Science, social science and the Dig Tree

Introduction

Every Aussie schoolkid knows the story of Burke and Wills and the Dig Tree: how these brave explorers tried to traverse the continent; how they crossed vast expanses of unknown country; how, in April 1861, they returned defeated, starving and exhausted to their depot on the Cooper Creek; and how they found there the tree emblazoned with the word DIG. It was to guide them to some buried supplies left by their support party who, sadly, had given up hope and departed only hours before. And everyone knows the tragic ending: how the explorers soon perished nearby, lacking the knowledge, the bushlore, to sustain themselves on the bush tucker all around.

Science is like that. Or at least, the relations between science and the humanities are. Here, there is a vast unexplored terrain between the sciences and the humanities - C P Snow's two cultures [1]. The disciplines of science, call them settlements or sometimes cities, are scattered along one seaboard of this unknown continent, and are well connected with each other through a busy network, for one of the key features of science is its continuity. On the far shore, the fatal shore perhaps, lie the humanities whose settlements are not so well connected. Here we find many city states, an Athens, a Florence or two, but also some Port Essingtons.

There are also expeditions exploring across this country from the science shore, for exploration is another key feature of science. They are usually called social science, since they try to apply scientific ideas to the phenomenon of man. There are depots and Dig Trees littering the landscape of the interior, and there are heroic failures as the wrong knowledge, maladaptive knowledge, is used in exploring this strange country.

In this paper, I will argue that it is time for a new expedition up-country, to find this strange thing called sustainable development. If it exists at all, it must exist there, because it demands the integration of biophysical and socioeconomic knowledge. Thus it lives, like Lasseter's Reef, somewhere out there in that unexplored country. But before we set off to traverse terra incognita, I want to see if we can learn anything from some of those earlier stalled expeditions, those Dig Trees.

Depots and Dig Trees

Social science has established two major depots out there today. Each has been provisioned by scientific ideas, each has got some way into the interior, each has faltered and can go no further, each has emblazoned a Dig Tree. One is usually called economics, and the other we can probably sweep up under the name psychology.

In order to understand how these great expeditions, these great intellectual adventures, went wrong, we need to understand something about why their provisions failed to sustain them for their journey. And for that we must go back a few centuries to the Renaissance when science first began to emerge as a distinctive way of knowing the world. The science of the renaissance pushed man from his privileged position at the geographic centre of God's universe. Galileo, with his telescope, showed that we inhabit a tiny planet, orbiting an average star in an unremarkable galaxy, adrift in a vast impersonal cosmos.

There were two quite different reactions to this momentous discovery in the Enlightenment, the era that followed. The first was a great flowering of science and scientific hubris [2]. It was the time of Sir Isaac Newton and the idea was abroad that Newton's science would eventually reveal all of nature's secrets. The hubris even affected the arts. Here is Alexander Pope:

Nature, and Nature's laws lay hid in night.
God said, Let Newton be! And all was light.

But there was also a great flowering of a philosophical movement to prop up the special role of man in the face of the onslaught of science. RenÈ Descartes codified a far older philosophical idea of the importance, the uniqueness of the human mind that led to a widespread dualism in the actual approach to understanding the world. It is but a short step from the idea of mind-body dualism to man-cosmos dualism, and the distinction between subjective and objective knowledge.

Unfortunately, perhaps, Charles Darwin comprehensively trashed this tidy and comfortable idea in the 19th century. The science of Darwin showed that man was not unique among God's creatures, but, like all other living things, an accident of fate, evolved by chance through natural selection, a process that Darwin described in a letter to the botanist Joseph Hooker as one of 'the clumsy, wasteful, blundering, low and horribly cruel works of nature' [3]. For the first time, man had a scientific framework with which to understand himself [4].

This little historical excursion allows us to understand why the two earlier expeditions stalled. They were essentially pre-Darwinian. Each enthusiastically embraced the ideas of Newton: the ideas of scientific laws, of an elegant physical simplicity underlying the apparent messiness of the workaday world. They also embraced the ideas of Descartes: the idea of objective knowledge, and of reducing complex things to their simpler components. But they missed one of Darwinís messages: that there is nothing special about man and all his works.

Economics, our first Dig Tree, missed Darwin spectacularly, and built an essentially Newtonian discipline in order to explain human behaviour. In a bid to become more 'scientific', economists proposed that economies be thought of as if they were the simplest physical systems with a single point of stability, and governed by a strictly linear dynamics - the simplest kind of Newtonian dynamics.

They even invented an 'invisible hand' to replace Newton's God to rule serenely over these mechanics, and have then packaged it all up into a central dogma called the 'general equilibrium theory'. The words of Leon Walras, the 19^th century originator of the theory, show clearly its scientific pretensions:

We all accept the current description of the universe of astronomical phenomena based on the principle of universal gravitation. Why should the description of the universe of economic phenomena based on the principle of free competition not be accepted in the same way? [5]

Pretensions that ignored the observed reality, as the most prominent economic heretic, John Maynard Keynes lamented:

The atomic hypothesis that has worked so splendidly in physics breaks down [in economics]. We are faced at every turn with problems of organic unity, of discreteness, of discontinuity ñ the whole is not equal to the sum of the parts, comparisons of quantity fail us, small changes produce large effects, the assumptions of a uniform and homogeneous continuum are not satisfied. [6]

But this self-evident complexity has not been sufficient to turn the broad church of economics from its defining idea. As a result, economics has marked a Dig Tree with a parody of Newtonian physics, instead of creating a scientific understanding of economic behaviour.

The second Dig Tree, the one I called psychology, is really broader than that. I am really thinking of that whole broad area of what might be called 'experimental' social science, the sort that revels in the collection and mathematical analysis of observations of the human condition. This also is mostly pre-Darwinian in conviction if not application, but pre-Darwinian in an interesting and subtle way, because it derives from a funny sort of pre-Darwinian biology. Let me explain.

This biology story, like biology itself, has a few more twists and turns than the physics story, but it is essentially the story of statistics and agriculture and Sir Ronald Fisher [7]. It is the story of experimental biology, the biology that sought to ape physics, but yet could not, until the idea of statistics had taken root. The clean serene equations of Newton's physics were just not reachable from the messiness of biology until the idea grew that the messiness might merely be statistical noise masking an ideal process or parameter, much as measurement error or noise in physics masks the true value. Such a platonic view of living things, of course, is instantly commensurate with the essentially platonic physics of Newton, and so it allowed a flowering of a particular sort of biology, once the appropriate tools had been invented. This sort of experimental biology depends utterly on the use of mathematical statistics, and on the mindset that goes with it. It was seen first in physiology and reached its apogee in agricultural research where the physical layout of experimental plots mimicked deliberately and exactly the layout of the statistical analysis. From there, in the early years of this century, it moved across to the 'experimental' social sciences.

We may note here that the provisions offered by this sort of biology were tools like Fisher's analysis of variance, which makes two heroic Cartesian assumptions: that the living phenomenon we are interested in may be usefully broken up into a set of more or less independent components; and that we know what those components actually are. Other less ambitious tools were also invented for those who shied away from the latter assumption, but still felt that no progress could be made without the former. These tools, like factor analysis and principal components analysis, assume only that living phenomena are decomposable and that, while we may not know the components yet, we might be able to discover what they are.

Both sorts of tools, by ignoring the contingent complexity of biological phenomena, including the human phenomenon, have caused great damage to both biology and the social sciences during this century comparable to the damage inflicted by physics on economics during the last. Ronald's Dig Tree points to provisions that cannot take us any further.

Complexity

What they both missed is that the complexity matters. By concentrating on the search for the supposed underlying simplicity of the world, by reducing the manifest complexity of the observed whole into more manageable, simpler, and more fundamental parts, these two expeditions followed an essentially Renaissance path to a knowledge of man. They were deaf to Darwin's call that the richness and diversity of life is the stuff of life itself. In his time, Darwin could do no more than record and explain it, but that was enough to change science and its view of man forever. By ignoring the complexity, the earlier expeditions were ignoring the bush tucker all around. By ignoring Darwin, they were ignoring the bush lore on how to use it.

Darwin changed science by bringing all the difficult complex phenomena back into the ken of science. Before Darwin, many messy physical phenomena, and particularly many biological phenomena, and especially most socioeconomic phenomena were considered refractory to the scientific method, and by implication were not the proper concern of scientists at all. They were handled by other methods: historical, narrative, ideological, rhetorical, philosophical and so on. But ever since Darwin, science has become fascinated by phenomena such as ecosystems and immune systems, economies and the human mind with their messiness, fuzziness, incompleteness, novelty, surprise, adaptation - in short, with their irreducible and contingent complexity. It has developed a wide range of tools not only for acquiring the data about such systems, but also for analysing and visualising them. These tools are usually computer intensive and rejoice in such techno-names as adaptive game theory, simulated annealing, neural nets, fuzzy logic, genetic algorithms, cellular automata, spin glasses, and agent based modelling.

Together with the tools has come an approach, called complex systems theory, which offers a qualitatively new way of doing science [8, 9, 10, 11, 12]. Where traditional science sees the search for simplicity and natural law as the goal, the new theory sees the search for emergent structures and dynamics. Where the old science gives primacy to testing hypotheses, the new encourages generating them. Where the old demands objectivity and the separation of observer and observed, the new sees no such distinction, encouraging interaction and recursion between them. The new science of complex systems tries to build exploratory tools, where the old constructs predictive ones.

The approach of complex systems theory, though, is ineffably scientific. It is not some woolly, 'anything goes since everything is relative' belief system. While it does say there may not be simple answers to the way the world is, it does not say any answer is as good as any other. While it does say that we may not yet have the right answers, it also says that many answers - nonscientific answers - are just plain wrong. On all those issues it is as one with traditional science. It fully acknowledges its scientific patrimony.

Learning

Complex systems theory is evolving fast. It already acknowledges a raft of complex physical systems, such as the weather or the oceans, as suitable objects of study. But it also distinguishes a special class of complex systems called complex adaptive systems. These include all living things and their parts - cells, say, or immune systems - or their assemblages - societies, economies, ecosystems and so on - that take this complexity further in exhibiting the peculiar characteristic of learning, of evolving, of adapting. The idea of complex adaptive systems is broader than just living things, however, since it includes other phenomena, such as 'artificial life' computer programs that exhibit adaptive and evolutionary behaviours.

In a lovely irony, it is in the use of these computer programs as exploratory tools, that a new understanding of human phenomena is emerging. We need, it seems to set a thief to catch a thief.

As an example, consider the recent work of Joshua Epstein and Robert Axtell of the Brookings Institution [13]. They talk about growing artificial societies and about social science from the bottom up. Using cutting-edge adaptive modelling techniques, they study the emergence of group behaviour, transmission of culture, knowledge and disease, the generation of trade and economic behaviour, and the development of combat. Their models are based on individuals, who themselves have none of these behaviours or adaptations. Instead, as in real life, the behaviours emerge from the interactions between and among individuals and their environment.

These are pioneering works, but they show great promise. They offer a profoundly Darwinian and scientifically robust alternative to the sterility of the pre-Darwinian Dig Trees. And they are richly recursive in their rejection of the false Cartesian dichotomy. As Epstein and Axtell point out [14]:

Just as the community of biologists had to learn to fully exploit the microscope when it was first invented, so we have only begun to explore the uses and limits of the artificial society as a scientific tool. We can only hope that the field itself will display the evolutionary process it studies - new agents join, and intellectual heterogeneity grows; social networks of scientists endogenously take shape; selection pressures operate; and from the social enterprise of agent-based social science, interesting things emerge!

Learning communities, striving for regional sustainability, might just find their approach and their tools right here.

References