Work in Progress: Observational and Theoretical Terms
Friday 29 April 2022
A couple of weeks ago I ended Work in Progress: The Concept of Basal Civilization by suggesting a way to get from a conception of civilization based on cities in relationships of cooperation, competition, and conflict to a conception of civilization in terms of its institutional structure. Another way to approach this division offers further clarification, and allows me to further articulate my conception of civilization in a context that is consistent with existing philosophy of science, as well as pointing beyond contemporary philosophy of science to ways in which science can be expanded and its concepts extended.
It is commonplace in contemporary philosophy of science to distinguish between the observational terms of a scientific theory and the theoretical terms of a scientific theory. Applied to a science of civilization, we would expect such a science to have observational terms and theoretical terms. My definition of civilization in terms of cities in relationships of cooperation, competition, and conflict is an observable idea, while my definition of civilization in terms of institutional structure as an economic infrastructure joined to a conceptual framework by a central project is a theoretical idea. A city is something that can be observed in the real world; an institution manifests itself in many ways in the real world, but you cannot observe it directly. If you try to observe the institution of the legal system, for example, and you observe courtrooms, you aren’t observing the institution itself, though you would be observing an aspect of the institution.
Thus given the distinction between observational and theoretical terms, city is an observational term in a theory of civilization, while institution or central project are theoretical terms. There is an extensive philosophical literature on the distinction between observable and theoretical terms; the distinction in its simplest and most rudimentary forms probably cannot be sustained. However we construe the distinction, making it puts us in the right ballpark and allows us to elaborate such a theory within a known scientific (and philosophical) framework.
One can later choose one’s preferred philosophical interpretation of the observable/theoretical distinction, and presumably it will be that formulation of the distinction that best suits the subject in question. That is as it should be. There is no a priori obligation that one and the same distinction should characterize every form of knowledge. Moreover, what is observable in one theory might be theoretical in another theory, and vice versa. We find this in logic all the time in terms of definition: what is in one logical language an undefined primitive term is, in another logical language, a derived term defined by logically antecedent primitive terms. In fact, we do this with the observational/theoretical distinction itself: if we define what it is for a term to be observational, then we can define theoretical terms as non-observational, or vice versa.
Identifying city as an observational term of a science of civilization also gets us in the right ballpark for the degree of abstraction of our observables (or, if you prefer, level of generality) that we would want for a big picture science like a science of civilization. I have remarked elsewhere that many books on cosmology identify galaxies as the “basic building blocks” of cosmology, and in this sense galaxy is an basic observable term for cosmology, but if you’re working in stellar evolution and astrophysics, the level of abstraction of cosmology is inappropriately abstract, so you have to descend the ladder of abstraction to something more concrete, like stars, or star systems.
Also, ultimately, you will need to cash out your cosmological explanation by going into the details of how galaxies are formed and how they evolved, and this will mean going into the details of stellar evolution and astrophysics, but while on the elevated plane of cosmology, what is most concrete and observable is the galaxy as a unit of cosmology, and it is left to those who delve into details to tidy up a cosmological theory by showing that is holds not only for galaxies but also for the stars that make up galaxies, and the matter than makes up stars.
One could similarly say that the cell is the basic building block of biology without also maintaining that the cell (or the study of the cell) is the whole of biology or the whole of life. It is not, but it is the basic building block of living things, so it is a good level of abstraction to take the cell as the building block of biology. Ecology might take the individual organism as the basic observable of its theories, and leave the details of these individual organisms to biology. It is in this way that we can think of the city as the basic building block of a science of civilization: an individual city is not coextensive with civilization, but it is the basic observable unit, and civilizations are, for the most part, built up from cities. The odd and exceptional cases — for example, large nomadic groups like the Mongols conquering and ruling (for a time) settled urban civilizations, or early medieval Europe, based more on the unit of the manorial estate than the city — can be explained as deviations from the model of city as the basic observable of civilization.
In the case of conquering nomads, we can observe that every successful nomadic conqueror was eventually assimilated to the social and cultural norms of the urbanized peoples they conquered. Goths, Vandals, and Huns wanted to become Romans even as they destroyed Roman civilization; a Mongol became the emperor of China, and by this time the Mongols had been integrated into Chinese society, bureaucracy, and cultural institutions; the Turks after conquering eventually became the Ottoman Empire, with mostly Persian civil servants running their empire for them, and then they eventually became the Sick Man of Europe.
In the case of the manorial estates (and monasteries) of early medieval Europe, we might note that these institutions took over most of the functions of cities and preserved what could be preserved of the urban civilization of Rome under very different circumstances. We would also want to note that the civilization of early medieval Europe was a very different kind of civilization from the epochs of Western civilization that preceded and that followed it, so that the substitution of manorial estates and monasteries for cities was not without consequence, and the basic unit of civilization being different meant that the civilization itself was different, though still within the boundaries of the concept of civilization.
The concept of a city is precisely the kind of “observable” that would cause problems for philosophical theories of observables, because it isn’t an observable in the same way that a cell is an observable, though the two are nicely analogous, because when we dig into the details of the function of a cell it is enormously complex — just like a city. At the level of biological thought, it is a convenient unit, just like at the level of civilizational thought, a city is a convenient unit. But if we pursue this further we get entangled in a web of inter-related philosophical problems like phenomenalism, apperception, theory-laden observation, and the further distinction of primary and secondary qualities. This is why philosophical questions can never stop short of metaphysics and totality, because each question leads to another question, until we are forced to make fundamental statements about the nature of reality as a whole (which is what metaphysics is). Science arrests this process by settling upon certain conceptual conventions that allow the process of philosophical inquiry to converge on good enough concepts so that empirical inquiry employing these concepts can go forward.
Jerzy Giedymin in his paper “On the Theoretical Sense of So-Called Observational Terms and Sentences” (in Twenty-Five Years of Logical Methodology in Poland, p. 111) characterizes observable terms in this way:
“Names of colours of things (macroscopic objects) are given as examples of observational terms: red, green, blue, etc.; names for their shapes: round, triangle, with a sharply outlined contour, etc.; relational names: left from …, congruent with …, longer than …, etc. Simple sentences (without quantifiers), containing only observational terms among their descriptive terms and negations of these sentences, are given as examples of observational sentences: ‘This lemon is yellow now,’ ‘The Eiffel Tower is higher than the building next to it,’ ‘The pointer of this instrument is now indicating zero,’ etc. Observational expressions are characterized according to the following definitions: A term M belongs to the observational vocabulary if, under suitable conditions, it is possible to decide (definitely), on the basis of (only) direct experience, whether the term M applies to a given situation or not.”
Now, Jerzy gives this above formulation only in order to use it as a stalking horse for the remainder of the paper, which criticizes this naïve conception of observables, but, as I said above, this gets us into the right ballpark, philosophically speaking. We could say, of a proposed science of civilization, that the term CITY belongs to the observational vocabulary because, under suitable conditions, it is possible to decide on the basis of direct experience, whether the term CITY applies to a given situation or not. That there are theoretical presuppositions of the idea of a city (and there are), and that some observations will embody systematic ambiguity, do not prevent us from using this as a rule of thumb which generally holds at the level of abstraction we are employing in our theory.
Given the distinction between observational and theoretical terms as applied to the study of civilization, we can move forward with more confidence, despite the philosophical problems which plague all suitably thoughtful formulations of science. In my case, the distinction allows me to integrate two distinct definitions of civilization in a single theoretical context composed of both observational and theoretical terms, each of which have their appropriate place, so that, in a sense, each justifies the other, rather than the two definitions being in tension with each other (even if it is a creative tension, from which there is much to be learned, you can be sure than someone will point out that you have used more than one definition for the same thing).
In all honesty, not many people will know the background theoretical problems of observable and theoretical terms, so if you get a shallow and unthinking criticism of your theory, you can silence a lot of this by just throwing out the line, “Well, that’s a theoretical term of the theory, so we wouldn’t expect it to directly relate it to observable terms. The observational terms I am taking as basic to this theory are…” and then you throw out the observational terms you’re most comfortable with, and it is likely that you will have strangled this irritation in its cradle. Your interlocutor will probably be happy to shift to some more concrete discussion of observables rather than get lost in the weeds (and risk being embarrassed).
A lot of doing science, especially founding a new science, is simply to get others to accept your presuppositions, if only hypothetically, until you get to the point that you can have a profitable and fruitful discussion about substantive issues, rather than going around in circles in a vacuous debate over terminology. Sometimes one must get creative in get to the substance, especially if your audience is unsympathetic to what you’re doing.
In my concrete definition of civilization as geographically co-located cities in relationships of cooperation, competition, and conflict, city is an observational term, and we can not only observe cities in existence at present, but archaeology allows us to “observe” defunct cities of the past. Showing the other terms of this definition to be observables would force us to adopt a rather liberal and expansive conception of observation, but if that’s what it takes, then that’s what it takes. For example, how do we “observe” trade and commerce in the historical past? Insofar as trade is an activity, and not an artifact, it poses problems. But cities, too, are as much function as they are structure, but it is the structure that we dig up and observe by means of archaeology.
If the historical record is well populated, it would not be difficult to convince skeptics that trade between cities took place, especially if there are written records of commercial transactions. But if we are studying the initial emergence of cities, and then the civilizations that supervened upon these earliest cities, we will not have written records.
There is a lot of interesting work that has been done on Neolithic trade networks. It is sometimes possible to determine the exact source of obsidian, for example, and to show that obsidian from a specific source is found on sites along a route, and that route may lead hundreds or thousands of miles from the original source of the obsidian. Where sea travel is practiced, trade goods may travel long distances without leaving any traces of intermediate stops. Shipwrecks like the Uluburun wreck have been incredibly valuable for the reconstruction of early trade networks, as they show us what commodities were stocked in the hull in trading quantities (and in standard units for trade), and the site of the wreck indicates some point between point of origin and intended destination. (In 1993 I had the good fortune to see the display on the Uluburun shipwreck at the Bodrum Museum of Underwater Archaeology in Turkey.)
A great deal of inference goes into reconstructing trade networks, and at some point we may ask when the evidence for trade is sufficiently strong that we would say that we “observed” that trade had taken place. Also, trade can be an instrument of cooperation or competition. A great deal of inference also goes in to determining past conflict. The site now identified as Homer’s Troy has been shown to consist of many layers of occupation (nine layers the last time I read about it), several of which show signs of violent overthrow (burned timbers, collapsed houses), presumably as the result of conflict. Similarly we could ask, at what point does the evidence rise to the level of retrodicted observation?
And it may well be that the evidence for cooperation, competition, and conflict is not any better than the evidence of some institution that we have identified as a theoretical term in our theory. This would seem to call into question the observational/theoretical distinction, but if we get to this point, we will already have thought this through pretty carefully and will have something to show for our efforts. In such a case, the observational/theoretical distinction becomes like a ladder that we pull up behind us after we have climbed it (to appeal to an image made famous by Wittgenstein), and we can forgo it and turn to new and no less interesting problems at that point.