Finn Sivert Nielsen Anthropologist
Home Search Contact På norsk

In Worlds of Mirrors

Models of Complex Societies and Anthropological Complexity

by Finn Sivert Nielsen

 Table of Contents 

Chapter 4. Text to Texture

Recycling bits of Ricoeur, Bateson, and Marx


At a remarkable seminar series at the University of Oslo in 1982, Tord Larsen introduced what has later become a catchword of Norwegian anthropology, the so-called "problem of translation" – "oversettelsesproblemet". The "translation problem" might be viewed as a local version of the international debate on "rationality" (cf. Wilson, Winch, Lukes etc.), but it has a wider field of application, in that it links problems of meaning to problems of social and institutional organization, economics, morality, and personhood. In both cases, however, the basic question that is asked relates ultimately to the problematization of "language" that was formulated by Wittgenstein in the 30's, and operationalized in the social sciences by such writers as Foucault and Derrida in the early 60's. "Poststructuralism" reached anthropology a decade later, and the "rationality debate" and the "problem of translation" were early explorations of its implications. Soon after, however, these debates were swamped by "postmodernism" which eclipsed many interesting questions that the earlier debates raised.

      The "problem of translation" was a practical one, and the answers Larsen proposed to it were pragmatic and practical. Can we translate between two cultures? Relativism denies it, universalism affirms it. But we know, experientially, that both assertions are equally absurd: sometimes translation is to some extent possible, sometimes predominantly impossible; and seldom, if ever, do we even approach the extreme forms. Communication takes place not in the domain of logic, but in the real world – not by means of "language" in the abstract, but by means of specific "language usages". We may speak of processes of "immunization" that increase the isolation of one language game from others, and processes of "mediation" that open or assert connectivity.

      But anthropological theories are themselves language games, thus we might expect the "translation problem" (and attendant processes of immunization and mediation) to play themselves out in their mutual relationships, as well as in communication between "real world" cultures. Larsen focused particularly on the see-saw movements between "formalist" and "substantivist" theories in a range of anthropological contexts, e.g. economic anthropology, ethnicity and kinship studies, and on totalizing diachronic schemes such as evolution and adaptation. A latter-day version of the same "dual opposition" is found in the dichotomization of practice theory and deconstructionism, in which the former attempts to "found" sociality in material content (body in environment), while the latter seeks to dissolve it into formal relativity. The opposition between these "schools" is rarely expressed in explicit terms – and, indeed, in the work of Wittgenstein, Foucault or Derrida, it is not an opposition: Wittgenstein deconstructs language to arrive at a concrete (materialized) situation. Nevertheless, in anthropological theory the oppositions are real, there is a real "problem of translation" between practice-oriented and deconstructivist theories. Though anthropologists such as Wagner and Bloch may pay homage to each other's positions, their epistemologies are in fact nothing but jazzed-up versions of formalism and substantivism. They can commune with each other, not as opposed analytical "positions", but through a constantly reformulated process of give and take. As Larsen recognized, the fundamental challenge in any dyadic debate of this kind is therefore to weave a pragmatic, median road between the extremes, strive to breach the absolute discontinuity without erasing it. To complicate simplicity is to "translate". Thus we can only hope to operationalize the unity of epistemological intent that underlies both practice theory and deconstructionism through protracted experience with concrete social analysis.

      I shall not attempt this project here (but see Nielsen 1998), but this essay grows out of a similar concern. For several years after Larsen's seminar, I worked on an operationalization of parts of his general scheme, which I later sought to apply to data collected during fieldwork in Russia in the late 70's and early 80's (see Nielsen 1987). The following is an attempt at lifting the teoretical structure I evolved in this empirical analysis, and letting it speak on its own terms.

      I shall refer to the type of complexity here described as textural complexity. The idea is a model similar to Larsen's, with two intersecting variational continua ("immunization" and "mediation"), stretching between four polar terms: the triadic model from the preceding chapter is thus superceded by a quaternary basic form. The two continua weave a shifting, billowing pattern through reality, they model its "texture". Where warp and woof are thick, there are great differentials of power, where they are thin, power is weak. Where warp and woof are dense, reality seems fixed, and objects, figures (or, as I shall call them here: "centers") seem to congeal spontaneously out of relationality. If the weave is open we assume that there is "nothing there", or that whatever is "there" is irrelevant chaos, rather than constructed order.

      In this conception, the fundamental problem is that the subject is eclipsed. As in the structuralist model (see Chapter 2; and to an even greater extent, in deconstructionism), the stability of experience converging around a center is renounced in favor of a relativistic stance. In the following I seek to avoid this problem in two ways: by proposing (a) a theory of "centers" (fixed entities congealing out of textural relativity), and (b) a theory of "environment", the "infrastructural" base that sociality is embedded in. As I myself see it, I have succeeded fairly well with the first theoretical patch (which asserts that "content" is derived from "form"); but the idea of "environment" remains at an embryo stage. This blind spot, I often think, was not a result of insufficient analytical application on my part, but intrinsic to the formal structure of the theory of "texture" itself.



How do we as anthropologists link the present to the past, the individual to the collective, the material to the spiritual? I would like to approach these questions from two points of view simultaneously, each of which seems to have great potential for subtle descriptions of complex system dynamics. On the one hand I shall discuss cybernetics, as it had been presented in anthropology through the work of such authors as Gregory Bateson, Roy Rappaport, and Fredrik Barth. The cybernetic approach is highly processual and dynamic, and perhaps for this very reason has a problem addressing issues such as force or habit. These questions, in contrast, are the hallmark of marxian theory. Change, here, is not a continuous flow through cybernetic loops. Marx devised a structural dynamics that describes history as a jostling of institutionalized aggregates against each other, like icebergs on a stormy sea. This convulsive imagery is fraught, not only with violence, but with "inertia", indeed, the two are often equated: history is a continuity of reified power. I shall ask what would happen if one were to integrate the lightness of Bateson with the ponderousness of Marx. Might not the result be an analytical instrument uniquely adapted to the complex world we inhabit?


Text and Dialogue


Society is enacted. But as Roy Wagner (1986) reminds us, it is also perceived – and when large social units change chaotically it seems increasingly imperative, and increasingly problematic, for the individual to maintain an adequate perception of the world. Society may then become (and indeed often is, even in tranquil social states) inscrutable to the actor, a landscape of a "quest for meaning", rather than a database of "resources" used to solve the practical and spiritual problems of life. Culture is in this sense not so much enacted as "read" and "interpreted", as a text.

      Paul Ricoeur (1971) highlights the difference between speech and writing, dialogue and text. The dialogue is realized in the fleeting present, it refers back to the subject speaking and reaches out to a concrete addressee. Its most basic reference is to the situation in which it occurs – when we talk, it is always "about ourselves" – about the relationships we participate in here and now. The text, in contrast, escapes the situation. It extricates itself from time and space, since it can – in principle – be read and interpreted by anyone at any time and anywhere. Most fundamentally, it leaves the intentions of the "speaking subject" behind:

"The text's career escapes the finite horizon lived by its author. What the text says now matters more than what the author meant to say, and every exegesis unfolds its procedures within the circumference of a meaning that has broken its moorings to the psychology of its author." (p.534)

Ricoeur's concern is not primarily with written texts as such. He proposes a point of view applicable to all interaction, equally to speech and writing. Dialogue and text are complementary ways of relating to any reality, and the difference between them is one of emphasis. In dialogue reality is free-floating, ephemeral, situational. In the text it is "fixed" – as the words on this page. Like writing, any act, speech or thought has an aspect of such "fixedness", it is imprinted in "external marks" (Østerberg 1963) which can be interrelated and contemplated outside the situation in which they arose. We know this from our personal lives. Thinking something, you remain free, but once you act the consequences tie you down: the letter you sent is different from the one you did not send, and therefore, when in doubt, you may choose not to send it at all – no "external mark" is often better than a "bad" one.

      But even thoughts will bind you. Once you've "stated it in your mind", you know – like it or not – and knowledge cannot be escaped. Having attained "knowledge of good and evil" Adam and Eve could no longer return to the innocence of the Garden – we might say they had learned to "write". But the myth also implies that all subsequent human culture is based on this lesson. Without "writing" there would be no continuity, no abstraction, no decoration, no building or tearing down, no truth or falsehood:

"In this sense, the inscription in 'external marks,' which first appeared to alienate discourse, marks the actual spirituality of discourse. Henceforth, only the meaning 'rescues' the meaning, without the contribution of the physical and psychological presence of the author." (p.535)


Rats and Punks


Ricoeur goes on to discuss the problem of "exegesis" – how a text should be understood and interpreted. We shall ask a somewhat different question. How does a text function? How does it attain the singular power by which it sustains and transforms the edifices of culture? In an unpublished manuscript from 1975, Reidar Grønhaug brings together some ideas with relevance for our theme. His basic terms – metaphor and metonym – are derived from Lévi-Strauss, who has them from poetics, where they are used to denote two types of verbal imagery. The metaphor is an idea put in place of another idea ("people are like rats"). The metonym is a part evoking the whole ("rats' feet on my shoulder"). This distinction is essential to communication. Let us take a meaningful concept, e.g. "rat". It arouses many feelings and images in us. We see dirty rats in slums, concentration camps and sewers, rats in the pantry, rats carrying bubonic plague, scurrying over our bed-clothes at night, tortured in Pavlovian labyrinths. Rats are not like mice, they are dangerous – but not in the same sense as lions or bombs. Maybe like snakes? But they're not so sleek and elegant. Their bald tails make them seem incomplete, like old men, babies, corpses – society's outsiders.

      Note how this train of associations proceeds. How do I define a rat? First I tell you where it belongs. Like any concept, this one figures in a limited number of contexts – it has its place. But when we say that the rat belongs to a context, we mean that it relates to other things which also belong to that context – food in the pantry, cats on the street, lions in the animal kingdom, bombs in the context of destruction. But this relation has two separable aspects. On the one hand there is some kind of connection, without which the rat would not be thought of "in" the context at all – either a concrete interaction, as with rats and cats, or a connection in the mind (rats and bombs). On the other hand there is a notion of separation, of difference. Rats are not bombs or cats – if they were we would not need a name for them.

      A concept has meaning because it belongs to a context. It interacts with other parts of the context and frames itself off from them. These two activities are inseparable aspects of all relationships. One cannot be part of something without being connected to it. But the connection must be specified, narrowed down, or one will merge with the other.

      Grønhaug's terms refer to these aspects of the relationship: metaphor is the aspect of separation, metonym of connection. Without either there could be no meaning, neither difference nor similarity, neither identity nor communication. The two aspects are indivisible. Nevertheless, to grasp their importance, we must understand how they can be manipulated as if they were separable. An example is a punk walking around town with a rat on his shoulder while shopping, applying for jobs and drinking beer. "I'm the rat," he's saying. "I'm the outsider: what you suppressed, violent and true." Though of course he's not literally a rat, the punk has done something simple, but very subtle. He has moved the rat out of its hidden contexts of death, dirt and destruction into the day-light we all inhabit. But in doing so, he only moves the "frame" around the rat, the aspect of it which makes it different from its normal environs. Indeed, it is only this difference, the rat's metaphor, that can be moved. For he can only move the rat if he can see it as a thing, a distinct entity. But the metaphor cannot be freed from its metonym in this way. Difference and connection are aspects of each other. So when the "frame" is moved, the connections it implies string along. Naming a man a rat is – in itself – nothing more than to label a dog a dog. What gives the statement power is all that it implies, the metonymic connections which compel us to see not only a small animal with a stringy tale, but all the contexts in which that animal appears – in reality, in folklore, in dream, in theory.

      Clearly, the punk knows what he's doing. But he's walking a dangerous line. In manipulating with contexts he is inviting all the world to read whatever they like into his metaphor. Not all contexts to which his rat is traced are contexts of which he would approve. His metaphoric little animal is a sluice, through which metonymic energies may be admitted. It's difficult to design that sluice and control it in such a way as to get across the exact meaning he wants. What we are discussing is therefore not just some wierd thing that punks do, but a gamble with high stakes and an art demanding delicacy and skill. It is, in fact, what we all do when we communicate. "It's twenty minutes of nine" may seem a neutral statement, but in my mother's family it is a legend, connected with the many significant things that have happened at that unusual time of day. "It's always twenty minutes of nine", is the native explanation, and that is indeed the essence of the matter. Any statement or act is a metaphor implying a metonym – and if it is used conspicuously enough this becomes obvious.


Interpretation and Change


Metaphors may be said to be the "words" out of which a text is composed, metonyms the content that connects them – the interpretative dialogue of "reading" or "writing". The more contexts a word has been connected to the more content it will have. When we read a text, our interpretation is just another context superimposed on those out of which the text itself grew and in which it has previously appeared. If we don't know these contexts our interpretation will be completely determined by the interpretative context itself – our reactions will be "naive". Neverthess, this is also an interpretation: we always "read" the rat on the punk's shoulder, no matter if we know its associated contexts or not. In order to read what he wants us to read, we must share some of his knowledge, have experienced dialogues which are equivalent to his. But we can read the text even without such experience by animating it with dialogues (metonyms) of our own. The important fact is that the metaphoric pattern of distinctions has a "fixedness", which enables us to lift it out of time and place, like the punk lifted the rat. The distinctions have no meaning in themselves however, they must be filled from some source – and this must be done by the "reader". As Ricoeur concludes:

"To say that meaning rescues the meaning is to say that only interpretation is the 'remedy' for the weakness of discourse which its author can no longer 'save'." (p.535)

But Grønhaug's concepts bring us a step beyond this general statement. A text is not a seamless whole into which we must "leap" blindfolded. It is systematically constructed by its "author" and contains a multitude of references to other contexts which admit and guide our understanding. As long as we in fact share some of the author's contextual references, the text is not closed to us – on the contrary, we are seduced into it. No member of Western society will be misled by the punk's rat – though many will dislike it. Serious problems of interpretation arise only when there is little or no contextual overlap between the worlds of "author" and "reader", as when we try to understand a foreign culture.

      Tord Larsen, at the 1982 seminar mentioned above, pointed out that when we discuss cross-cultural "translation" we tend toward one of two absolute and untenable positions: either cultures are based on unique paradigms that are incomprehensible to each other, or the deepest paradigm of all cultures is the same and diversity mere ornament (see also Larsen 1979a; 1980). Common sense tells us that these views must be complementary and translation depend on both. But this is too vague. Larsen therefore restated the problem in a way which may be illustrated as follows. Take cultural evolution. The idea that all societies develop along the same general lines implies that they undergo some common metamorphoses of structure, on the basis of which communication may be possible. On the other hand, if they diversify with time the opposite would seem true. As Marshall Sahlins notes (1960b), both mechanisms are at work. In the first case certain properties of social organization increase linearly, such as population size, productive power, military force, institutional specialization. These factors are measurable and comparable. They enable communication and translation by emphasizing the textual aspect of society. But the specific coloration or style of these phenomena also changes: all state societies have means of coercion at their disposal but the way in which coercion is exercised differs. In this sense social systems diverge, become more and more unique, depart into closed, situational dialogues. So we are back with the same vague complementarity. But while before we were trapped in an idealized dichotomy, we now have two empirical processes that may be discussed on the basis of what actually happens in society. Larsen dubbed these processes immunization and mediation and proceeded to draw examples of both from social theory, poetry, politics, media, history, religion and philosophy. His idea was that they are not opposed, but independent movements, working at "right angles" to each other, like the woof and warp of a loom, which together "weave" the fabric – the text or texture – of the reality we know.

      There is a marked affinity between Grønhaug's concepts and the processes described by Larsen. Like immunization, the metonym closes a context, ties meaning to it and refuses to let go. And like mediation, metaphors open contexts, they move between them, are read by audiences separated in time and space. But the similarity goes only so far. Metaphor and metonym are not processes, but the elements out of which processes are constituted. Metaphor distinguishes, metonym connects. But immunization increases connection within a distinct context, while mediation connects contexts by moving distinctions between them. Both processes employ both elements, but in different ways. To understand their differences we must now take a closer look at the elements they utilize.


Rules and Flow


We have described the metaphor as a frame or boundary separating an entity from its context. This imagery must be qualified. Let us start with an act. Where are its "boundaries"? They are not external to the act but in it – as the beat is in the tune. They are rules to which the act conforms. These rules are of various kinds. Some are physical constraints – the body's strength, the walls enclosing us, gravity, distance. Others are subtler, like skill, ignorance or habit. Others again are defined by the situation in which the act occurs, or by norms common to a group – morality, status, influence, aspirations, concepts of what is possible and impossible, incomprehensible or normal. Common to all rules is some kind of constraint. But their constraint is never total. The point of rules is to structure action, not preclude it. Rules are in this sense more like sluices than boundaries. They admit action of certain kinds – certain intensities, durations and tempos. Again we are reminded of rhythm in music.

      Bound up in this association is another trait which all rules share. The beat does not exist outside the tune. Rules which are not acted on, not obeyed, do not exist. Thus rules not only permit action – they vanish when the action stops. Certain physical objects may remain, but they are no longer meaningful. If a car is dropped in the Amazon jungle, without roads, gas-stations or drivers, it is no longer a car. The natives may make something else out of it, read its text differently, but even if we know how the car is used, its meaning as a car is gone, its purpose can no longer be acted on – its rule cannot be obeyed. The rule is "in the act", in the absolute sense that it is nothing but its effect on the act. It is, as Bateson puts it, "a difference that makes a difference" (1970, p.453), or in the dry poetry of Giddens "an absent set of differences, present only in their instantiation" (1979, p.64).

      It is this abstract, immaterial quality of rules which – paradoxically enough – gives them such power. Since they are abstract, they can be moved from one context to another. They can be reconstituted out of nothingness, as long as you know how. They can be replicated endlessly, made to apply everywhere and always, independent of time, space, person, situation or consequences. Money is a rule of this kind. Power is another; theory a third. Rules governed by other rules form hierarchies of dominance or abstraction. Different rules placed side by side make for flexibility, ambiguity, paradox and humor. Rules are what give any object, fact, concept, person, action or thought its form, its identity, by setting it off from all that it is not.

      The rule or metaphor is a beat. The music it drives is the metonym, the aspect of connection. But connection is never static – for all that is static is distinction, and thus a rule. The metonym is movement, flux, flow. But to connect, flow must not simply move, it must cycle to and fro between the points it connects, drawing them together like a whirlpool – assimilating every part into the whole. Flow thus operates on the very opposite principle of rules. It obliterates distinction. It is pure movement without identity, quality or meaning. And since it is cyclic, flow may be thought of in cybernetic terms. We may take a simple cybernetic circuit – a room with an oven and a thermostat. The oven warms the room, the thermostat turns the heat off when a certain temperature is reached, turns it on again when it gets too cold. The cycle of heating and cooling is a circuit of flow. All props surrounding it, holding the circuit in place, are rules – the size of the room, the capacity of the oven, and the setting of the thermostat. Without rules, the cycle would collapse. But the unique balance, the negative feed-back achieved in this case is not the product of just any rules, but of specific ones. It won't work if the windows are open or the thermostat is not set properly.

      By equating part and whole, by levelling all distinction, flow resists and seeks to assimilate the rules governing it. But without the focusing and directing power of rules, flow loses all identity. It does not vanish in thin air like a rule without flow. Instead it "leaks out" and diffuses into random vibration. When it is governed however, it ties the rules together into a continuous, autonomous system – a society, situation or context. Without flow, thermostat, room and oven would be dissociated atoms drifting in limbo. Without rules, flow would be the Brownian drift on which the atoms are born.

      Without the order imposed on reality by rules and flow there would thus be no reality we could know at all. For order has no given reality outside ourselves – we create it with our mind and senses and it is always and unavoidably a construction. What we know is never reality in itself, but our interpretation of it. What we do not know is in this sense unknowable, because by learning to know it we change it. It is Kant's Ding an Sich and Schrödinger's cat. But since it precedes all knowledge and order it is also the primordial Chaos of myth, out of which rules and flow build all knowable worlds. As we have seen however, these worlds are vulnerable. Flow and rules must balance each other for the web of Texture to be woven. If their equilibrium is lost the fabric comes undone and we are precipitated into a void with neither form nor content, meaning or direction. The philosophical abstraction of the Ding an Sich thus becomes an existential possibility, threatening to unravel the very fabric of reality. In this sense therefore, the unknowable can be known. When our ability to know and order the world breaks down, Chaos invades and we are plunged into anomie, psychosis, estrangement. But – as Nietzsche saw – this is a challenge, not only a threat:

"Man muß noch Chaos in sich haben, um einen tanzenden Stern gebären zu können. Ich sage euch: ihr habt noch Chaos in euch." (1887)

The discussion of rules and flow has thus provided us with a world-view in which the Ding an Sich has a natural place as a substratum of Chaos – or freedom of Nirvana – underlying the maya-web of social texture.


Feuds and Poison


We have seen that texture is constituted of rules and flow, distinction and interconnection. Before that however, we observed that the processes of immunization and mediation may be described as particular ways of combining flow (metonym) and rules (metaphor) – i.e. they are processes which produce texture. We shall now combine these two trains of thought by seeking to describe exactly how immunization and mediation utilize rules and flow, in order to arrive at a theory of textural genesis and change.

      We may take as our starting point two empirical situations – Evans-Pritchard's (1937) description of poison-oracles among the Azande of central Africa, and the evolution of European feudalism as presented by Perry Anderson (1974ab). In both cases a particular kind of conflict is resolved. Among the Azande, stress and frictions in a local community result in accusations of witchcraft. These are "judged" by the oracle – a poisoned chicken, which answers "yes" or "no" to the questions posed it by either surviving or passing away. In the second example, pre-feudal Europe, the conflict was between the remnants of Roman cities, which specialized in commerce and artisanship, and a Germanic aristocracy basing its power on control of agricultural land. In Europe this conflict – which in similar historical cases has almost always been resolved by land-holders suppressing the cities – led to the social transformations that culminated in industrial capitalism.

      We shall ignore the contrasts in scale and historical depth in these two examples and instead try to define the role played by rules and flow in each case. In the African example, we have on the one hand a repetitive sequence of interaction or flow, periodically resulting in conflict. We also have a rule for how to tackle such conflicts – the oracle. The basic conflict in this case is therefore the opposition of interactive flow and controlling rule. On the face of it, this conflict seems unavoidable. Any cycle of flow varies in volume and force; the rule however, is a rigid, but perfectly abstract frame. How can a brittle abstraction contain real motion? Why is it not simply broken apart? The answer is simple – it survives by being not one, but many rules. Evans-Pritchard brings this out clearly. The oracle is a very complex system of rules. To ask a question, intricate ritual procedures must be observed and any deviation will mislead the oracle. The poison may be badly prepared, taboos may be broken, an unknown witch may be influencing the chicken. Any pronouncement by the oracular "court" is therefore subject to a diffuse evaluation by the supplicant. A trustworthy person will never be suspected of witchcraft in the first place and since the oracle only answers yes and no, he does not risk being pointed out as a witch. If he should be so anyway, he may be saved by the fact that the oracle must always answer twice consistently, and since chickens are expensive it may never be given a second chance. Even if the pronouncement is confirmed, witchcraft is not seen as a conscious activity on the part of the witch, but an involuntary reflex. So the witch is approached discretely and warned. Only if he is repeatedly accused or known as a bad trouble-maker does he risk punishment. Thus the system functions to channelize and defuse conflict, to sanction real deviants in a systematic and highly selective way. But it only works because there are a multitude of rules governing interaction. Flow may be erratic, but there is always a rule there to "catch" it. We may thus envision a process – derived from the inherent conflict of a rule with the flow it governs – leading to a gradual accretion of many rules to the same interactive cycle of flow.

      The evolution of feudalism presents a very different picture. City and countryside are opposed. The basis of their conflict is that urban artisanship and commerce potentially threaten the power of the land-holding class, as they demand increased mobility of labor and land, and undermine the aristocracy's hold on the peasants. This conflict is universal in state societies and may be traced in the annals of many non-European civilizations. However, one factor in the European situation was specific – the extreme fragmentation of political authority. Early medieval Europe was an unstable conglomerate of two incompatible social systems – on the one hand a disintegrating Roman culture, centered in the cities and in the city-based Church but also influencing organization of rural manors. On the other hand an unstable Germanic culture, which originated with mobile warrior bands led by petty chiefs and their retinues. As the invaders settled, the chiefs partitioned out estates to their followers, but these retained a high degree of autonomy. The resulting hierarchy was a conglomerate of ambiguities and overlapping claims, from the local level of individual manors up to the continental fragmentation into a multitude of belligerent states. Into the cracks and interstices of this structure the cities were inserted in an ad hoc fashion. This gave urban culture a breathing space and permitted an autonomous development of trade and urban production which the rural lords could not prevent. The process here described differs radically from the Azande case. The European opposition is not between a single cycle of flow and its governing rule. Instead, we have two discrete systems in conflict, two separate cycles of activity, each controlled by different kinds of rules – one Roman and urban, the other Germanic and rural. The conflict is also resolved differently. The conflict of city and countryside is not neutralized and perpetuated, as with the Azande oracles. This is what happened in the more centralized Chinese state, where the emperor held the cities permanently on a tight leash. But the European cities ultimately conquered both urban and rural sphere – and in doing so transformed both and created a new, capitalist, system, which unified them. Thus, out of the conflict of two sets of rules, each governing autonomous circuits of flow, there grew a new, over-arching rule, which coordinated the rules from whose conflict it was originally derived.

      In the Azande case a conflict of rules and flow generates a process of increasing multiplicity of rules on the same level. In the European situation a conflict of two rules generates a hierarchy, in which an over-arching rule governs and transforms the rules on lower levels. Under certain circumstances a conflict may be escaped – or suppressed if one of the parties is much stronger than the other. But in the examples described above this does not happen. The oppositions we are dealing with are not merely conflicts, but "contradictions". This is a technical term much debated by marxists, and I will not enter into their discussion here. I shall instead propose my own definition – that a contradiction is any conflict which is "locked". The concept of the "lock" will be clarified in the following, but for the time being I shall simply define it as any circumstance – material, social, emotional, logical – which "surrounds" a conflict between parties of equal strength in such a way as to make it unavoidable (i.e. the participants can neither "pull out" nor "win"). The interaction between prisoners in a cell is "locked", so is the conflict inherent in the following sentence: "I always lie." (If it's true, it's false – and vice versa.) In the first case, the lock is a physical barrier, in the second a syntactical one (enclosure in one sentence). Furthermore it is clear that some locks are more permanent than others. The prisoners are only locked as long as they are confined together, the sentence only as long as it catches our attention. There are thus different kinds of locks – and we must now define the difference between the lock in the poison oracle and in European history.

      In the Azande case, flow and rules are opposed, and since rules and flow are inseparable, conflicts of this kind are therefore always contradictions – they are locked by the very way in which we constitute reality. We may refer to this as a logical lock. In the European example is of different kind. It may be that the urban and rural systems are logically incompatible, nevertheless, their conflict very seldom develops into a contradiction. What forced the parties to fight it out to the bitter end in this specific instance was the historically unique and highly improbable coincidence of organizational, cultural and geographical circumstances that prevailed in pre-feudal, post-Roman Europe. This lock is not logical but situational. It has one necessary trait however, which we must notice. Somehow it is strong enough to withstand the force of the conflict it contains. It has some crucial component which is "heavier" or "more real" than the conflict itself. We might for example suggest that the European contradiction was confined to a very restricted land-mass – so conflict was forced inwards rather than being dissipated into a boundless periphery.

      We can now return to the processes of immunization and mediation. The Azande case exemplifies the first of these. It elaborates, decorates and diversifies – enhancing the uniqueness of the Azande world and its impermeability to comparison and communication with what is outside. The development of feudalism is an example of mediation, since local, particularistic contexts are integrated and unified under the auspices of a single, over-arching rule. Both processes proceed from locked contradictions. But the nature of their locks is different – in the first case it is logical, in the second situational. The first process takes place automatically in all social situations as long as they remain stable for any length of time. The second is precipitated by exceptional circumstances and is very rare. Once it succeeds however, it tends to spread, since it produces systems with power to dominate their surroundings.

      Both processes are dialectical in that they generate a synthesis on the basis of two factors in contradiction. But our discussion casts a rather unusual light on this concept. In classical marxism, dialectics are the driving force behind a unilinear, deterministic process of social evolution, leading by necessity from classless, "primitive" societies to "modern" societies dominated by power. Our discussion turns this whole conception inside out. There is not one process, but two. One of these (mediation) may well be equated with evolution, as the European case shows. But this process does not proceed by necessity, but is a rare freak of history. The other process is not like evolution at all. It does not increase the power of society, but its flexibility, its ability to absorb and cushion random variation – it may be termed adaptation. In stable situations adaptation is necessary, but not unidirectional, since it increases variation.

      The stochastic character of evolution (see Bateson 1979) contradicts the heritage of 19th century determinism from Darwin and onwards. Godelier (1978a) is one of many who have been perplexed at this. He notes the "stagnant" development of oriental empires such as China and goes to great lengths to explain it. However "stagnation" is not the exception, but the rule – as any overview of the past excluding the last few hundred years will show. As Carl Sagan (1977) has pointed out, a mere 0.74% of human history has passed since agriculture was introduced, and no more than 0.46% since the first cities. Throughout 99.26% of our existence as a species all humans were hunters and gatherers and most continued that way until the most recent present.


Power and Flexibility


Adaptation (immunization) encloses flow in an intricate cocoon of rules, cushioning it and increasing its flexibility. Evolution (mediation) erects a hierarchy of rules connected by power. It should by now be obvious that the two processes are not opposed, but complementary. A system may be powerful but also flexible, it may be flexible but not powerful, etc.. We may therefore envision evolution and adaptation as two intersecting axes of variation and change. Each of the four points of the cross they form denotes an extreme state to which a texture may be driven by adaptation or evolution: On a macro-level, evolution proceeds from "primitive" to "modern" systems. But since mediation is just as relevant on a micro-level I shall avoid the traditional terms and refer to an evolved texture with an extensive hierarchy of rules as "deep", its opposite as "flat". A flexible, well-adapted texture is crisscrossed by multiple rules and will therefore be denoted "dense", while mal-adaption and inflexibility is "open". (Figure 1.)


      Flexibility is defined by Bateson as "uncommitted potentiality for change" (1970b). A circuit of flow enclosed in a dense cocoon of rules has such potentiality, in as much as only one rule is activated at a time. The rest remain latent until the flow changes and encounters them. The correlate of rule flexibility is thus long-term flow stability, or rather dynamic equilibrium, as in the example of the thermostat. Conversely an open system will lack these attributes – its rules will be few, and therefore inflexible and brittle. Flow will be liable to escape from their grasp by gradually "leaking", or by exploding them altogether. Density thus conserves flow, controls action efficiently, while openness is "wasteful" and makes for inefficient and erratic control. A dense economy produces and circulates goods with little waste. Dense political structures are legitimate, the exercise of power is mediated and predictable rather than abrupt, violent and irregular. Dense social relations are stable, or change in a rhythmic or regular fashion. A dense machine is not liable to break down – but the most intricate and balanced machine is a crude stone axe compared to the density of the human body – even an amoeba has greater efficiency.

      The social effects of density are may be clarified by further reference to Evans-Pritchard's discussion of Azande beliefs. They do not, he says, believe in witches because they do not know the causes of misfortune. Witches are responsible for death, sickness and tragedy – but not in a causal sense. If a man is killed by a collapsing storage house the cause is clear. The termites ate the wooden supports. The man went to sit in the shade. What is not so clear is why the two chains of cause and effect should intersect then and there. We call it chance – Azande attribute it to witches. The difference is mainly that chance to us is an anonymous and purposeless force. It gives no meaning to the man's death. This brings out a crucial effect of density. It cushions our experience of life, makes it comprehensible, insulates us from anomie and the nameless terror of the Ding an Sich. Peter Winch (1970) says that the Azande have "contemplated human necessity". They cannot avoid pain and death any more than we. But they accept it, in the sense that they cannot and do not combat it. In this passive, waiting state contemplation is possible and misfortune acquires meaning – for without stability, adaptation fails. Beauty is another attribute of density. Observe the intricacy of a copse of birches in the fog. The interwoven, bare branches waiting for spring. The pale trunks rising from the ground. Each bit of the scene is subtly balanced. There are so many rules here that no rule can be discerned, so many interlocking wills that it seems purposeless – pure being for its own sake.

"Music, the movement takes possession of all our soul – and this movement is nothing but immobility. As in the spectacle of the wave the moment when it begins to break is the very moment of concentration and beauty." (Weil 1970, p.84)

Openness also has beauty, the wierd beauty of factories, slag-heaps and abandoned cities. It is a sad beauty, though it is important for our understanding of ourselves. Turner (1964) has shown this in a sensitive discussion of rituals of change among the African Ndembu. He points out that transitional stages in life (birth, marriage, death) are points where social texture is strained and open and threatens us with anomie. It is typical of stable societies that they enclose all human necessities in dense webs of ritual and symbolism. But stability can never be complete, and however comfortable, it is ultimately denied by the stark realities we must all face. Therefore the Ndembu ritual web does not cover up openness, it conserves it, opens access to it, focuses our attention on it, even exaggerates the Chaos it represents. Thus we are forced to contemplate it, and in confronting necessity to renew ourselves and society.

      We now turn to the process of evolution, to characterize the differences between deep and flat textures. What the term "power" implies in the model I am proposing is a specific type of relationship between rules. When an over-arching rule is established, it must (to be real) be obeyed, it must contain flow. In itself the rule of power is merely an abstract compromise between the contradictory rules it seeks to unite. To become real it must limit the authority of the original rules, penetrate their barriers, so to speak, and siphon off part of the flow they govern. This flow may then be merged into a wider circuit which obeys the over-arching rule. Power is the means by which this is accomplished. A simple example is taxation. A state may be seen as an aspect of an over-arching rule which, to govern, must secure the means of its own existence. These means are extracted from the productive population, collected at the center and re-channeled to the multiple activities in which the state is engaged. Two conclusions may be drawn from this model. First, since the rule of power derives its "substance" from lower-level rules, it is dependent on them. Without their (willing or unwilling) contribution, it would not be real. Power limits the autonomy of the local interaction on which its rule rests, and is therefore always resisted. As Giddens (1979, p.145-50) points out power is always a two-way relationship. A "deep" hierarchy therefore always entails a state of polarization – of power and resistance – which a "flat" system avoids.

      On one level it is obvious what this means. The universal principles by which a state governs are derived from the particularistic interests of its citizens, but at the same time estranged from them and therefore resisted – explicitly (tax-evasion) or implicitly (staying home from elections). But polarization is not an undesired or external consequence of power, it is its essential nature and permeates every aspect of a deep texture. And so it is not confined to politics. Life itself is polarized in modern society. We move to and fro between spheres which pertain directly to our own lives ("friendship and hobbies"), and spheres which are in principle completely impersonal ("contracts and money"). But even this description is superficial. An "impersonal" career can in fact be personal, paying our fare on the bus can be more than an abstract monetary transaction. Conversely, children may be brought up "for society's sake". Polarization is therefore not external to the act, but internal to it. What is polarized in a deep texture are not only different kinds of acts but different aspects of the same act. Almost any act in the modern world has this split character. It relates to the general and the specific, the abstract and the concrete at the same time. It has obvious consequences for ourselves, here and now. But it is also coordinated with millions of other acts in vast patterns which none of us understand. This is not true in a flat texture, where hierarchies are shallower and most effects of an act are immediately visible.

      In The Savage Mind (1962), Lévi-Strauss shows that polarization permeates our conceptual world as well as our activities. He contrasts a modern and a primitive mode of thought, typified by the engineer and the bricoleur – the jack-of-all-trades who repairs our car with whatever materials are at hand. The engineer works with specialized concepts pertaining to concrete things, which are coordinated by generalized abstractions. But the bricoleur's concepts are simultaneously abstract and concrete. An example is the stone axe used by native Australians (Sharp 1964). This was their only tool and served an almost infinite variety of practical purposes, from chopping down trees to delicate sculpting. It was not specialized in the sense a bread-knife or chain-saw is. But its multiplicity of functions must not either be misconstrued as a result of a "primitive" inability to think in abstract terms. The axe was abstract and concrete at the same time. It was not merely a tool but a living thing, a spirit, a totem, a statement about society and human relations with the most general implications. When Europeans supplanted it with cheaper, but by their nature specialized steel axes, the very basis of generalization was threatened.


An Algerian Soldier


Texture is woven by evolution and adaptation, but is in itself nothing but an abstract pattern of distinctions, a vast composition of crisscrossing rhythmic patterns, superimposed for nuanced expression or coordinated for increased force. Beneath this pattern lies the Ding an Sich, that which is not patterned and not ordered because it cannot be known. But this image of the world includes no objects or people or institutions, not even words or thoughts or things. These entities congeal out of texture, they are the harmonies and disharmonies of the music, not the music itself. I refer to them as centers.

      To take an example, Barthes (1975) describes a black Algerian soldier saluting the French flag. This poster, which was used for propaganda purposes by the imperial French state, is a center. It becomes a "thing" for us by virtue of a subtly manipulated pattern of contrasts and connections, rules and flow. It is a fragment of a larger web of evolution and adaptation, power and flexibility. Barthes points out some of its qualities. The soldier, qua soldier, is a real person, with a life, a history. He belongs to a context. He is moved out of that context as a metaphor, into the new context of the picture. But not only is he moved, he is subordinated to an over-arching rule. In the picture he is not himself, but a sign of the allegiance of Algeria to the empire. Still, he is not just dominated, he resists: he remains himself. What we see is a polarized center. But it was not so easy to see this if one was French and lived at the time when there still was a French empire. The picture is well composed. It checks and balances our objections. The soldier looks human, sincere, the sun is shining, the flag flapping prettily in the wind. The distractions make its polarization unobtrusive. It is propaganda, but it is not crude. Its polarization is subtle, flexible and stable – it is a dense center.

      This tension between density and depth goes far towards describing the center so we see it. But there is more here. First there is the soldier himself. Then there is his image. Then the image of the image, which produces the polarized, power-ridden effect. The over-arching rule of empire which slices into this last picture dominates the soldier by reducing him to a sign of empire. But at the same time the power of that sign is dependent on the soldier himself – as a person. It is because he in fact is real that the derived reality of the rule of power seems real to us. Barthes reacts to this as an insult. The image of the image is a corruption – because it feeds off the autonomy of the original, "pure" sign. I agree, this is what power is based on. But what convinces us is not this in itself but the elegant, unobtrusive way in which it is brought off – the center's density rather than its depth. Barthes is offended by its depth because he sees the original image as somehow uncorrupted and true, as meaning pure and simple, and therefore "more real" than the image derived from it. There is something very problematic, but at the same time intuitively right, about this. As an abstract stand-point it is clearly wrong. No meaning is uncorrupted. There is no such thing as a pure sign. What the polarized center does is no more than Barthes himself does when deriving generalized abstractions from concrete examples. Still, in some sense something "more real" has in fact been lost to something "less real", and the "less real" feeds off the "more real" and could not survive without it. This however, leaves us with the question: what do we mean by saying that something is "more" or "less" real?


The Real and the Ideal


This was the question that occupied center stage in the previous chapter, and although we did not note it at the time, it was also implied by Ricoeur's statement that the text is "fixed" in external marks, since these "marks" must be "more real" than the text itself. Thought binds us like action. But there is a qualitative difference in the strength of the bond in the two cases, because the "external marks" of action are "more real" than those of thought. It therefore seems that the "external marks" in which texts are inscribed have different "weight". We are always chained to continuity, but some chains are fastened to concrete objects, other to images in the mind. There thus seem to be different kinds of reality, some more "solid" or "heavy" than others. "Light" things like meaning are inscribed onto "heavier" things like acts, which are imprinted on still "heavier" things, like brick walls or bodies. Perhaps reality seems real to us because of this chain of anchoring in more "fundamental" realities. But this assumption seems to contradict the notion of Chaos, which we saw was the underlying paradigm of texture. In the first case the real seems to be a hierarchy with increasing solidity the deeper we get. In the second case all reality is constructed and the entire hierarchy rests on the quicksand of Ding an Sich. Must we conclude that the seeming hierarchy is an illusion, that all parts of reality are equally "light", equally a product of the order we have clothed them in? This would – if we accept it – imply an all-out relativism, in the light of which our impression of some things as "more fundamental" than others is pure superstition.

      Because of its highly abstract nature this problem has repercussions throughout social science. Marxism is a case in point. Doubtless, the greatest asset of marxist theory is the insight it provides into relations of power and hierarchy – i.e. into conflicts between more and less "real" worlds. Still, it is often unclear what exactly the marxist concept of hierarchy implies. On the one hand we have the idea of a pyramid of ruling and dominated classes. These are different groups of people with conflicting interests, because of their different access to the means of production. On the other hand we have a "functional" hierarchy of infrastructure, which produces the means of survival, and superstructure, which legitimizes the power organizing and appropriating the fruits of production. But this "functional" hierarchy is itself confusing. True, in the "vulgar materialist" version, it presents no problem – infrastructure is the material world, the economy, which dominates the world of spirit or superstructure. More subtle thinkers (among them Marx himself) have discerned a dilemma in this. The "economy" is clearly not just material, but a system of action. To act however, one must have knowledge and skills, which in turn imply categorization and meaning – all of which are "spiritual" terms.

      These problems are typically solved in one of two ways, exemplified in the work of Hamza Alawi (1973) and Maurice Godelier (1978b). Alawi asserts that producing the material necessities of life is more fundamental than thinking about them, and for this reason the institutions concerned with production are more basic than those concerned with thought. Classes are such institutions, and thus the hierarchy is simply a series of groups – some (the lower classes) directly involved in production, some only indirectly involved through their power to exploit the producers. Godelier takes the opposite view, that infrastructure and superstructure are not institutions at all, but "functions", defined by the type of task they perform (e.g. "production" or "legitimation"). They are thus neither material nor spiritual, but systems of action, which "use" any props and prerequisites that action requires (knowledge, skill, tools, personnel, buildings, institutions etc.) as means to their own ends. Thus, while to Alawi "institutions" are a priori "facts of life", to Godelier nothing is more basic than anything else and the concept of power loses its meaning. Both solutions solve one half of the problem by distorting the other.

      We must face this issue squarely. Either some realities are in themselves "more real" than others, in which case the text can free itself from the context where it arose. Or else all realities are equal, all discourse situational, and there is no continuity. Either reality gets more solid and "heavy" the deeper we go in it or it's all a fiction, an arbitrary order we impose on unknowable Chaos. Or to formulate the problem in more positive terms: Texture is a web of distinctions woven by adaptation and evolution. But as yet we have no criteria of "reality" in this web. How do formless connection and abstract distinction produce a world of centers – solid, tangible persons and things?


Locked Realities


To understand how this "fixation" (Ricoeur) comes about, we must return to the idea of "locking". The lock is the crucial factor in both adaptation and evolution, the "lens" which focuses each process and gives the contradiction from which it springs its incontrovertible and compelling character. Nowhere in our discussion have we encountered a concept with such solidity, and in our search for the reasons why some realities seem "more real" it is therefore natural to turn to the lock.

      We have described two basic kinds of locks – the logical lock, on which adaptation is based, and the situational lock of evolution. These "tie down" reality in different ways. The logical lock may be likened to an imperative demand for consistency within texture itself. An example might be the contradiction involved in my acting in one way and thinking of myself in another. There is a dialectic between these two aspects – my actual doing (flow), and the "label" (rule) I identify with. It is a conflict I cannot evade, to which I will always seek a solution – and the primary characteristic of the solution I seek is consistency. I will never attain it, because what I do is dynamic and ever-changing while my "label" is a framework of fixed distinctions. The concepts are by their nature irreconcilable. But the problem they pose gives my life a purpose and meaning so imperatively "real" that I am forced always to try to resolve it.

      The situational lock "binds" reality differently. As we saw above, this lock must contain some element which is "external" to the contradiction – more fundamental than it, in the sense that the physical extent of the European land-mass is more fundamental than the organizations subsisting from it. In general terms such locks might be said to reside in the "environment" of social textures rather than in the textures themselves. This however, begs the question – for what is "environment", and how can it be said to be "more real" than the system resting on it? If all reality is constructed, an ordering of Chaos by means of rules and flow, are not all realities equally arbitrary? The question must be answered in the affirmative. From the standpoint of adaptation – i.e. from a perspective based on logical contradiction and the need for consistency – all reality is indeed nothing but a more or less densely woven web of texture. But from the standpoint of evolution this is not true. In our treatment of the Algerian soldier we observed that the "image of the image" rests on the "image", which in turn rests on the real soldier. From an evolutionary standpoint this cryptic "rests on" has an exact meaning. The different levels of imagery constitute a hierarchy tied together by power and resistance. To survive, high-level rules must curtail the autonomy of lower levels by "siphoning off" flow from them. They are dependent on lower levels because they themselves cannot produce the flow on which their existence is based. In this sense some rules are indeed "more fundamental" than others and the term "environment" takes on an exact meaning. What is environment in relationship to any system, is a level of rules which produce reality more directly than the system itself. Production is in this sense the environment of exchange, the ecosystem the environment of society, the lifeless physical world the environment of the ecosystem. Beyond that, as modern physics has shown, our ability to know breaks down. We cannot "see" deeper than to the "particles-which-are-waves" – the rules and flow out of which physical reality is constituted.

      But though higher levels of rules are dependent on their environment they also exert power over it and thereby change it. The environment is therefore not simply a "realer reality", it is a "realer reality" which is deformed and marked by the system it carries (as the physical reality is transformed – though still physical – in a city). Any given level of texture thus rests on imprints of itself in the levels of texture on which it depends (its environment) – just as the "image of the image" of the Algerian soldier rests on the Algerian himself.

      But this invites yet another conclusion. If production of goods is the environment of exchange of goods, then surely the production of meaning is more fundamental than communication. But is then the production (and exchange) of goods the environment of the exchange and production of meaning? Common sense seems to tell us that "goods" are more real than "ideas". In this connection Jürgen Habermas (1973, p.27f) has made an important point, which – though obvious – is seldom realized. He reminds us that the environment of society is not merely the "nature" surrounding us, but also the "nature" within us – our bodies and psyche. We may object that the psyche is a social product, and this is true enough. But the social product resides in a pre-social "container", and this "un-socialized mind" is not a tabula rasa. Modern neuro-psychology has proven beyond doubt that the brain, before it has ever learned a thing, is a complex, functioning system, with explicit capabilities and equally explicit limitations. So the environment of social texture resides "within" us as well as "without". This means that texture is imprinted not only in external nature (physical objects and forces like roads, machines, books), but also in internal nature. Most obviously, it is imprinted in subconsciousness, which, as Bateson (1967) has convincingly argued, is not merely a Freudian cabinet of suppressed horrors, but a repository of skills and knowledge so necessary and universally useful that it would be self-destructive to question their validity.

      In this way the distinction between material production and the production of meaning disappears. Goods and meaning are produced simultaneously, objects and categories are aspects of each other. What is a physical thing in the outer world is a thought or emotion or category in the inner world. Indeed the very concepts of rules and flow are enhanced by this insight. We see now that the separation and connection which these two concepts describe is most fundamentally a connection and separation of the inner and outer environment. This is of course just another way of saying that society is nothing but action – that rules and flow cannot be constituted at all if they do not bridge the gap between a human being and the world surrounding him or her.

      The evolutionary point of view thus states that high-level rules have power, in the sense that they are imprinted in realities on which they depend, and that these realities – the environment – are equally internal and external to the human organism. What locks and "fixes" reality in this perspective is therefore not the need for logical consistency of the adaptational lock. Instead it is our dependence on those parts of reality we dominate. It is, in a very true sense, our need for sustenance – things to rest on, food to eat, emotion and experience to think about.




Texture is thus woven by processes arising from the locked contradictions of rules and flow. These contradictions have an imperative power over our lives which originates in our need for consistency and survival. They "fix" the texture so any more or less homogeneous area in the fabric may become the focal point of our search for survival and consistency. In that case it is locked – logically and situationally – and acquires the characteristics of an object, person, role, symbol, dream, concept, institution – or even society. It becomes a center – an entity with a life of its own, which we can communicate with, think about and manipulate.

      But the center is derived from texture, it is not primary. It is common to think of centers as if they were independent "things", with an a priori existence that need not be questioned. In social analysis we relate to such "things" as black boxes. They move around, behave in certain ways, are manipulated or change. But as they are black boxes their inner nature need not be taken into account and they may be categorized on the basis of purely external similarities: All societies are societies, all persons are persons, all symbols symbols – and may be expected to behave, be manipulated or change as persons, symbols or societies "always do". This is not only misleading, it is wrong and engenders persistent misunderstandings and controversies. Schools of philosophy have been built on this misconception: Idealism and materialism, macro-analysis and actor-orientation, structuralism and role theory. Each is based on the fallacy, that "things are things", but some things are "more real" than others. The point this essay seeks to make is that "things are not things". On the contrary, they only appear to be things and it is the processes constituting them, not the face they present to us, which is their main point of interest.

      We can now see that the standpoints of Godelier and Alawi which we referred to above, are complementary. Godelier's functional hierarchy is a category of texture, while Alawi's hierarchy of groups is a "fixation" of texture into centers. This casts light on several important questions: First, it becomes clear that a ruling class may itself be productive, just as the lower classes are given opportunity for leisure. This is because the basic hierarchy is textural – what "governs" society is not a group, but an over-arching rule, controlling a complex circuit of action. The important datum then is not who performs the actions of government, but that they are performed. They may be narrowly focused in a small clique or widely dispersed – in which case the external hierarchy of groups may be explicitly denied, as in a democracy. But the degree to which power is concentrated in a group is not a measure of the depth of hierarchy in that society. The concentration of power in ancient Egypt does not make it more hierarchical (and hence deeper, more modern) than our own society. What the concentration of power may indicate however, is the degree to which the ruling power is legitimate. If power is legitimate (i.e. if society is dense) power may be freely dispersed without risk to the basic function of coordination. In an open – illegitimate – texture however, power must often be concentrated to compensate for the inefficiency and lack of flexibility of the total system. This, as I have argued elsewhere (Nielsen 1987), is the basic contrast between the state systems of the Soviet Union and the western world.

      Finally, when the two concepts of hierarchy are clearly separated the notion of power itself becomes clearer. An over-arching rule has power in the sense that it siphons off flow from the rules it governs, collects this flow and utilizes it to coordinate the system as a whole. This in itself is neither moral nor immoral, it is a necessary attribute of any deep system, and removing it would lead to complete chaos. It is by virtue of this coordinating function that the productive power of society has increased so vastly in the course of social evolution – and the underlying reason why the modern world can support a population many thousand times larger than in the Paleolithic. Obviously one cannot remove power in this sense of the word without simultaneously offering an Endlösung more dramatic than Hitler ever hoped for. The marxist dream of a communist utopia is thus just that – a dream, strangely out of touch with the reality Marx himself so subtly analyzed. But in the hierarchy of groups power acquires a different meaning. A "ruling class" can take on important tasks implicit in the over-arching rule, and – by virtue of the position thus acquired – indulge in a life of luxury while grinding the population underfoot. Such a group has defined itself very effectively as a "thing". It in fact typifies the formation of centers as we have discussed it: the rule of power has imprinted itself on the rules it governs, on the organization of society, on peoples' minds, and on nature. Its imprints are the lock which sustains the rule, fixes texture as a text to which all readers must relate as objective fact. If you control access to this lock, you can potentially establish yourself as an equally objective and necessary given. But there is a further step to be taken before you become a tyrant or a saint. You must contemplate the necessity in which you have placed yourself, the logical paradox between the rule you represent and the flow you actually engage in: you have identified with something impersonal and abstract – but you are still a person and concrete. Contemplating this logical lock does not enhance your power, it increases your flexibility, makes you impermeable to outside influences – among them such factors as common decency and self-evaluation. By combining these two processes and anchoring yourself in the locks on which they rest you achieve that position of fame or infamy, which is all that the text of history will remember of you.