Korzybski: A Biography (Free Online Edition)
Copyright © 2014 (2011) by Bruce I. Kodish
All rights reserved. Copyright material may be quoted verbatim without need for permission from or payment to the copyright holder, provided that attribution is clearly given and that the material quoted is reasonably brief in extent.
Uncertainty was built into the abstracting process. Perhaps the most ‘certain’ (least uncertain) knowledge came when we learned how a map did not fit a given territory:
Around the time Korzybski was writing this, Karl Popper in Vienna was developing related notions while writing what would become his first book, Logik der Forschung, to be published in 1934. Popper’s book, later published in English as The Logic of Scientific Discovery (1959), was his opening volley in the theory of knowledge and scientific methodology. His work centered around the importance of falsification (disconfirmation). According to Popper, a theory that couldn’t conceivably be tested and shown false was not scientific. The best tests of a theory applied the most rigorous, potentially falsifying challenges to it. We could learn something definite when a theory did not pan out. A successful scientific theory, one with better predictivity than others at a given date, had simply better survived attempts to test it.
A minimal ‘maybeness’ pertained to even the best, nigh certain scientific theories. As Popper wrote in the closing pages of his book:
I have found no evidence that Korzybski was ever aware of Popper. Popper apparently didn’t know or think much of Korzybski’s work (according to Stuart A. Mayper, chemist, korzybskian scholar/teacher, and a former editor of the General Semantics Bulletin, who studied with Popper at the University of London).(25) Nonetheless, their complementary views both pointed to an uncertaintist perspective.
With similarity (not ‘sameness’) of structure providing the sole link between different levels of abstracting, one could only have varying degrees of similarity of a map with a territory, giving varying degrees of predictability. You could not have a ‘perfect’ map. But you could strive for maximum probability of predictability in a map at a given date. In 1931, Korzybski formulated this in terms of a “general principle of uncertainty”:
Korzybski agreed with Einstein: indeterminism was not a scientific option. He argued that ‘indeterminism’ would involve a denial of structure or relations, an inherently contradictory viewpoint since, as he wrote to William Alanson White in early 1931,
Unlike Einstein, Korzybski did not see causality as at odds with chance. He embraced the fundamental role of probabilities in our knowledge of anything. Accepting the generalized uncertainty of all statements (not only the statements of quantum mechanics) did not require abandoning the search for knowledge and predictability. It did require abandoning the “one cause, one effect” determinism, which seemed based upon two-valued ‘certainty’. This had to be refashioned into a probabilistic, many-valued determinism which would require a many-valued logic of probability in which two-valued logic remained as a special case.* (Korzybski argued for such a probabilistic logic, but didn’t develop one himself.) Second-order certainty of degrees of uncertainty remained. “If so, then invariably and always so” had to give way to “If so, then probably so, which could depend upon multiple factors, e.g., x, y, z, etc.” Knowledge and uncertainty existed together, inexorably intertwined.
* Korzybski only rejected the universality of two-valued, either-or ‘logic’ as a general orientation. Within a multi-valued orientation, either-or distinctions sometimes appeared appropriate. As he wrote to Keyser soon after the book was published, “... mathematics in the main (1934) ( only) would be impossible if today 1 + 1 = 2 and tomorrow 1 + 1 ≠ 2 . So when we want sharp tools we use two-valued orientations, but that’s VOLITIONAL. I try to establish some sharp tools for efficiency in human orientations, and once in a while I must use two-valued orientations...” [AK Collected Writings, p. 186]. He had already made this point in Science and Sanity [Korzybski 1994 (1933), pp. 94, 195, 405, 760–761], and continued to do so in later writings. Still, petty criticism about his supposed rejection of two-valued ‘logic’ would also continue.
Even with some degree of uncertainty, reliable knowledge remained possible. Indeed, under some circumstances—when developed through a scientific approach—higher order abstractions (also known as generalizations or inferences) could provide the most reliable knowledge possible at a date. Some people later got the notion that Korzybski was against making generalizations. This conclusion could not have come from a careful reading of his book. “It is no mystery that when we want to look further into the past and future we need higher and higher order abstractions.”(29)
Linking knowledge with uncertainty also related to a conscious recognition—at the heart of non-elementalism—of the importance of non-additivity. “As a structural fact, the world around us is not a ‘plus’ affair, and requires a functional [relational] representation.”(30) Yet it was especially easy to misrepresent complex, non-additive, nonlinear processes, relationships, organizations, etc. as ‘plus’ affairs. Since the early 1920s—long before the formal development of systems theory, ‘chaos’ theory, ‘complexity’ theory, etc.—Korzybski had been concerned with this additive or plus tendency in people’s evaluating, reflected in their language. The addition of one ‘small’ factor in a situation didn’t necessarily lead to the ‘same old thing’. One mother and one father ‘plus’ one small baby involved a whole new, complex set of relations as a couple became a family. An uncertaintist perspective, informed by non-additivity, could prepare one to expect the unexpected and thus reduce undesirable shocks from the new.(31) Korzybski devoted an entire chapter of the book “On Linearity” to the ramifications of this. After the book was published, “New factors: the havoc they play with our generalizations” became a prominent focus of his teaching and writing.(32)
Interestingly Leibniz had written that, “Those great principles of a sufficient reason and of the identity of indiscernibles, change the state of metaphysics. That science becomes real and demonstrative by means of these principles; whereas before, it did generally consist of empty words.”(33) Korzybski’s parallel principles of structure and of non-identity—with their attendant redefinition of knowledge in terms of structure, generalized uncertainty, probabilistic ∞[infinite]-valued determinism, non-additivity, etc.—seemed to Korzybski to change the state of 20th-century metaphysics, epistemology, etc., into something real and demonstrative, and of potentially great human significance—a foundation for a science of ‘man’.
Notes
Uncertainty was built into the abstracting process. Perhaps the most ‘certain’ (least uncertain) knowledge came when we learned how a map did not fit a given territory:
All knowledge is hypothetical, in which…[t]he most important facts must be negative. When the [linguistic and empirical] structures do not match, then we learn something quite definite about the empirical structures. (22)
Because words are never the things we speak about, the sole link between languages and the objective world being structural, the only ‘positive’ facts about this world are of the old ‘negative’ character. (23)
Around the time Korzybski was writing this, Karl Popper in Vienna was developing related notions while writing what would become his first book, Logik der Forschung, to be published in 1934. Popper’s book, later published in English as The Logic of Scientific Discovery (1959), was his opening volley in the theory of knowledge and scientific methodology. His work centered around the importance of falsification (disconfirmation). According to Popper, a theory that couldn’t conceivably be tested and shown false was not scientific. The best tests of a theory applied the most rigorous, potentially falsifying challenges to it. We could learn something definite when a theory did not pan out. A successful scientific theory, one with better predictivity than others at a given date, had simply better survived attempts to test it.
A minimal ‘maybeness’ pertained to even the best, nigh certain scientific theories. As Popper wrote in the closing pages of his book:
The old scientific idea of episteme—of absolutely certain, demonstrable knowledge—has proved to be an idol. The demand for scientific objectivity makes it inevitable that every scientific statement must remain tentative for ever. It may indeed be corroborated, but every corroboration is relative to other statements which, again, are tentative. Only in our subjective experiences of conviction, in our subjective faith, can we be ‘absolutely certain’. (24)
I have found no evidence that Korzybski was ever aware of Popper. Popper apparently didn’t know or think much of Korzybski’s work (according to Stuart A. Mayper, chemist, korzybskian scholar/teacher, and a former editor of the General Semantics Bulletin, who studied with Popper at the University of London).(25) Nonetheless, their complementary views both pointed to an uncertaintist perspective.
With similarity (not ‘sameness’) of structure providing the sole link between different levels of abstracting, one could only have varying degrees of similarity of a map with a territory, giving varying degrees of predictability. You could not have a ‘perfect’ map. But you could strive for maximum probability of predictability in a map at a given date. In 1931, Korzybski formulated this in terms of a “general principle of uncertainty”:
…on objective levels we deal structurally with absolutely individual stages of processes and situations and by necessity we speak in higher order abstractions and generalities and use many multiordinal terms (without the use of which no speaking is possible), so any possible statement about the objective levels must be only probable in different degrees, which introduces a fundamental and entirely general Ā principle of uncertainty. Heisenberg’s restricted principle in physics appears only as a special case. (26)The implications of this for Korzybski seemed profound. For one thing, it resolved the problem of “indeterminism” brought to the fore by the development of quantum physics. Many scientists and philosophers were ‘reeling’ from the replacement of classical, two-valued determinism with statistical approaches for understanding the submicroscopic world. Many had leaped to assuming the end of causality and perhaps the loss of confidence in scientific knowledge. Many, like Bridgman, appeared to despair or rail at this conclusion. Einstein, for example, accepted the findings of quantum physics but rejected the fundamental nature of the statistical approach (although he had helped develop it). He continued to embrace classical determinism, searching for hidden variables behind the probabilities of quantum physics. He could never accept that ‘Herr Gott’ might play dice with the universe.
 |
| Does God Play Dice? |
…the language of causation which in a subtler analysis requires series (mathematics) is a characteristic of human rationality and [as such] has little to do with the world, but it is also the foundation of relations and structure and a non-aristotelian system. If empirical data lead us to indeterminism verbally, such language is in structure non similar to the world and our nervous system and simply should be changed to another language which retains determinism. (27)Here was another echo of Leibniz. One of Leibniz’s guiding principles had been that of “sufficient reason”—“nothing happens without a sufficient reason why it should be thus rather than otherwise.”(28) According to Leibniz’s 17th-century scientific ‘faith’, anything that happens exists within the compass of potential human understanding and knowledge. Korzybski could be seen to have held an updated form of Leibniz’s principle: any knowledge—‘scientific’ or otherwise—involved a search for structure. Determinism provided the test for structure. But for Korzybski, determinism must involve probabilities.
Unlike Einstein, Korzybski did not see causality as at odds with chance. He embraced the fundamental role of probabilities in our knowledge of anything. Accepting the generalized uncertainty of all statements (not only the statements of quantum mechanics) did not require abandoning the search for knowledge and predictability. It did require abandoning the “one cause, one effect” determinism, which seemed based upon two-valued ‘certainty’. This had to be refashioned into a probabilistic, many-valued determinism which would require a many-valued logic of probability in which two-valued logic remained as a special case.* (Korzybski argued for such a probabilistic logic, but didn’t develop one himself.) Second-order certainty of degrees of uncertainty remained. “If so, then invariably and always so” had to give way to “If so, then probably so, which could depend upon multiple factors, e.g., x, y, z, etc.” Knowledge and uncertainty existed together, inexorably intertwined.
* Korzybski only rejected the universality of two-valued, either-or ‘logic’ as a general orientation. Within a multi-valued orientation, either-or distinctions sometimes appeared appropriate. As he wrote to Keyser soon after the book was published, “... mathematics in the main (1934) ( only) would be impossible if today 1 + 1 = 2 and tomorrow 1 + 1 ≠ 2 . So when we want sharp tools we use two-valued orientations, but that’s VOLITIONAL. I try to establish some sharp tools for efficiency in human orientations, and once in a while I must use two-valued orientations...” [AK Collected Writings, p. 186]. He had already made this point in Science and Sanity [Korzybski 1994 (1933), pp. 94, 195, 405, 760–761], and continued to do so in later writings. Still, petty criticism about his supposed rejection of two-valued ‘logic’ would also continue.
Even with some degree of uncertainty, reliable knowledge remained possible. Indeed, under some circumstances—when developed through a scientific approach—higher order abstractions (also known as generalizations or inferences) could provide the most reliable knowledge possible at a date. Some people later got the notion that Korzybski was against making generalizations. This conclusion could not have come from a careful reading of his book. “It is no mystery that when we want to look further into the past and future we need higher and higher order abstractions.”(29)
Linking knowledge with uncertainty also related to a conscious recognition—at the heart of non-elementalism—of the importance of non-additivity. “As a structural fact, the world around us is not a ‘plus’ affair, and requires a functional [relational] representation.”(30) Yet it was especially easy to misrepresent complex, non-additive, nonlinear processes, relationships, organizations, etc. as ‘plus’ affairs. Since the early 1920s—long before the formal development of systems theory, ‘chaos’ theory, ‘complexity’ theory, etc.—Korzybski had been concerned with this additive or plus tendency in people’s evaluating, reflected in their language. The addition of one ‘small’ factor in a situation didn’t necessarily lead to the ‘same old thing’. One mother and one father ‘plus’ one small baby involved a whole new, complex set of relations as a couple became a family. An uncertaintist perspective, informed by non-additivity, could prepare one to expect the unexpected and thus reduce undesirable shocks from the new.(31) Korzybski devoted an entire chapter of the book “On Linearity” to the ramifications of this. After the book was published, “New factors: the havoc they play with our generalizations” became a prominent focus of his teaching and writing.(32)
Interestingly Leibniz had written that, “Those great principles of a sufficient reason and of the identity of indiscernibles, change the state of metaphysics. That science becomes real and demonstrative by means of these principles; whereas before, it did generally consist of empty words.”(33) Korzybski’s parallel principles of structure and of non-identity—with their attendant redefinition of knowledge in terms of structure, generalized uncertainty, probabilistic ∞[infinite]-valued determinism, non-additivity, etc.—seemed to Korzybski to change the state of 20th-century metaphysics, epistemology, etc., into something real and demonstrative, and of potentially great human significance—a foundation for a science of ‘man’.
You may download a pdf of all of the book's reference notes (including a note on primary source material and abbreviations used) from the link labeled Notes on the Contents page. The pdf of the Bibliography, linked on the Contents page contains full information on referenced books and articles.
22. Korzybski 1994 (1933), p. 324.
23. Ibid., p. 365.
24. Popper, p. 280.
25. Stuart A. Mayper, Personal communication.
26. Korzybski 1994 (1933), p.760.
27. AK to W. A. White, 2/4/1931. AKDA 23.246.
28. Leibniz 1979 (1951), p. 222.
29. Korzybski 1994 (1933), p. 483.
30. Korzybski 1994 (1933), p. 605.
31. Korzybski believed that a basic study of permutations and combinations could help give a feel for the complexities that arise from increasing the number of elements in a situation. Later, in his seminars, he would recommend Stanley Jevon’s chapter “On The Variety of Nature” from The Principles of Science as a good introduction to this.
32. See “Introduction to the Second Edition”, Korzybski 1994 (1933), p. lv-lvii.
1 comment:
Re: Popper. Don't let it slip by that Korzybski did NOT consider himself a Positivist as Popper did of himself. Tricky but crucial point.
Post a Comment