The Twin Poles of Order and Chaos

Click to increase image sizeClick to decrease image size

Notes

We might stretch the metaphor a bit further to suggest that one goal of psycho-analysis is to accept disorder as inevitable—conflict shall always be with us—while at the same time understanding that every bit of chaos has its hidden order and expectability. Stretched to this point, at least metaphorically, this is a very contemporary idea—order within chaos rather than chaos as the inevitable dissolution and outer bounds of order. See, for example, Prigogine and Stengers, 1984, and Sardar and Abrams, 1998.

This statement is a bit of a historical oversimplification. Ideas that children actually developed from simple to more complex abilities and that psychological capacities emerged over time were in the air as early as the late eighteenth century. Over two hundred years ago, Tiedemann published observations of his son from birth through his second birthday (Horowitz, 1987). Like all ideas that dramatically shift perspectives and paradigms, early ideas of ontogeny and epigenesis were taking shape by the early 1800s in biology and related fields.

I need to make explicit what I mean by disorder when applied to development and behavior for the very word itself generates a host of associations and meanings that are surely laden by our inherent reactions to instability and disorderliness. Throughout this paper, I shall use the term “disorder” in two related ways—to indicate variability that is not expected or explained by a particular theoretical perspective and deviation from a norm or average way of behaving, thinking, or perceiving believed to be appropriate for a given age.

It is worth noting that, like evolution and natural selection, development was (and is) quickly incorporated into social, economic, even political institutions and expectations (see Kessen, 1990). Indeed, evolution, development, and progressivism are a secular trinity for the late nineteenth and twentieth centuries. To evolve and develop is to embrace progress. To regress is to retreat. To be delayed is to be shamed, outside the mainstream. The “cultural internalization” of ideas as broad as evolution or development and their linkage with the “cultural ideal” of progress are probably inevitable—or at least understandable. But that trinity very often distorts, even disfigures, the idea of development beyond recognizable utility. At such points, it becomes important to reclaim and refigure the idea of development, to evolve another form.

Perhaps nowhere is the disjunction between development and growth more apparent than in instances of severe mental retardation in which children may grow physically but not psychologically (at least at the same rate).

Of course, this difference in efficiency is not a trivial fact, and many an error or miscalculation in drug dosage or attribution of understanding is made by not recognizing that young, immature systems do not function at the same level as mature or adult systems. Again, in a popular sense, just understanding these kinds of differences in functional efficiency is “thinking developmentally.”

Self-organization or new forms of typing abilities on a global scale may also be seen in the design of the typewriter. In 1873, an engineer named Christopher Scholes designed a typewriter keyboard with the familiar QWERTY layout which was intentionally inefficient. If a typist on a standard typewriter moved too fast, the keys jammed. The QWERTY layout slowed the typist down sufficiently to diminish this possibility. However, when this particular layout was mass-produced and typists across the world began to use it, their efficiency improved, other companies imitated the same keyboard layout, and soon this intentionally inefficient typewriter became the standard that typists used with great efficiency (example from Waldrop, 1992).

Psychoanalysis offered one of the earliest and most elaborate stage-based theories of development, but there are of course many others, dealing with cognition, language, motor skills, and social development.

Indeed, it is not uncommon for children learning a new skill to lose temporarily gains in other domains. Thus, recently gained motor skills may seem to be lost as new language skills came on-line. At least, gains in one area may be paralleled by the temporary slowing of gains in other domains.

The creation of new form and function is not limited to childhood. It may occur throughout life, as with the learning of a new language or a musical instrument in adulthood. Each of us struggling to learn a new skill as an adult has had the experience of painfully plodding progress initially, followed gradually by increasing proficiency and then by a sense of automaticity. Actions or thoughts initially taken in slow, conscious steps are smoothly integrated into a seamless whole that occurs nearly without conscious vigilance—a new skill has developed.

Wolff (1987, p. 240) has suggested that “the induction of novel behavioral forms may be the single most important unresolved problem for all the developmental and cognitive sciences.”

This too is an oversimplification, for we often make inferences about parts from our observations of the whole, whether these are animate or inanimate, physical or psychological forms. For example, we infer from the turbulence in a stream that there may have been an obstruction upstream. We infer the central location of a cerebral hemorrhage from a paretic gait and slurred speech. We infer a specific fantasy configuration from the stream of associations in an analytic hour.

Seeing an analytic hour as a complex system may be one solution to the very troubling issue of how we can ever know that what the analyst reports about what happened clinically in any one hour or series of hours represents a “verifiable” account. If we take the analyst’s report as one perspective on the complex system of analyst- analysand, so-called inter-observer reliability refers not to agreement about the analysand’s observed behavior but rather to how other analysts might think about the material as presented by the treating analyst.

Of course, even the terms “macroscopic” and “microscopic” are relative. Studying the chromosomal patterning in a genetic disorder is surely a more microscopic approach to studying the behavior of a child with a particular mutation. On the other hand, studying the length of the telomere or the process of cleaving DNA into relevant segments is a more microscopic perspective than the chromosome patterning. In other words, while there is often a implicit higher value placed on more detailed (e.g., more microscopic) data as being more likely to reveal the true nature of the sys- tem in question, the degree of resolution is more accurately judged by the question needing to be addressed. Counting the number of looks away may or may not be a relevant measure of the conversational synchrony between two individuals, though it may be one that is easily verified and agreed upon. The data required to study the behavior of any complex system will necessarily be more macroscopic than those data relevant to the function of individual components of the system. Data regarding the behavior of a complex system must take into account the interactions among a number of components.

Indeed, many theories of development from the late nineteenth and early twentieth century drew extensively on the laws of thermodynamics, particularly the second law, which allowed for the inevitable increase in disorder in the world—or the constant increase in entropy.

The science of “complexity” emerged in the early 1980s with its roots in chaos theory. See Gleick, 1987; Prigogine and Stengers, 1984.

The simplest example of a dissipative system may be a rolling ball. Classical views of the energy needed to start a ball rolling expressed a simple relation between the mass of the ball and its speed. But these simple, linear perspectives do not take into account such dampening forces as friction, which makes a simple linear problem non-linear inasmuch as the impact of friction will change the relation between mass and speed as the ball continues to move. Thus, more energy will be needed to keep the system moving. Otherwise, the ball will stop. This is the difference between a so- called closed system and an open one. The former is a system that is stable, at a state of entropic equilibrium. No new behavior can emerge without the addition of energy. Open systems, while often behaviorally stable, are always far from thermodynamic equilibrium because of the continuous flow of energy to keep the system functional.

There is, of course a familiar, perhaps even historically regressive tone to this kind of language—energy, equilibrium, steady states. It is reminiscent of the hydraulic, mechanical language of the earliest psychoanalytic models of mental functioning—drives seeking discharge, excess quanta of energy, mass, and force. It is also reminiscent of the applications to human behavior of principles of thermodynamics, particularly the second law regarding the inevitability of increases in disorder or entropy, to human behavior if the restraints of defense and ego are adequately in place. Even contemporary models of cognitive functioning use mechanical or computational metaphors, though this is beginning to change in some quarters (see review in Siegler, 1996). One difference in these views of complex dynamic systems is in the focus on variability, not as a signal of regression or dissolution, but rather as a desirable, necessary property of the system.

In a way, chaos is a paradoxical term, for in its technical usage it indicates not so much instability and disarray as another level of orderliness, which is hidden in apparently random behavior, a kind of “organized disorder.” The term also indicates a phenomenon that defies prediction, not because it is random and disorderly, but because it is not linear. Chaotic structures, functions, and behaviors follow non-linear dynamics and non-linear models. Development is inherently non-linear even though our long accepted ideas of stages, phases, and apparently dramatic transitions in developmental abilities lead us to think of a relatively stable, relatively programmed process.

The “butterfly effect” was first noticed in studies of weather patterns and weather forecasting. Very small changes may be quickly magnified through various feedback loops so that a tropical rain cloud generates a hurricane hundreds of miles away. In other words, theoretically, even large-scale changes can come from small perturbations such a gale force winds can eventuate from a change in air currents produced by the flapping of a butterfly’s wings. The elegance of the so-called butterfly effect is that it involves randomness—that is, because of such sensitivity to external conditions, weather patterns (and other behaviors) are never quite predictable—but at the same time, accounts for richly complex, ordered (though non-linearly) behavior. For an accessible description, see Gleick, 1987.

The idea of infinite complexity within finite constraints emerged from a remarkably simple, yet intuitive, question and classic paper, “How long is the coast of Britain?” (Mandelbrot, 1977). The answer depends on one’s perspective. If we mea- sure in miles, the answer will be different from a measurement made in inches or centimeters. The closer we look, the more detail and variety we see. What we observe de- pends on where we are positioned as observers and how we make our measurements. What looks flat from afar is three dimensional up close. Thinking in fractions of dimensions or fractals (e.g., 2.2 dimensions, not just two then three dimensions) captures the idea of moving continuously closer or farther away as an observer and thus seeing more or less of the targeted observational field.

Perhaps the clearest example of an attractor state comes from watching a ball rolling in a bowl. Unless disturbed, the ball eventually settles to the bottom of the bowl. This is the attractor point or state. Consider too that pendulums tend to swing regularly, it does not usually snow in July in temperate climates, and infants tend usually to babble before they speak in words. Each of these “on average” behaviors de- fines a set of attractors or preferred states.

The dynamic, changing behavior of complex systems along their attractor states may also be represented graphically in so-called phase space. We are accustomed to looking at representations of three-dimensional objects, such as a building, in two-dimensional drawings. But behavior can also be represented in two dimensions such as the swinging of a pendulum, where one axis shows the horizontal location and the other the vertical location. At any point in time, with these two points, we can locate the pendulum in space. Phase space takes this representation of motion one step further and represents change or motion. One axis represents, for example, the velocity of the pendulum, another the location on the vertical axis, another the point in time. Phase space diagrams take information from moving, changing parts and represents the behavior of the whole system over time. If the behavior is steady and unchanging, the phase space diagram is a single point. If the behavior is repetitive or periodic, the phase space diagram shows a loop. If the behavior is a mixture of repetitive and changing patterns (e.g., order and disorder), the phase space diagram assumes the shape characteristic of many complex systems and represented prototypically by the Lorenz attractor (see figure 2; Lorenz, 1995). The Lorenz attractor, initially described in studies of weather patterns, was dubbed “strange” (Ruelle, 1993) because of the intricate phase space diagram that depicts non-linear, complex, non-predictable processes. The more degrees of freedom in the system, the more complex the phase space diagram and the “stranger” or more complex the set of attractor states.