Society and Knowledge

V. G. Childe

• Publisher: Routledge
• Year: 2021
• ISBN:
  9781000516913
• Edition: 1
• Topic: Philosophy
• Subtopic: Social Philosophy

Originally published in 1956, the well-known archaeologist here takes on the role of philosopher. The author argues that knowledge is a social phenomenon, and that our intellectual life is the product of social heritage: reality is the product of different opinions of various societies.

VII. The Categories as Constructional Tools

TO EXPLAIN the method of filling up gaps in experience and anticipating future developments, I must have recourse to another analogy. I propose to compare the reproduction of an ideal world of knowledge to the rebuilding of an ancient monument, say a ruined Saxon church, by a team of archaeologists, architects and craftsmen. Most of the material is there already: stones tumbled from the walls, window frames, capitals, voussoirs, roofing tiles are lying about the site, chipped and battered in disorderly heaps. These it is planned to re-use and replace in their original positions. But pieces are missing; some stones have been carted away, others have crumbled into dust, but many are just buried in the debris and will come to light as this is removed and replaced in its former position. Immediately the missing parts will have to be replaced—at least temporarily—by substitutes of concrete or similar material suitably shaped. Some of the wall foundations and column bases are still in place and suggest the outline of the ground plan, but, as work proceeds, fresh wall footings may be exposed under tumbled stones and may impose modifications of the suggested plan. The archaeologists know from other churches of the period that survive in better preservation, the general design and superficial appearance of the coniplete edifice. The details are, however, totally unknown; no one knows how many windows were on each side nor exactly how the roof was constructed in this individual instance. That they hope to find out on the one hand by recovering the mouldings and springers, on the other hand by experiment. For they plan by actual building to find out by trial and error how, within the limits set by their general knowledge of this church’s plan and the common design of such churches, the surviving fragments can be fitted together into a stable edifice. As it proceeds, the work will reveal gaps caused by missing parts and these will have to be filled by substitutes. A scaffolding will be needed to support window sills or capitals till the actual walls have been raised high enough to receive them and at the same time to guide the, masons and others co-operating in the enterprise. The scaffolding is in fact the skeleton of the design, provisionally outlined by the architect in the light of the archaeologists’ deductions. But it must be flexible since the provisional design may have to be modified to accommodate the surviving fragments which must all be fitted in, to ensure the structure’s stability, or in the light of discoveries made in the course of the work.

Let the surviving pieces of the ruin represent the perceptual data in our ideal reproduction called knowledge, and let the church as built in a.d. 950 stand for the structured external world it must reproduce. Then, just as in the church the missing parts had to be replaced by substitutes shaped like surviving structural elements, so the gaps in pooled experience must be filled with symbols, expressing ideas imagined on the model of ideas, already socially approved and objectified. As the architect had to anticipate the design that the original and substitute pieces are to form and to lay out a scaffolding accordingly, so knowers must anticipate the uncompleted pattern of reality in imagination, using partial patterns already known as a frame to support their hypotheses. But this frame must be flexible since hypotheses must be modifiable in the light of practice.

The imagined ideas or fictions, used as substitutes for empirical jiata, are all too familiar. They are expressed by symbols which, as often as not, have no referents and so cumber all languages with words that may have no meanings and yet look deceptively meaningful. But many are, or have been, really serviceable, and all were probably once necessary. A whole host of eminently respectable symbols used in science, as in everyday life, were produced to express imagined meanings, and have only later acquired referents as a result of experiments, i.e., actions, that could not have been planned without their aid. Others that were likewise useful for that purpose have been discarded for terms whose meanings correspond better to reality. Other symbols were invaluable for the communication of information and for description and are still found convenient for that purpose even though they prove to have no referents and even after better descriptions have been found. Many electrical phenomena, particularly conduction, were found to be quite satisfactorily described in the terms applicable to flowing liquids. The term, “electric fluid,” current, or shortly, “electricity,” was thus coined to form the grammatical subject in descriptive sentences which yet conveyed useful information, i.e., knowledge. Though we know now that there is no fluid flowing along copper wires so that the symbol strictly has no referent, the term is as we saw already still found useful in practice, and its reference remains a convenient idea.

Psychoanalysts have created a whole pantheon of mythical entities—the Unconscious, the Censor, the Ego, the Libido . . . It is doubtful whether these symbols have any better referents than $\sqrt{-1}$. But like the latter, they are convenient and useful in communication and in action. With their aid, observed phenomena can satisfactorily be described, and rules for action have been deduced that have undoubtedly relieved or cured neurotic and psychotic patients.

“Centaur” may once have been just as useful. It was no more a creature of the imagination than the “electric fluid,” and no less. As the idea of an “electric fluid” imagined as water was created to fill a gap in experience, so was that of combined man-horse. It may have served a prehistoric peasant society, unfamiliar with the use of horses for riding, as a convenient symbol to denote the first mounted warriors it encountered and to communicate information about these unwelcome strangers. But by Homer’s time the centaur was already a mythological creature. In this case it is not only the failure of humanity in three thousand years to find an instance of “centaur” empirically that has induced society to abandon the word as a symbol describing reality; a fuller appreciation of the biological and psychological characters of men and horses shows such a combination to be inconceivable. In other words, once the connotations of “man” and “horse” respectively are adequately understood, it appears that “man-horse” can have neither connotation nor denotation; “centaur” is literally meaningless with neither reference nor referent.

It is needless to dwell on the hosts of demons, ghosts, spirits, magic forces that have plagued humanity in the past. All, like the electric fluid or the Aether of Space, were created by imagination to fill gaps in experience as substitutes for missing structural parts without which the rebuilding could not proceed and rational action would be impossible. None were created out of nothing; all were imagined on the analogy of authentic perceptual data though those data may have been distorted by illusions and delusions. All have been, or still are, objective in as much as a social convention has invested the symbol with meaning, and some society by incorporating the word in its language has imposed the idea on its members. They are, or were, objective too in that society, believing in the reality of the fiction, behaving as if it did form an element in the external world and obliging their members to act accordingly; otherwise co-operation with their fellows is impossible. If a society’s ideal reproduction of the external world be built too largely of imagined substitutes without referents, it may be impossible to correct its errors as long as the society holds together; for its members will be literally incapable of perceiving data that do not fit into the design of the edifice! But the trouble is more often due to a too rigid scaffolding—to a misuse of categories.

A category connotes the outline of a pattern, the kind of relation holding between elements in a pattern that is itself presumed to be a component pattern of the external world. If not perhaps under this title categories are familiar to every reader—I mean “space,” “time,” “causality,” “substance,” and so on. Each denotes a way in which empirical data are supposed to hang together to form a pattern and the kind of pattern thus formed. They are clearly very abstract ideas indeed. I have compared them, not very happily I fear, to the scaffoldings used in rebuilding the Saxon church to guide the builders and to support the walls during their re-erection. They follow the plan of the church as disclosed by the surviving wall foundations, but, it will be remembered, these were only partially exposed and the clearance of the debris may reveal a rather different plan. At the same time the scaffolding might enable the builders to anticipate where part of an arch or other structural element might go provided suitable supports for it were subsequently found and fitted.

It is just such an anticipatory function that a category should serve in building up a serviceable ideal reproduction of the external world. It is used in the belief that this outline at least of the latter’s pattern will persist and preserve its identity in the interval between knowing and doing. The total pattern is changing at every instant, but component patterns may continue to follow the same outlines without practically significant deviations. In so far then as the ideal reproduction follows such a pattern’s outline, it will for practical purposes continue to correspond to the pattern of reality. For instance it is plausibly assumed that the constituent parts of the external world are and will be arranged spatially or “in space” though of course the actual arrangements of such parts are ever changing. This assumption yields the category of space. So causality assumes that there will be a pair of events, A and B, such that B “invariably” follows A.

Philosophers have maintained that the categories are necessary preconditions of knowledge, not derived from experience but given a priori, and existing eternal and immutable in their own right. If the first claim means that the external world exhibits a pattern, it is just a restatement of what was said on page 63 and must be endorsed. If it mean that knowers must somehow know the outlines of patterns before they can know their contents, it would seem to invert the sequence. Yet just this is implied in the assertion that the categories are given a priori. Finally the alleged immutability of the categories is rebutted empirically by the history of thought; categories have changed during the course of written history as will be illustrated shortly. On the other hand, these philosophers are right to this extent that the categories are not given as such in private experience. Patterns are perceived, their abstract outlines have to be discovered. Perception is a private affair; the discovery of the general outlines of patterns contained therein is the result of social co-operation, and it is society that objectifies the categories as thus discovered. These points must be amplified and clarified by two concrete examples-—space and causality.

For Kant space and time were the a priori preconditions not only for knowledge, but even for perception, the raw material of knowledge. And from the concept of space all the propositions of Euclid’s geometry could be deduced a priori. This curious belief was due in no small measure to an atomistic psychology assuming that the several senses—sight, touch, etc.—supplied discrete messages to the (individual) Mind—a fiction—that then somehow combined and arranged them in a pattern. It is the merit of the Gestalt school of psychologists to have shown that experience does not present a lot of disconnected bits but a pattern; objects are apprehended as forms (Gestalten). Men, and presumably animals and even insects, behave as if they perceive things and events as in front or behind, above or below, to right or to left, before or after.

Nevertheless this private space, perceived as occupied by things, by the objects of action and desire, is not the category of space. It must be objectified by society, and co-operative action must gradually reveal its properties, i.e., how things can be arranged spatially. Space as a category is not that in which things are perceived, but that in which members of a society co-operate and act together on things. In such social behaviour the private spaces, perceived by individuals, are welded together, and transformed into a public space, an idea transcending all society’s several members and imposed upon them. By themselves “above” and “below” are meaningless symbols; in co-operative action they acquire both references and referents, and are given context. In a cave dwelling of the Old Stone Age the children had to learn where they should sleep just as in the collective hunt or ritual dance each adult must occupy his appointed place. In such activities a common spatial world is constructed; the spatial pattern is realized just in moving to take up the socially designated positions and occupying them. Thus men gradually discover by experiment how things and persons can be arranged spatially, so defining an idea of space. As such it must find a symbolic vehicle and be expressed.

The verbal symbol, “space,” in English, is a convenient expression like “electricity,” but it too is a dangerous one and may mislead. The category may get turned into a “thing” -and be hypostatized. We talk of “measuring space,” but what we really measure are always fields, or strips of rope or lines on paper, that are spatially extended. The word inevitably suggests some sort of frame into which things have to be fitted. It should mean rather a kind of pattern that men can make if only symbolically. In this correct usage the symbol’s meaning has demonstrably changed in the course of history. A radical transformation of Kant’s eternal and immutable category is not much over a century old: Euclid’s three-dimensional space has grown into the n-dimensional space of Lobochevski andRiemann. More recently Relativity has familiarized most people with the latter idea or at least its symbolic vehicle. If few people understand the concept, most will admit that it has been triumphantly justified by successful operation.

Causality (cause and effect) is a very reputable category, but proves no more immutable and still less a priori than space. It was regarded by Kant as the presupposition that made science possible. It has certainly proved a very serviceable scaffolding within which the whole edifice of mechanics, and physics since Galileo has been built. But the scaffolding has been bent by the weight of accumulating data it supports and to accommodate unexpected discoveries. And now the distortion has become so acute that it has to be replaced by a fresh scaffolding. Atomic physicists have abandoned causality in favour of mathematical probability. They have been induced to adopt this drastic expedient not by the criticisms levelled against the old category by logicians, but to accommodate fresh empirical data. Their reason suggests that the category of causality is based on experience rather than its prerequisite.

Of course causes of the rather mystical kind excogitated by metaphysicists can be perceived even less than metaphysical space. But action presents every sentient being with a pair of perceptible events that are “causally” connected. By exerting your muscles in pushing against a movable object—an event perceived with the aid of muscular and tactual sensations—the object changes its spatial relation to you and other fixed objects—a second event disclosed by vision. This is the basic perceptual pattern the outline of which, conceptualized and,socialized, becomes causality. It might even acquire at the perceptual level the element of “necessity” that was supposedly essential in the category until probability proved a more serviceable idea. This primary pattern, given in private experience, is objectified as soon as it is used as a guide in the distinctively human kind of action represented by the manufacture of tools. The palaeolithic flint knapper not only must have perceived that when he delivered a suitable blow on a flint nodule a flake came off, but he also behaved as if he confidently expected a similar blow on another nodule to be followed by the detachment of a second flake. Now, the act of striking a flint nodule is itself a pattern—a pattern of the kind termed “causal.” The operator knows the pattern, if not before or even after, in the act itself. Our operator’s confidence was inspired not only by his own memories of successful repetitions of the feat, but still more by the social endorsement of his expectation. For, the reader will recall (page 11), every tool is a social product, and this flint knapper, like every other, had been taught by his society what to make and how to make it. In other words, society had transmitted to him, whether by precept and example or by the latter alone, the idea of the flake to be produced and the idea of how to make it. In his successful act the idea was realized. In other words the correspondence between the ideal reconstruction and the external world was re-established. Therewith the correspondence between the causal pattern as conceived socially and a pattern of the external world was publicly vindicated. Collective experience thus converted the successes of repeated individual experiments into a necessary consequence of socially inculcated actions. The outline pattern discovered and justified in operation becomes a category, a tool for planning further actions.

If generalized into a rule, “Every event has a cause,” though long before any such rules were formulated, causality thus conceived inevitably engendered a horde of fictitious beings—gods, spirits, demons ... to act as causes. It is indeed still replenishing language with their symbols—”gravity,” (“the pull of gravity”), “chemical affinity,” “the libido” are just attenuated descendants of Jove and Juno! Of course the category has been refined and depersonalized by generations of philosophers and logicians. Even so as Ritchie remarked in the Natural History of Mind (1936) “a nasty flavour of pushing and pulling clings to causality.” In any case the refinements have been due as much to craftsmen and technicians as to philosophers. In the Old Stone Age the only way men could move things was to push and pull them with their own muscular energy. Neolithic men had domesticated oxen, asses and horses and could make them do some of the pushing and pulling. Scarcely two thousand years ago Greek engineers designed water mills for grinding grain, and therewith men began to control and use for the first time an inanimate motive power. In the last thousand years the Greco-Roman water mill has given birth to varied progeny of complicated machines actuated by inanimate motive powers (water wheels, steam engines, etc.) and applying rotary motion to the performance of an immense variety of repetitive operations formerly executed by hand. Familiarity with machines that European societies could construct and operate completed the transformation of the personal: or animistic category of cause into a mechanistic one that since Newton has done such signal service in physics. The outline pattern, imagined on the analogy of machines that men do make but far outstripping human ingenuity to realize, is itself just a vast machine whose motion, being cyclical because all the essential parts just revolve, produces only change in the relative positions of the components, but no really novel configuration. The latest transformation of the “eternal” category in turn is similarly related on the one hand to the invention and construction of electronic machines that are no longer cyclical and on the other to the creation of symbolic-vehicles-for intellectual tools, e.g., for expressing, and operating with, mathematical probability.

Besides its liability to personification, causality suffers from a further defect. “Events” as perceived are seldom simple, but can be usually analysed into a whole train of events. To the lazy motorist it is a push on the starter button that “causes” the engine to “go.” Any mechanic can recognize a battalion of complex processes in delicate mechanisms intervening between the pushed button and the firing cylinders, and a scientist could identify a regiment of subtler reactions behind them! Again even the simplest member of a causal sequence may appear complicated by irrelevant accidents. In trying to make flint implements from a nodule a modern archaeologist often bashes his thumb and adds the utterance of an imprecation to his manual activities. Very likely the Stone Age flint knapper did the same, but perhaps he mistook his expletive for an element in his causal activity as efficient as the movement of his hands. If he transmitted this belief with his practical lore to his apprentices, if, that is, society endorsed his illusion, the imprecation became a magic spell, as essential a “cause” of the finished tool as skill and muscle. Be that as it may, among illiterate societies today effective and rational productive operations, for instance in fishing, in potmaking and in iron smelting, are habitually accompanied or preceded by what we regard as superstitious rites—abstinences, the recital of spells, even bloody sacrifices. These barbarian tribes behave as if they made no distinction between the manual and the ritual action, but regarded both, indissolubly united, as the “cause” of the desired result.

When scientists justifiably boast of their success in discovering causes, they generally mean disentangling an efficient cause from such irrelevant accidents. Still, exponents of every branch of science stop disentangling somewhere— when they have found experimentally the kind of event within the limited field of their particular discipline that “invariably” precedes the particular phenomenon they wish to produce. Causal connexions do correspond to outline patterns of the external ...