, the cat must be in one of its two possible states, and which one it is in is a matter of chance (for a layperson’s account, see Gribbin 1984).
6. FROM A BROADER POINT OF VIEW
1. At this juncture I should point out that relatively recent cognitive studies are generally based on the dualistic premise that the mind can be studied independently of the brain, that psychology’s focus on “software” is independent of neurophysiology, which focuses on “hardware” (see Fodor 1968, 1975; Johnson-Laird 1983; Miller, Galanter, and Pribram 1960; Oatley 1978; Putnam 1960). The course of this trend has been changing during the past few years, however, about which I will have some brief observations below.
2. See Murphey’s (1961) interpretation of Peirce, though Murphey was criticized, unjustly I believe, by Riley (1968), who argues that Peirce was never an epistemological or ontological idealist—as Murphey claimed he was—during his early years, and later dropped his idealism for an epistemological realism.
3. This is tantamount to the paradox, introduced in chapter 1, of the selective and nonselective domains. The problem inevitably arises when two incompatible frameworks—finitude and the infinite in this case—are artificially thrown into the same mixing bowl. It is the problem, as I have intimated often throughout this essay, existing at the root of all holistic, cosmological visions.
4. Almeder concludes, nonetheless, that, contrary to Peirce, both epistemological realism and objective idealism as they are traditionally defined cannot be consistently endorsed. Epistemological realism presupposes a slash between mind and the world’s furniture, while objective idealism dissolves that slash with the claim that everything is mind. One asserts what the other denies; one is a full-blown dualism or pluralism, the other a radical monism. They appear to defy synthesis, which is precisely what Peirce claims to do. Regardless of how one reads Peirce, Almeder argues, a “tragic flaw” remains, namely, his belief that the property of mind is irreducible to a property of matter. As Peirce would have it, however, everything in the universe consists of some sort of psychic energy, which apparently subsumes epistemological realism into an idealism devoid of true individuals. In this sense epistemological realism, according to Peirce’s use of the term, is radically distinct from the term’s ordinary usage. Even if one embraces Peirce’s unorthodox nomenclature, Almeder continues, it is difficult to comprehend how one can take the next giant step to construe everything as mindlike: if some things are mindlike and therefore irreducible to a property of matter, would it not equally follow that some things are not mindlike? Almeder (1980:184-85) concedes that though Peirce’s belief is not necessarily false, he nevertheless did not effectively argue his point.
5. Saussure (1966:82-88) used the example of the stem of a plant considered in terms of either its longitudinal development or a cross-section at a given point along the stem for illustrating diachrony and synchrony respectively. I could use the term system instead of organism with respect to Saussure, but system would be more relevant to synchrony, in the strict Saussurean sense, which has been associated with structure—though Saussure did not use the latter term.
6. There have been heated debates among cognitive psychologists and intelligence researchers concerning the process of thinking. Some hypotheses from artificial intelligence assign an epiphenomenal role to the perception of images on the assumption that the base components making up an image consist in atomic “bits” of information which combine to form abstract propositional structures. That is, according to this hypothesis, picturelike entities are not stored in memory. They can, however, be constructed by recombining a set of atomic “bits” to form holistic patterns (analogous to the ideal language of the logical positivists which begins with fundamental atomic propositions and combines them into logically cogent arguments). In this sense our “seeing” relations between things in our perceived world is dependent upon our ability to “say” them by means of propositions and with an explicit code—that is, the relations are dependent upon our possession of some sort of language (Anderson and Bower 1973; Collins and Quillian 1972).
Critics of this view reject such a complete dependency of images or “pictures” on propositional structures. They argue in favor of “dual coding” according to which conceptual and propositional messages are conveyed by means of a linearly generated digital language consisting in discrete entities, while analog (iconic) images form a distinct but complementary mode of communication. There always exists a degree of interaction between the two, but neither is clearly dependent upon the other. According to this notion, “seeing” and “saying” are distinct but complementary activities (Neisser 1967; Paivio 1969, 1971).
On the other hand, some theorists working in the information-processing tradition have suggested that there may exist a form of representation which differs radically from both sensory iconic patterns and linear verbal-propositional descriptions. In this light, Z. W. Pylyshyn (1973:7) concurs that
the need to postulate a more abstract representation—one which resembles neither pictures nor words and is not accessible to subjective experience—is unavoidable. As long as we recognize that people can go from mental pictures to mental words or vice versa, we are forced to conclude that there must be representation (which is more abstract and not available to conscious experience) which encompasses both. There must, in other words, be some common format or interlingua.
According to this hypothesis, a relatively simple set of symbol structures can provide for a parsimonious account both of thought images and of the deep structure of language. There must be some common principle underlying our “seeing” relations and our “saying” them. At some level, “seeing” and understanding what is “seen” is equivalent in structure to “saying” and understanding what is “said” (Fodor 1975).
7. The moiré effect is appropriate to the model of semiosis developed in this inquiry. A particular style of “op art,” moiré results from the creation of optical effects through the manipulation of geometric forms, color dissonance, and kinetic elements to exploit visual illusions. The effect is produced by a surface kinetic which sets off a two-dimensional surface into an apparently three-dimensional pulsation, much like a collection of Necker cubes on a sheet oscillating between the two possible actualizations. It is comparable also to the hologram, Bohm’s image of the holomovement, and Jenny’s dancing grains of sand; there is perpetual scintillating movement, but no overall change (see Ulmer’s [1985] pedagogical application of the moire and other related phenomena).
8. Commensurate with irreversibility, this blackened grain in the camera is recorded, a hypothesis is supposedly verified, a paper is written and published, the scientific community applauds, the author eventually gets a write-up in Scientific American and a few lines in the New York Times and Time magazine, and plush government grants are suddenly forthcoming. All because of a solitary photon! Ah, the power of the sign.
9. Physics I is that of Galilean-Newtonian mechanics, physics II of relativityquantum theory, and physics III Wheeler terms “meaning physics,” which calls for an exploitation of language itself, which is, in the final analysis, the only “reality” we can know, that is, the “semiotically real.”
7. THE SPECTRUM OF MIND-SIGN
1. Among the various compatibilities and divergences between Peirce and Popper to which I have alluded above, Peirce’s “refutation” constitutes one of the most evident zones of overlap. In Peirce’s (CP:1.120) words:
The best hypothesis, in the sense of the one most recommending itself to the inquirer, is the one which can be the most readily refuted if it is false. This far outweighs the trifling merit of being likely. For after all, what is a likely hypothesis? It is one which falls in with our preconceived ideas. But these may be wrong. Their errors are just what the scientific man is out gunning for more particularly. But if a hypothesis can quickly and easily be cleared away so as to go toward leaving the field free for the main struggle, this is an immense advantage. (CP:1.120; see Rescher 1978 for further discussion)
2. By the end of the nineteenth century it was considered in many quarters that the quest for Absolute Truth was nearing its end point, and that the physicist’s task was merely that of, so to speak, calculating the next decimal point. Prima facie this calls to mind Peirce’s asymptote model. While Peirce was somewhat off the mark in this respect (see Suppes 1984), there is a fundamental difference between his conception of the investigative community and the general turn-of-the-century attitude toward the future of the classical episteme. In the first place, Peirce was already a vehement critic of the classical paradigm, especially regarding its mechanistically oriented determinism. Second, Peirce was always explicit concerning the impossibility of any finite number of human beings’ ever coming into possession of the Grand Prize. And third, Peirce held that, though the asymptote generally described a fairly smooth curve, he also admitted to fits and jerks, and even catastrophes, all of which serves further to substantiate his indeterminacy of meaning.
3. I refer here not simply to mysticism in contrast to rational thought or to the Freudian consciousness/unconsciousness dichotomy per se, but rather to the distinction conveyed in general by the likes of Conant (1964), Cassirer (1953-57), Geertz (1973, 1983), Langer (1942), and Northrop (1946) (see also the synthesis by Wilber [1977]). Nonetheless, as implied above, this general concept is in a certain manner germane to much Eastern thought. It is revealed in D. T. Suzuki’s (1971:131) words that, in the beginning,
the will wants to know itself, and consciousness is awakened, and with the awakening of consciousness the will is split in two. The one will, whole and complete in itself, is now at once actor and observer. Conflict is inevitable; for the actor now wants to be free from the limitations under which he has been obliged to put himself in his desire for consciousness. He has in one sense been enabled to see, but at the same time there is something which he, as observer, cannot see.
4. In general, the “radical meaning variance” theory of science propagated by the likes of Feyerabend (1975), Hanson (1958), and Kuhn (1970) posits that any and all theories are invariably language-bound, world view-bound, and dependent upon particular cultural contexts of discourse.
5. The use of symbolism and symbols in this regard is not that of Peirce’s symbol, which, in conjunction with the index and icon, makes up his most general class of signs. I do not, of course, wish to disparage Peirce’s notion of the symbol, but to foreground the West’s penchant for abstraction, for hypertrophying symbols to the extent that they are presumed to be coterminous with the “actually real.”
6. Contemporary conventionalism, of which the Sapir-Whorf hypothesis is a powerful expression, had its beginning in the nineteenth century. It was formulated by, among others, Poincare to account for the emergence of non-Euclidean geometries and phenomena that could not be explained by classical physics. In the twentieth century, conventionalism was taken a giant step further with the corollary that language uniquely determines the nature and structure of the perceived and conceived world, and that by internalizing different languages, one is thereby creating different pictures of “reality.”
7. Arthur Burks (1980:283) puts forth the homology that “an individual man is to the ultimate community of knowledge seekers as an individual sign is to the complete and living system of signs, or language.” He goes so far as to say that language in man is a genetic program, and that man is ultimately an algorithm. This reductionistic computer model stretches the analogy between man and sign to the breaking point. Peirce was actually against mind as an algorithm machine, especially at the abductive level.
8. Fairbanks (1976:19) suggests that we accept this and other such apparently outlandish statements literally “to see their more interesting philosophical implications.” I don’t know whether Peirce intended his words to be construed literally or not, but I would guess the latter.
9. I refer the reader once again to Dunne (1934) and Matte Blanco (1975), whose consciousness within consciousness reduplicates the image of the map paradox.
10. Perhaps no contemporary philosopher has stressed the fundamental importance of distinguishing the two modes of knowing more than Whitehead, whose voice has occasionally, though fleetingly, been overheard in this book. Abstraction and a bifurcation of the seamless fabric of the universe belong to the Western way which has mistaken those abstractions for concrete reality, thus committing the “fallacy of misplaced concreteness.”
11. This is much in keeping with J.J. Gibson (1950, 1966, 1979) and Quine (1960, 1969), who refer to perception as reactions at nerve endings.
12. Significantly enough, in this regard, Sebeok (1979:xiii) observes that all “survival-machines,” all cells, are only a sign’s way of making another sign, which recalls the words of François Jacob that the dream of a bacterium is to make another bacterium.
13. Sachs goes on to point out that Buber’s relations are comparable to Bohr’s complementarity. See Agassi (1986) for a further discussion of Buber and complementarity.
14. Margenau, I must hasten to add, concedes that he indulges in a measure of speculation that will be questioned by many readers, though he maintains that his speculation is logically rigorous and consistent. The same conclusion, he adds, can be rephrased “in less mathematical form” in the other two of his three approaches.
15. Interestingly enough, a collection of papers on the “many worlds interpretation” of quantum theory, edited by DeWitt and Graham (1973), opens with an aphorism taken from Borges’s short story “The Garden of Forking Paths,” which is a strikingly faithful image of Hugh Everett’s hypothesis.
16. Thus Schrödinger (1967:165) states that the multiplicity of individual minds “is only apparent, in truth there is only one Mind.” And he uses that most common of analogies, the mirror, to illustrate the split of one into two, subject and object, interpreter and interpretant, which is actually but one. As Varela (1975:22) puts it:
In finding the world as we do, we forget all we did to find it as such, and when we are reminded of it in retracing our steps back to indication, we find little more than a mirror-to-mirror image of ourselves and the world. In contrast with what is commonly assumed, a description, when carefully inspected, reveals the properties of the observer. We, observers, distinguish ourselves precisely by distinguishing what we apparently are not, the world.
This, of course, comes within a hair of Rorty’s “Mirror of Nature” fallacy, as does Peirce’s allusion to “man’s glassy essence,” to which I referred in chapter 3. However, see Varela (1984a) for an “antifoundationalist” statement regarding his hypothesis.
8. NOTHING AND EVERYTHING
1. Sunya plus the suffix ta can be translated as “ness”—”nothingness,” voidness, openness. But to contrast “nothingness” to somethingness is to fall into binary “thing” thinking once again. “Nothingness” is, most simply, Plato’s shadow on an unilluminated wall of the cave. The problem is that we generally tend to see “things” as either A or not-A. If a child is good, he is not bad; this is the power and pathos of conceptualization. “Nothingness,” in contrast, is free of all things and concepts.
2. As mentioned briefly in the previous chapter, the laws of nature, rather than static, are the result of evolution (CP:6.13; 7.514-15), hence we cannot pin these laws down for all time (CP:7.514). In fact, not only logic but time itself is the product of the evolutionary process (CP:6:189, 6.200, 6.214, 8.318). Moreover, for Peirce the evolution of the universe is hyperbolic (CP:8.317), which is commensurate with his asymptote model.
3. This flattening of language by razing all hierarchical metalinguistic schemes can be viewed as an extension of the “limitative theorems” of Gödel and others (DeLong 1970), the “radical meaning variance” theorists’ critique of logical positivism’s metalanguage hypothesis and its distinction between observation language and theoretical language (Feyerabend 1975; Hanson 1958; Kuhn 1970), and the complementarity principle’s requiring distinct context-dependent languages within an encompassing framework (Heelan 1983).
4. The origin as absence is not the mark of prior presence. Whiteness (the blank page) is the presence (juxtaposition) of all colors, and blackness the absence. The white sheet is a potentia from which all colors (signs) of varying wavelengths can be generated. But the sheet itself is nothing; it is “nothingness,” which is perhaps most adequately stated as both presence and absence, both blackness and whiteness.
APPENDIX 2
1. For brief comparisons of Peirce and Sheffer, see Bochenski (1961), Church (1956), and Kneale and Kneale (1962).
We use cookies to analyze our traffic. Please decide if you are willing to accept cookies from our website. You can change this setting anytime in Privacy Settings.