Korzybski and Bateson: paradoxes in 'consciousness of abstracting'.

ALFRED KORZYBSKI maintained that training in "consciousness of abstracting" would lead people beyond the paradoxes of abstraction. Gregory Bateson, on the other hand, argued for the inability to achieve this thoroughly. He writes, "Korzybski was, on the whole, speaking as a philosopher, attempting to persuade people to discipline their manner of thinking. But he could not win. When we come to apply his dictum to the natural history of human mental process, the matter is not so simple." (1979, pp.30-31) This point can be brought out more generally, where Bateson suggests,

   We do not, any of us, achieve rigor. In writing, sometimes, we can
   take time to check the looseness of thought; but in speaking, hardly
   ever ... I know that I personally, when speaking in conversation and
   even in lecturing, depart from the epistemology outlined in the
   previous chapter; and indeed the chapter itself was hard to write
   without continual lapses into other ways of thinking and may still
   contain such lapses. I know that I would not like to be held
   scientifically responsible for many loose spoken sentences that I
   have uttered in conversation with scientific colleagues. But I also
   know that if another person had the task of studying my ways of
   thought, he would do well to study my loosely spoken words rather
   than my writing. (1968, p.230)

Finally, toward the end of his career Bateson concluded that, "It seems to be a universal feature of human perception, a feature of the underpinning of human epistemology, that the perceiver shall perceive only the product of his perceiving act. He shall not perceive the means by which that product was created." (Bateson, 1977, p.238)

I will not, as if by fiat and pronouncement, take sides on this issue. Rather, I here attempt to explicitly walk through a series of illustrations that help bring out how the "logic or method" of Korzybski remains correct but the conclusion of Bateson should not be underestimated. I try to demonstrate that we, in our warnings and suggestions regarding the troubles with abstraction, inevitably make the very kinds of mistakes that we attempt to extirpate.

Even if we desire to speak about entities which are "not yet" classified (such as the concepts of "the unspeakable level" or perhaps "the infinity of 'things' not yet talked about"), we so easily forget how we already (i.e., therein) have classified them. (1) I apologize for my utter literalism here, but I would like you to carefully examine the previous sentence. In particular, the words "entities" and "things," to the extent that they make reference, refer to what are already within a class. This means that if we say what something is called rather than "is," we might thereby conclude that it is not yet classified. Syntax is such that it enables us to forget that we already have called it "something" and also "thing." And even there, in that previous sentence, we seemed likewise to forget that we called the thing "it." And there it was called "thing." Across these few simple sentences, we hop from one word to another, obliviously taking some to be labels for things while taking others to be not more labels; it is as if we make reference to what we then pretend we didn't talk about.

Consider an even more direct and immediate manner of explaining the term "abstraction." I could stand before people, hold a cup of coffee in my hand, and then state that the thing I am holding and drinking from is not a "cup." "'Cup,'" I further say, "is the English word we use to refer to this thing." Then, silently tapping the cup and dramatically wrapping my hands around it, I say, "This, what I now am holding in my hands, is not a 'cup,'; 'cup' is word, a kind of mapping, that enables us to categorize things, in this case what I am holding." What is critical is that although my silent action accompanies my words, the words themselves actually increase in abstraction: The word "cup" is labeled as an abstraction while simultaneously the word "thing" seemed to be used as if it were the less abstract term. By syntactical combination, integrating different levels of abstraction though an overlapping reference, we produce utterances in which we seem to talk about what would be independent of our talk about it; higher abstractions are thus taken as if they were the "thing" more directly.

These paradoxes of abstraction occur because we conflate the difference in logical type between abstract words (e.g., "things," "entities," "objects") and metalanguage or words that explicitly refer to the verbal order (e.g., "speech," "language," "words"). Although non-metalinguistic words (abstract words) can posit that their referents preceded the words we used to refer to them, we are able to do so only because of syntax: Terms such as "language," "words," "speech," or even "verbal level," when syntactically combined with abstract words such as "object" or "thing," operate as a mode of "overlapping reference" (cf. Holenstein, 1976). They function as a "code to message" reflexivity that allows us to make sense of the claim that "objects" and "things" preceded the words by which we refer to them. Hence, metalinguistic references enable a reflexivity that becomes taken-for-granted in the claim that "things came before language." By the very syntax of our utterances, we say that things precede language, and, this does make sense, but only because we have used metalinguistic references and thereby already have referred to the verbal order.

The roots of such difficulties can also be found where we attempt to thoroughly separate a class from its name. If we confuse a class with its name, we obviously suffer from logical-typing errors. But the question remains: is it even possible for this to be thoroughly avoided? Doesn't an unnamed category seem not to be a category at all? What, that is, would an unnamed category be a category of?

Perhaps a more illustrative example would help: Common sense suggests that actual physical apples must have preceded the abstract class of apples, and yet, if we do not yet have the class of apples, then how could any one apple be counted as an apple? As Lee Thayer (1997) suggests, the difficulty is that "To have one of anything, we already must have a category." (p.75) Thayer's point is that individual entities do not precede the categories by which we class them. Take, as one more example, Bateson's opening remarks regarding the problem of "play." Bateson (1956) states, "We live in a universe of namables. Within that universe we make classes." (p.145) This is certainly clear enough, but could the namables come before the classes? Is it not obvious that the namables are already within a class, the class of namables?

At this point, a critic still might try to argue that language is not needed for the existence of kinds or classes of things. 'It is only for our convenience, it merely aids us in labeling naturally occurring types,' the critic might argue. The critic may further state: "It does not matter what you call the thing. Whether you call something a 'cup,' a 'drinking vessel,' or an 'object,' the thing is still here." This does seem to make sense at least initially. And yet, as I have tried to show in several ways, we have not made good sense as much as we have enabled ourselves to overlook our non-sense. It is worth recalling that when the White Knight meets Alice, in Through the Looking Glass, he tries to cheer her up by playing some music. But first he states,

    'The name of the song is called "Haddock Eyes."'

    'Oh, that's the name of the song is it?' Alice said, trying to feel
    interested.

    'No, you don't understand,' the Knight said, looking a little vexed.
    'That's what the name is called. The name really is "The Aged Aged
    Man."'

    'Then I ought to have said "That's what the song is called"?' Alice
    corrected herself.

    'No, you oughtn't: that is quite another thing! The song is called
    "Ways and Means": but that's only what it's called, you know!'

    'Well, what is the song, then?' said Alice, who was by this time
    completely bewildered.

    'I was coming to that,' the Knight said. 'The song really is "A
    Sitting On A Gate": and the tune's my own invention.'

It appears that Alice eventually got to the bottom of all this (also cf. Wilden, 1978), but did she? Could the song be 'A Sitting On A Gate'? Is this not a kind of erroneous identification, a subtle act of taking the words for the thing? Do we now know the song rather its title? More critically asked: what does the White Knight mean by "the tune"? My inquiry focuses not upon what the name of the tune is called, nor what the tune is called, nor even is it an attempt to know the name of the tune. I would like to know how I know (or why I seem to think) that the tune is the song. And, end of the day, I would like to know (i.e., to be able to "re-produce," even if only partly and in memory) the actual song.

The textual imbroglio we find in Carroll's humorous tale forces our attention to what Bateson called the "premise intransitivity" that characterizes naturally communicative frames. Bateson (1955) argues, "It is conventional to argue that if A is greater than B, and B is greater than C, then A is greater than C. But psychological processes do not obey the transitivity of asymmetrical relations. Proposition P may be a premise for Q; Q may be a premise for R, and R may be a premise for P." (p.185) Our everyday talk habitually hops around and between various abstract words and, in doing so, we leave such intransitivity covered over. Thus, when Bateson suggests that he has a "desire to know about those processes whereby organisms pull themselves up by their bootstraps" (1956, p.216), he turns our attention to the ways that communication syntactically imbricates intransitive yet asymmetrical distinctions and thus in-builds different layers of abstraction.

Now, obviously, Korzybski's Science and Sanity (1933) discusses at length the non-identification between "words" and the "un-speakable objective level," and he succinctly summarizes his ideas with the pithy one-liner: "Whatever one might say something 'is,' it is not" (p.409). In this very quotation Korzybski's actual utterance goes against his insights as he states them. By the syntax of the utterance he implies that the words "something" and "it" are not already something said. Korzybski undoubtedly would defend himself and say that this exactly is his point, as he sums it up elsewhere: "It is evident that every time we mistake the object for the event we are making a serious error, and if we further mistake the label for the object, and therefore for the event, our errors become more serious" (Korzybski, 1949, p.245). Here we again find the same difficulties: he uses the words "object" and "event" to state his insights and thereby is forced to use the very resources that he calls into question.

We cannot propose a non-identification--nor call identification into question--without subtly embodying the errors that we wish to challenge. Regarding the paradoxes of abstraction, we thus continue to make the mistakes--and thereby to illustrate--the very difficulties that we attempt to bring under critical attention.

NOTE

1. This point is nicely brought to head by Alan Watts, who compares Korzybski's views with Zen Buddhism. Watts writes, "However, it would seem that Korzybski still thought of the 'unspeakable' world as a multiplicity of infinitely differentiated events. For Zen, the world of 'suchness' is neither one nor many, neither uniform nor undifferentiated ... it teases the mind out of thought, dumfounding the chatter of definition." (pp. 130-131) This issue is taken up also by Walker Percy (1954).

REFERENCES

Bateson, G. (1979). Mind and Nature: A Necessary Unity. New York, NY: Bantam Books.

Bateson, G. (1977). "Afterword," In About Bateson. (Ed., J. Brockman). New York, NY: E.P. Dutton.

Bateson, G. (1968). "Information and Codification: A Philosophical Approach," in Gregory Bateson and Jurgen Ruesch Communication: The Social Matrix of Psychiatry. New York, NY: W.W. Norton & Company.

Bateson, G. (1956). "The Message, 'This is Play'" In Group Processes. (Ed., B. Schaffner). New York: Josiah Macy, Jr. Foundation.

Bateson, G. (1955). "A Theory of Play and Fantasy: A Report on the Theoretical Aspects of the Project for Study of the Role of Paradoxes of Abstraction in Communication." Approaches to the Study of Human Personality. American Psychiatric Association. Psychiatric Research Reports, no 2. Reprinted in 1972 as "A Theory of Play and Fantasy," In Steps to an Ecology of Mind. New York, NY: Ballantine Books.

Holenstein, E. (1976). Roman Jakobson's Approach to Language: Phenomenological Structuralism. (Trans. C. Schelbert & T. Schelbert.). Bloomington, IN: Indiana University Press.

Korzybski A. (1933). Science and Sanity: An Introduction to Non-Aristotelian Systems and General Semantics. Lakeville, CT. The International Non-Aristotelian Library Publishing Company.

Korzybski, A. (1949). "Fate and Freedom" in The Language of Wisdom and Folly: Background Readings in Semantics. (Ed., I. Lee). New York: Harper and Row Publishers.

Percy, W. (1954). The Message in the Bottle: How Queer Man is, How Queer Language is, and What one has to do with the Other. New York: Farrar, Straus, and Giroux.

Thayer, L. (1997). Pieces: Toward a Revisioning of Communication/Life. Norwood, NJ: Ablex Press.

Watts, A. (1957). The Way of Zen. New York: Vantage Books.

Wilden, A. (1987). The Rules are No Game. New York: Routledge & Kegan Paul.

COREY ANTON*

* Corey Anton is the author of Selfhood and Authenticity (SUNY press, 2001), which received the Erving Goffman Award by the Media Ecology Association. He is an associate professor in the School of Communications at Grand Valley State University, Allendale MI. A much larger and modified version of this paper is to be included in The American Journal of Semiotics as part of the Bateson Centennial. The larger piece is called, "Playing with Bateson: Denotation, Logical Types, and Analog and Digital Communication."