"Models of social complexity"
- What is theory? -

Finn Sivert Nielsen

Paper read at the Magleås seminar, May 15th 2001

© Finn Sivert Nielsen. Distributed by www.AnthroBase.com.
To download, print, or bookmark, click: http://www.anthrobase.com/txt/N/Nielsen_F_S_04.htm. To cite, quote this address and the download date. Not for commercial use. Please do not remove this notice from digital or paper copies of this text.

This paper has two titles, because I am going to do two things in it. First, I want to tell a story about a book - or rather, a collection of articles - that I started working on some 15 years ago, and which still exists only as an unfinished manuscript. Secondly, I want to try to answer the question what theory is.

During this seminar, we have seen a number of examples of theoretical constructions at work. We have seen them twist our commonsense perceptions of the world, connect things that normally are not connected, separate things that seem inseperable, draw conclusions about the world on the basis of esoteric metaphors and / or do-or-die logic. But what is actually a theory?

We tend to think that a theory is a kind of tool, like an axe or a knife; indeed, the idea of a "theoretical toolbox" is rather common among anthropologists. The "tool" (and here we should remind ourselves of our Monday morning session on "technology") is commonly thought of as a means to increase the efficacy of human action. When we cannot lift the rock, we apply a lever to it - and lift it anyway. But the ethnographic record shows conclusively that even "mere" physical tools are - behind their utilitarian facade - in fact, far from simple. Complex metaphorical structures adhere to them, and make them as much a means of elaborating and estheticising the world as a means of exploiting it.

Let me briefly examplify this point:

"Olive Took an Axe" - is the name of a recent television serialization of a novel. We know without ever having seen the film that Olive was not merely going out to chop wood (not much to base a movie on, after all). It goes without saying that she must have a more sinister motive - to kill someone, most likely. The title itself poses the questions: "Who will she kill - or attempt to kill - and why?"

Here we see the docile, utilitarian Axe developing a strange and horrifying aura. We sense that Olive has somehow perverted the pure purpose of the Tool, by turning it - our servant - against us. And indeed, the idea of the "perverted Tool" seems to be an intrinsic theme of modernity - see Goethe's Faust, Mary Woolstonecraft's Frankenstein, Marxian social theory, and modern gas chambers, nuclear arsenals and greenhouse effects.

An article that is read by every first-year student of anthropology in Norway is Lauriston Sharp's classic Steel Axes for Stone-Age Australians, from the 1950's. Sharp's narrative (recently commented beautifully on by Tian Sørhaug) of the well-meaning missionaries that distributed steel axes to stone-age Australian tribes, and thereby disrupted their entire moral universe, is another story of the "perverted tool". But Sharp dwells extensively on the "non-perverted", pre-contact stone tools of the aboriginees. His point is that the traditional stone axe was intimately interwoven with almost every aspect of social and cultural life. This gives first-year students an excellent introduction to the idea of holism. In the present context, we should take it as a reminder of the fact that tools are often highly multifunctional - so highly, indeed, that the very idea of them having a "purpose" seems dubious.

The entire present seminar, it seems to me, is premised on the idea that radical deconstructionism is on the wane, and that it will soon again be legitimate to engage in synthesis and (perhaps even "grand") theory building. Leaving the question of whether or not this is literally true to the side for the moment, we should note that the idea itself rests on a rather curious presupposition. If a theory is a tool, the lesson of postmodernism is that human cultures (in this case the culture of anthropology itself) go through "tool-constructive" and "tool-deconstructive" phases. Exactly what this means for society at large - outside of anthropology - is not so easy to understand, although such movements as iconoclasm (cf. the recent example in Afghanistan) or puritanism, are reminders that technology-bashing is a time-honored human activity.

So where do these examples leave our idea that "theory is a tool"?

This was the question I posed in the book I started working on sometime back in the early 1980's. Its title was originally "Models of Social Complexity" - a reference to Barth's widely read "Models of Social Organization". This paper by Barth was also my initial critical reference: "Organization", I suggested, is a unidimensional term, which cannot do justice to the multidimensional complexity of social reality. Barth, and many others, have sought to reduce this complexity to simple, formal models. The strength of such models is the elegance of the analyses they produce - Barth's Pathan monograph is a paradigmatic example of the genre. The weakness of formal models is that they describe a multidimensional reality in unidimensional terms. Formal models do not tolerate inconsistency, and they therefore exclude alternative perspectives on the world. Barth thus excluded structural analysis from his model, rather than incorporating it into a broader, less streamlined, less elegant, whole.

Here I returned to the idea of the theoretical toolbox. Let us consider two kinds of toolboxes: First, the well-organized kind, with all the screwdrivers and pliers and drills in all the standard dimensions lined up in orderly rows. Secondly, the kind that contains the residue of years of active use: bits of wire, an odd screw, a Canadian drill bit for very tight places, all jumbled together, but still, in some strange way, immediately accessible. Barth's idea of a toolbox was definitely of the first kind. I wanted to see what we might learn from the second kind of toolbox.

So I sat down and wrote nine articles, each of which was designed to illustrate a particular theoretical paradigm in action. Each article would furthermore show that it had its own - truthful - message to convey about social reality. Social reality was so complex, I thought, that it had to be grasped from many independent points of view simultaneously. Anthropologists were like the famous blind men feeling various parts of an elephant and quarreling about how it really looked. Instead, I suggested that we should sit down and discuss our accumulated theoretical experience, and try to piece together a metatheory of the elephant as a whole.

My role model here was Bateson, who was always an eclectic, and a theoretical opportunist. Naven seemed to me a most amazing book, since it was an actual attempt to explore several simultaneous but separate points of view on the same social phenomenon.

I actually got the nine articles written and put together. I have been advised to build bridges between them and to integrate the chapters better. There should, is the subtext, be a consistent argument running through my book. This is probably sound editorial advice, but it contradicts the purpose of my book.

The book's purpose was to demonstrate that a complex world demands complex descriptions. I was inspired by Bateson's immortal: "Two eyes see better (not more) than one". Being fluent in several languages, I reasoned, might enable the development of a metalanguage, a "theory of theories", that would encourage any and every new theoretical innovation, and incorporate them all into the continuity of its idiomatic repertoire, as weapons in its arsenal, instruments in its orchestration, tools in its toolbox.

Now we return to the question: "What is a theory?" Clearly, if we want to develop and master a true orchestration of anthropological theory, this question must first of all be addressed.

Once again, Bateson has some very suggestive ideas. In one place he writes: "An explanation [i.e. a theory] is the mapping of the pieces of a description onto a tautology." A theory, this informs us, performs two simultaneous movements: a breaking up of the world into "pieces", which are then brought together and framed by description and then, a "mapping" of these "pieces" onto a tautology.

Now what is a tautology?

A tautology is a particular pattern of logic, in which all terms are logical transformations of each other. The use of tautologies for explanatory purposes is not considered comme il faut in the empirical sciences, since their truth value is independent of empirical facts. Bateson's definition of theory therefore at first sight seems quite unsuitable for empirical analysis. But then consider mathematics. Mathematics is nothing but a highly complex, and completely tautological, formal language. It has nothing to do with empirical reality. Yet it is applied with remarkable success by physicists, engineers, astronomers, etc. And the explosive developments in biology since the Second World War are directly traceable to the fact that computation technology had by then proceded to the point where it began to become possible to model biological processes mathematically.

No social science has yet been reliably mathematized, and indeed, it is unclear if this will ever be possible. In anthropology this is all the more true, despite the limited use of quantitative methods in certain subfields, and despite the rather widespread use of formal models - such as games theory or structuralism, that might in principle be mathematized, but in fact, no such attempt is ever made.

What we see in the latter cases is that mathematics is used not directly, through computation, but as a metaphor in social science. This observation will allow us to rephrase Bateson's elegant definition of theory into a more long-winded, but rather more exact form:

An explanation, we shall state, is a metaphorical mapping of the pieces of a description onto some other area of reality that is perceived, by the mapper, as being independent of the described reality, and as having a certain self-sufficiency and coherence.

In line with this, David Parkin argues that theory in general is nothing but a subspecies of metaphor. All theory is based on the classical metaphorical trick of saying that something - X - is "like" something else - Y. (The famous biological metaphors - from Evolutionism to Durkheim to Bateson - are here brought to mind, along with the legal metaphors of Radcliffe-Brown, the logical metaphors of Marx, the linguistic metaphors of Lévi-Strauss and the literary metaphors of postmodernism.)

Parkin's further discussion need not concern us here. It is enough that we agree with his initial assumption, which allows us to mobilize the accumulated interdisciplinary force of metaphor theory in our search for a "theory of theories".

The literary analysts Franklin and Mary Ann Roberts have written very suggestively on metaphor, stressing several points, two of which I will dwell on briefly below:

Metaphors are relationships of "resemblence", which is not the same as relationships of "identity". When something is said to be "like" something else, it occupies an intermediary space between identity and difference. But exactly what intermediary space does it occupy?

Here it becomes necessary to differentiate between "good" and "not-so-good" metaphors. Some metaphors "click", some don't. Saying that "the lion is like a tiger" is clearly a metaphor of the latter kind. The similarity between the two animals is too great to be interesting. The statement is uninformative about the empirical world. The same is true, but for the opposite reason, with the metaphor "the lion is like a pencil". Here the difference is too great. Somewhere in between these two examples lies the classical "the lion is like the king", endlessly portrayed in front of countless royal and national palaces. Still more exactly positioned, though between other extremes, lie such metaphorical strokes of genius as T. S. Eliot's:

Our dry voices
When we whisper together
Are tired and meaningless
Like wind in dry grass
Or rats' feet over broken glass
In our dry cellar.

How to we hit this point? This is where the art of metaphor use, and the art of theory building, comes in. Rogers and Rogers refer to this as the art of finding and utilizing an "ambiguity favorable to resemblance" between described and descriptor. It is the art of constructing a paradox - a glittering arrangement of likenesses and differences - that evokes, in flashes of insight, a new understanding of the terms being compared.

It follows from this that metaphorization - and theoretization - is a non-determinate, fundamentally unpredictable process. Theory does indeed produce new knowledge about the world - it even supplies us with a backward kind of guarrantee of the objectivity of this knowledge, in the sense that the flashes of insight that a good theory evokes in themselves signify that a vital relationship has been established between the model and the world. But the perspective from which knowledge is obtained is always that of the blind man (or the blind flea) on the elephant. Each theory supplies knowledge, but very partial knowledge, of the world. And the extent to which this partial knowledge is relevant for an understanding of the larger whole varies dramatically from situation to situation. If Marx is to become useful for analysis of face-to-face interaction, his theory will obviously have to undergo some rather fundamental alterations. Applying Bateson or Goffman to an analysis of power would be equally foolhardy. The by now classical quarrel between Barth and Geertz is also a classical case of complementary points of view competing rather than cooperating.

And finally, I would like to quote a story that I cite in Chapter Three of my book, which I feels expresses part of what I have been trying to say. It is an example of academic creativity from one of the "real", hard sciences, which may perhaps contain a lesson to us "soft" sciences as well. Back in the 1920's, the physicist John V. Atanasoff became preoccupied with the idea of building an electronic computer. He worked on the project for years, pondering alternative solutions and deciding on some general principles - but by the winter of 1937 success seemed further off than ever. We shall let him tell the rest of the story himself, as q uoted in Scientific American:

"The evening had not begun with particular promise. It had, in fact, been so frustrating that he left his laboratory, got into his car and began driving eastward from the college at Ames at high speed, concentrating on his driving to take his mind off his troubles. After several hours he ended up some 200 miles away in the state of Illinois, where he stopped at a brightly lit road-house for a drink. 'It was extremely cold and I took off my overcoat,' he recalled... 'I had a very heavy coat, and hung it up, and sat down and ordered a drink, and as the delivery of the drink was made, I realized I was no longer so nervous and my thoughts turned again to computing machines. Now, I don't know why my mind worked then when it had not worked previously, but things seemed to be good and cool and quiet... I would suspect that I drank two drinks perhaps, and then I realized that thoughts were coming good and I had some positive results.'" (Mackintosh 1988, p.73)

That evening, Atanasoff essentially invented modern computing.

I would like therefore to make a call for a less puristic, more eclectic, theoretically more cooperative and intellectually more tolerant, anthropology.


Parts of this paper reflect ideas that were presented and discussed during an MA course I held at the Institute of Anthropology, University of Copenhagen, during the winter of 2000-2001. The course was entitled: "Er antropologi et komparativt samfunnsfag?" - "Is Anthropology a Comparative Social Science?" I would like to thank all the students who participated in this course for the many stimulating discussions we shared.


Barth, Fredrik (1959): Political Leadership Among Swat Pathans, London: Athlone Press.

Barth, Fredrik (1966): Models of Social Organization, London: Royal Anthropological Institute. Royal Anthropological Institute Occasional Papers, No. 23.

Bateson, Gregory (1936 [1981]): Naven. A Survey of the Problems suggested by a Composite Picture of the Culture of a New Guinea Tribe drawn from Three Points of View, Stanford: Stanford University Press, Second edition.

Bateson, Gregory (1972 [1976]): Steps to an Ecology of Mind, New York: Ballantine.

Eliot, Thomas Stearns (1925 [1963]): "The Hollow Men", in: Collected Poems 1909-1962, London: Faber and Faber.

Mackintosh, Allan R. (1988): "Dr. Atanasoff's Computer", Scientific American, August, p.72-78.

Parkin, David (1987): "Comparison as the Search for Continuity", in Ladislav Holy (ed.): Comparative Anthropology, p.52-67, Oxford: Blackwell.

Rogers, Franklin R. and Mary Ann Rogers (1985): "Literary Form. The Topology of Metaphor", in: Painting and Poetry. Form, Metaphor, and the Language of Literature, p.105-124, Lewisburg: Bucknell University Press.

Sharp, Lauriston (1952): "Steel Axes for Stone-Age Australians", Human Organization, Vol. 11, p.17-22.

Sørhaug, Tian (Hans Christian) (1996): "Hvordan ting blir sagt med ting", in: Fornuftens fantasier. Antropologiske essays om moderne livsformer, p.90-111, Oslo: Universitetsforlaget.