 |
 |
Introductory models & basic concepts: complexity theory
|
|||
|
Introductory models & basic concepts: complexity theory
|
During the 1980s the study of complexity, often more popularly known as chaos theory (though some argue that, strictly, there are differences between the two which I won't go into in this brief overview - and probably don't understand anyway), took off very rapidly at a variety of research institutions. Foremost amongst those is the Santa Fe Institute in the USA, where genuinely interdisciplinary research programmes are taking place. The 
Santa Fe Institute brings together many highly original researchers - often Nobel laureates - from a variety of different fields. If you want some names to follow up, then Gell-Mann (Physics), Holland (Biology and just about everything else), Arthur (Economics) and Langton (Artificial life) will serve as well as any. Arising initially from the 'hard' sciences, the study of complexity rapidly spread to economics and it seems it is being applied to the 'fuzzy' sciences of Sociology, Psychology etc.
You may recall that one of the criticisms we have made of much communication theory is that the models are unrealistically linear. Now, the term 'linearity' employed in this critical sense has become almost a term of abuse in some areas of cultural studies, where it seems to be used to refer dismissively to an attachment to supposedly outmoded and discredited 'linear' Enlightenment thinking, as opposed to some new kind of preferable 'non-linear' thinking, which allegedly parallels the insightful thinking of chaos theory's study of 'non-linear dynamics'. There is an unfortunate wooliness in this use of the term, since 'linear', when used as a criticism of certain communication models, is used in a non-technical sense, and yet the highly technical science of non-linear fluid dynamics is pointed to as the justification for the criticism.
To criticize some communication models
for being 'linear' in the non-technical sense strikes me as a rather crude criticism since virtually all models which derive from Shannon-Weaver do include some kind of feedback
loop. Indeed, I think one of the great contributions of cybernetics to our understanding has been to increase our awareness of the importance of such circular processes.
Confusingly, many cultural theorists who find in chaos theory an exemplar of the
'postmodern science' they advocate highlight its alleged challenge to Newtonian mechanistic
linearity, whereas in fact
Newton's 'linear thought' uses equations that are perfectly non-linear; this is why many examples in chaos theory come from Newtonian mechanics, so that the study of chaos represents in fact a renaissance of Newtonian mechanics as a subject for cutting-edge research. Likewise, quantum mechanics is often cited as the quintessential example of a 'postmodern science', but the fundamental equation of quantum mechanics - Schroedinger's equation - is absolutely linear.Sokal & Bricmont (1998 : 135)
Transmission models are criticized also for presenting an oversimplified view of the process of communication. I'm not sure that that's an entirely acceptable criticism either. After all, models are not supposed to be a complete representation of the world; they are supposed to be a simplified abstraction from the available data. The Shannon-Weaver model, when applied to human-to-human communication, may look simple, but, when the feedback loop is added in, it is immediately suggestive of great complexity in the real world.
A more significant criticism of models drawn from information theory or first-order cybernetics is perhaps that they tend to be concerned primarily with equilibrium maintenance/homeostasis. There is primarily a concern with restoring a system's equilibrium whenever it is disturbed by external influences and there is thus a greater emphasis on negative than on positive feedback. Some cultural studies theorists, oversimplifying greatly, claim that this form of cybernetics has in common with the Newtonian conception of an orderly universe that it seeks mastery of the environment and relies on the assumption that, though it may be difficult to achieve, such mastery is obtainable.
Complexity is the science of 'non-linear dynamics'. That terminology immediately suggests that it must be of some use to the investigation of communication (though the suggestibility of the term may depend on a 'common-sense' understanding, rather than on the distinction in mathematics between linear and non-linear equations (see Sokal (1998 : 133 et seq.) for a discussion of this). I have been able to find some examples of its application to information theory, but I am as yet unclear about its application to communication studies.
What follows, therefore, is stuff I don't know much about and don't understand very well. All I have done is to single out from complexity theory some of the features which seem relevant to us. I hope it will be of some help to you when dealing with questions on the relevance, suitability and applicability of communication theory. A word of caution, though: I am often skeptical about the enthusiastic adoption of theories from hard science and their application to the soft sciences, to a large extent because the 'soft scientists' simply don't understand what they're talking about.
If you're beginning to feel at this point that this is all becoming a bit confusing, well, you're not alone. Cultural studies, in its more postmodern incarnations, frequently draws on concepts from the 'new physics' and from 'postmodern science' to support the theses being advanced, but equally frequently uses the terminology in ways confusingly different from the ways it is used in science. Now that might be acceptable, if the purpose were to draw a rough analogy between some aspect of cultural theory and scientific theory, or to draw on the science as an enlightening metaphor, but in most cases it seems rather to be an excuse for mere verbal pyrotechnics, a show of cleverness and erudition (often quite hollow) and an excuse for even greater obfuscation than is otherwise typical of postmodern texts. So we find in Lacan a mass of incomprehensible waffle about mathematics and topology, in communication theory fundamental misunderstandings of entropy and redundancy and a wholesale trendy adoption of concepts such as indeterminacy and uncertainty taken in a slapdash manner from quantum mechanics. Much of this can lead to half-baked ideas masquerading as science, so be aware (especially since I am formally certified as totally innumerate) that what follows may well be just as half-baked. I am very conscious in writing this of Sokal and Bricmont's in my view wholly justified demolition of the nonsense spouted by Luce Irigaray, nonsense supposedly supported by her knowledge of fluid dynamics. So I am also very much aware that I am probably going to get a lot of it round the back of my neck. However, it is the case that many of the scientists involved in research into chaos theory and complexity theory themselves claim for their research area an applicability to numerous soft sciences. Whilst I am aware that the drive to secure research funding may lead them to make exaggerated claims, an attempt to state what their claims appear to be seems justified, even though I undertake it with some trepidation.
Newtonian mechanics is entirely deterministic. This was particularly emphasized by the popularisers of Newton in Enlightenment France. The scientist Laplace summed this up by saying that if he knew the current state of all the molecules in the universe he would be able to predict the future of the universe for ever (and, by extension, the whole history of the universe up to
now - it's worth noting in passing that Laplace did not actually imagine that we
would ever have that perfect information and that he made his comment in the
context of an essay on probability theory ). We human beings might feel rather uncomfortable with that because, of course, we are also made up of
molecules. If Newton's laws of motion mean that the current state of all molecules determine the state of all molecules at the next instant, then free will is a meaningless concept. Voltaire, in typically acid style, summed up the consequences:
It would be very singular that all nature and all the stars should obey eternal laws, and that there should be one animal five feet tall which, despite these laws, could always act as suited his caprice. It would act by chance and we know that chance is nothing. We have invented this word to express the known effect of any unknown cause.
There's the rub, of course - unknown cause. The simple fact is that we don't know the current position of all the molecules in the universe. For a single spoonful of air we would need to know the position and velocity of 1020 molecules bumping into one another about 6 x 109 times a second. The physicist Michael Berry considered a collection of oxygen molecules at atmospheric pressure at room temperature. He imagined a single electron placed at the edge of the known universe (somewhere around 1010 light-years away). After how many collisions would a given molecule in the oxygen miss a collision with another molecule which it would not have missed had the electron not been there? Remember that the electron is affecting the oxygen only by its gravitational field, which must be so weak that we can virtually discount its effect. You might think so, but in fact Berry calculated that the oxygen molecule would miss its collision after a mere 56 collisions. The amount of information we would require and the number of calculations we would have to perform to arrive at Laplace's prediction is so huge that we simply couldn't do it, presumably not ever. Note, though, that the fact that we can't perform the calculations in practice does not mean that the universe does not behave according to entirely deterministic laws. In practice of course, this doesn't normally matter much. We may not be able to predict the position of a given molecule in a gas when we compress it, but for the gas as a whole Boyle's law applies. If we make a 10% error in measuring the volume of the gas or the pressure applied, our predictions of what happens when the gas is compressed will be 10% out.
Some researchers in complexity seem to be convinced that their study represents a major paradigm shift in the sciences and in human knowledge more generally; indeed it is sometimes referred to as 'the new science'. Prigogine emphasizes that the mechanistic and deterministic Newtonian world-view - emphasizing stability, order, uniformity, equilibrium and linear relationships between or within closed systems - is being replaced by a new paradigm. This new paradigm is more in line with today's accelerated social change and stresses disorder, instability, diversity, disequilibrium, non-linear relationships between open systems and temporality - the sort of terminology you will find cropping up repeatedly in the writings of postmodern theorists, though I'm not personally convinced that they understand it any better than I do.
One of the principal focuses of complexity study is complex adaptive systems. As far as I can see, it's not at all clear how complexity is to be defined, some writers describing a snowflake as 'complex', others describing it as 'merely complicated'. Here are what Holland considers to be the main features of complex adaptive systems:
A butterfly flapping its wings over Cuba in August could influence the course of a hurricane heading for Florida in September. Virtually everything and everyone in the entire world is caught up in a vast nonlinear web of incentives and constraints and connexions. The slightest change in one place causes tremors elsewhere. Under the right circumstances, the slightest uncertainty can grow until the system's future becomes entirely unpredictable - i.e. chaotic.
Again, I should perhaps emphasize that this does not mean that the system does not behave according to deterministic laws; it 'just' means that we cannot have enough information to figure out what the effects of those laws might be.
Sensitivity to initial conditions was sketched out by Poincaré in 1903:
It may happen that small differences in the initial condition produce very great ones in the final phenomena. A small error in the former will produce an enormous error in the latter. Prediction becomes impossible, and we have the fortuitous phenomenon.
Poincaré quoted in Peterson (1993 : 167)
However, this 'butterfly effect' is normally associated with Edward Lorenz, who in 1961 was using computers to model meteorological systems. One day he wanted to make a closer examination of one particular sequence on a previous printout. Instead of starting the run all over again, he simply entered the data from midway through the run. He set the computer running and went off for a cup of coffee. When he returned he found that the results were completely different from the original printout. The 'weather' had developed in an entirely different way. On examining the data, Lorenz realized that he had typed in .506 as his starting value, whereas he should have entered .506127. He had assumed that such a slight variation would make no difference. In fact the difference was such that he might as well have entered a randomly chosen number. (source: Gleick (1987))
To come back to Laplace's molecules for a moment: if you are attempting to measure the position of one of them with a ruler, what happens if you place the molecule on an irrational number on the ruler, say 
? Expressed as a decimal, such a number goes on for ever, so you have to chop it off at some point. As we have seen, truncating it will have
unpredictable consequences for our calculations. The 'Laplacian' ideal is in practice impossible to realize.
In the social sciences, we might expect complexity theory to show the following features:
Artificial life suggests that a population of simple elements following simple rules of interaction can display exceedingly complex behaviour.
Neural networks suggest that the more densely the neurons are interconnected the less likely they are to cycle through a limited number of states or ever repeat the same state. The same probably applies to human networks: the more interdependence grows, the less likely it becomes that history will ever repeat itself. It therefore becomes increasingly difficult, if not impossible, to make any predictions on the basis of previous experience.
The above might suggest that communication models and theories have, at best, explanatory power, but not predictive power. In other words, if you accept that view, communication theory can offer a broad, retrospective explanation of how a given communication or set of communications evolved, but cannot predict how a new communication will evolve. This is not to suggest that complexity, chaos theory, complex systems theory or whatever you want to call it can predict nothing. Quite the contrary is the case. Complexity theory has already had considerable impact on our lives - for example, fractal mathematics has greatly improved compression of data, thus greatly speeding up data transmission and making more efficient use of available bandwidth; our understanding of heart arrhythmias and the functioning of the brain has been greatly improved by complexity research; genetic algorithms are being succcessfully applied to economics and stock market movements and complexity theory is being applied increasingly to the study and management of modern organizations. What I am suggesting, though, is that I suspect complexity theory says to the student of human communication that we are likely to be able to predict only in broad outline how successful a communication is likely to be and that we are likely to be seriously disappointed if we expect the process to be as predictable as Newtonian billiard balls.
If at any future point I feel I have a clearer understanding of this, I'll add some more here - but don't hold your breath.
The Lasswell Formula
The Osgood and Schramm Model
Gerbner's Model
Santa Fe Institute
Nonlinearity and Complexity Homepage (disappeared?)
Artificial Life and Genetic Algorithms at Brunel University
|