Work in Progress: Problems of Concept Formation
Friday 11 March 2022
I take an on-again, off-again interest in the problems of concept formation, and have often consulted Heinrich Rickert’s The Limits of Concept Formation in Natural Science and Carl Hempel’s Fundamentals of Concept Formation in Empirical Science for whatever ideas on concept formation I can plunder from them. I do not as yet have any systematic views on concept formation that I feel the need to propound, nor is there some bête noire of concept formation that I feel the need to slay because I believe it to be hampering progress in science and philosophy. This is perhaps why my progress in concept formation has been slow and episodic at best. But I do think that concept formation is important, and deserves more attention than it gets.
One of my many projects that I have thought about casually but not invested real work in is an essay to be titled, “Concept Formation in the Early Space Age,” which would discuss the many new concepts that had to be formulated (in a relatively short period of time) in order to talk about new technologies of space flight, new activities of space flight, new human experiences as a consequence of space flight, and so on. Space exploration was so unprecedented in human history that its introduction occasioned a new vocabulary, and has sparked debates (like the debate over where exactly the atmosphere ends and outer space begins) over the concepts that had been introduced.
Of course, there were many antecedents to space exploration, no matter how unprecedented. There was decades of science fiction writing, which developed a great many concepts before Sputnik was even placed in orbit, and some of these ideas have yet to be realized, i.e., yet to be tested against experience, and thus yet to be substantively debated because we don’t yet really know what is at stake in these ideas. Some ideas especially important for the Space Age ancient, like the idea of the planet, are ancient. The idea of a planet has mutated many times over its history, and even today the definition and meaning of a planet is not settled. Back in newsletter 114 I discussed what is called the geophysical definition of a planet, which is only one of many definitions of a planet, and not the one employed by the IAU.
With ancient concepts like “planet” we are faced with a different problem than the problem of concept formation simpliciter, but the problems overlap. Many concepts were introduced into our conceptual framework long before the advent of science, and we continue to use these concepts even as their meanings have mutated under the influence of science. Thus we say “sunrise” even though we know that the rotation of Earth is bringing us around into the light of the sun, and the sun isn’t rising at all (Buckminster Fuller liked to use the examples of sunrise and sunset.) What is happening with these concepts of ordinary language is that new scientific concepts are being introduced and then applied ex post facto to existing terms, and this can create a lot of confusion. This confusion is then given a boost by journalists who write things like “the God particle” or “mitochondrial Eve,” with the idea that a new scientific concept can be choked down easier if it is swallowed with a spoonful of sugary traditionalism.
Of course, in rigorous science, we don’t have this problem, because we set up a formal system and define our terms and our operations, but then the scientist must attempt to communicate this formally rigorous knowledge to a wider audience, and we get things like the absurdities of the previous paragraph. And definition is by no means a settled matter; it is an active area of debate in philosophical logic, and I have always had a particular interest in these debates over definition because of their relationship to concept formation. Stipulative definitions can be understood as explicit instances of concept introduction.
When we have one intuitive mode of thought embodied in ordinary language, which seems to point to a kind of intuitive metaphysics, and another formal mode of thought embodied in formal systems and scientific knowledge, which also points to a metaphysics, but a different metaphysics than intuitive thought, and often an incommensurable metaphysics, then we have a problem. One way to handle this problem has been to dismiss our intuitive thought as “folk whatever,” like folk psychology and folk biology, etc. This is the eliminativist position. It doesn’t work very well, because although intuitions can be changed, they change very slowly and only when a better intuition is available to displace an older intuition.
So we can see from the above that “science communicators” deal with some pretty weighty issues, metaphysics among them, but I really doubt that anything like this can be taught. Perhaps a Socratic midwife might help a science communicator to be born, but most of the work is not being done by the midwife.
I have been facing some of these issues vis-à-vis civilization, since the word used as a noun was introduced during the Enlightenment, when it primarily had moral and religious associations, and it has taken a long time, and appropriation by archaeologists and others, to be able to use the word without being assumed that one is talking about moral and religious conditions of society. I introduce concepts all the time in my attempt to get closer to saying something precise about the nature of civilization.
Last week I was thinking about Karl Löwith’s philosophy of history (which Richard Wolin called Löwith’s “emphatic rejection of philosophies of history”), and I realized that, implicit in Löwith’s critique of secular philosophies of history as counterfeit versions of earlier religious concepts was an assumption of pristine concept formation. For Löwith, modern philosophies of history were, essentially, fake, because they were traditional religious concepts of history stripped of their original religious meaning and pressed into service as though adequate despite their deceptive appearance. Moreover, the philosophers who used these concepts didn’t even know what they were doing, so they were fools in addition to offering fake ideas of history.
It would seem that, for Löwith, for a concept to be legitimate, it must be introduced de novo, and not borrowed from any previous source. Thus a legitimate concept is pristine. How closely can pristine concept formation be approximated in actual cognitive experience? Apparent pristine concept formation may in reality involve multiple layers of self-deception, in which one is borrowing a familiar concept and applying it to a familiar object, while concealing this familiarity from oneself, believing that the concept of pristine and the that object is new to experience.
Löwith’s critique can be extended backward in time to impugn the religious concepts that Löwith presented as though being authentic and pristine. Jessie Weston in her classic From Ritual to Romance argued that much in medieval mythology was due to earlier pre-Christian mythological material that had been given a Christian whitewashing by medieval poets. This idea, once controversial, is now accepted and we can push this as far back in time as we have records to make the argument. Few if any concepts are pristine. If we are to believe Ecclesiastes, there is no new thing under the sun.
This takes me back to an old idea of mine that I have never written up, but occasionally I return to in my imagination. The earliest ideas would have started to form in prehistory, and were handed down though successive languages and communities and eventually civilizations after civilizations formed. And these ideas are still evolving and getting passed down through history to us. Identifying any one period in the development of an idea as its pristine and authentic state is, in this context, misleading to say the least.
Let’s take the idea of sainthood. The medieval Catholic church formalized the criteria for sainthood, and formalized the process for officially recognizing individuals as saints, but the idea of an individual being “saintly,” that is to say, being especially holy, being closer to the divine than most other human beings, is most certainly an ancient idea that was appropriated by Christians and formalized within the context of Christian doctrine. If we take a narrow conception of sainthood, it would be anachronistic to call anyone a saint prior to the procedures for officially recognizing sainthood. In other words, there were no, nor could there have been any, saints in prehistory, or even in early antiquity.
Now suppose some far future time when humanity still exists but it is after the Christian Epoch and there is no longer any institutional Christian church. Under these conditions, and given a narrow conception of sainthood (which we might also call a formal conception of sainthood), there could no longer be any saints. But we know that human experience continues even when institutions change or go extinct, so that there would still be individuals who lived in a saintly manner. Thus I see the idea of saintliness as transcending the formal procedures of any particular institutionalized church for defining or recognizing sainthood. Saintliness is the form that this perennial form of human experience is given during the Christian Epoch, but the same human experience and human behavior was to be found prior to the Christian Epoch, and will be found after the Christian Epoch, though we may not call it by the same name, and, obviously, the institutional formalities of the past will be irrelevant.
From this point of view, Löwith is simply insisting on a narrow conception of particular concepts in the philosophy of history that were used in a religious context, but which later came to be used in non-religious contexts (or, more accurately, used in not specifically religious contexts), and we can posit that something like these concepts were also present prior to their first explicit use in religious contexts, although likely in a much less developed and sophisticated form.
Even when a human experience is perennial — like the experience of sainthood — later instances of the experience will be colored by conceptualizations of earlier instances of the experience. There is a certain retained conceptual complexity, as well as a growing body of narrative that surrounds the concept. One might further postulate that a concept that grows in this way might become too large and unwieldy, and eventually there would be a need to streamline and simplify the concept or it would become too sophisticated for ordinary use. Concepts that fail to undergo this simplification process on a periodic basis become dead concepts, analogous to dead languages. Dark ages, when the conceptual resources of earlier ages become inaccessible or frankly incomprehensible, thus serve as an historical opportunity for the resetting of overly-complex concepts, which undergo a forced simplification due to epistemic impoverishment.
One takeaway from the above is to distinguish, as far as possible, between perennial concepts and concepts specific to a particular place and time, i.e., specific to a particular cultural milieu. There is no reason to insist that all concepts are perennial (the position of Ecclesiastes, narrowed to a doctrine of concepts) nor to claim that all concepts are peculiar to a given social milieu. Perennial concepts might also be called transcendent concepts, because they transcend any given social milieu, so that we could call particularistic concepts immanent concepts. Transcendent concepts evolve; immanent concepts are introduced. Our conceptual framework is a mixed bag of transcendent and immanent concepts. The transcendent concepts preserve some continuity to with the human past, while the introduced immanent concepts define our distinctive milieu.
Directional development of the conceptual framework comes about with what we may call sempiternal concepts, which are concepts that, once introduced, because a permanent possession of the human conceptual framework — or, at least, a semi-permanent possession of the conceptual framework. The conceptual framework, like a Schumpeterian economy, is always in a state of creative destruction, with new concepts being introduced and old concepts being eliminated. But there is an ongoing tension, though which perennial concepts continue to be re-introduced in simplified forms when some great disruption seems to purge them from the conceptual framework. The conceptual framework as a whole, and not only individual concepts, can become too large and unwieldy, and it too must be simplified. In a disruptive simplification (i.e., a dark age) a lot of immanent concepts are lost, some are transformed into newly simplified sempiternal concepts, and most perennial concepts undergo drastic simplification.
