In the early 1960’s Edward Lorenz, an MIT meteorologist who just died April 16, 2008, was working on computer simulated weather forecasting models. His testing of what was then a twelve-dimensional model required reiterated time series analyses, one of which went wildly awry while Dr. Lorenz went off for a coffee break. In sorting through what happened, Lorenz discovered that a small entry change in the initial data entry for the last simulation had produced large effects. A miniscule rounding error set the simulation off in a completely different direction.
This idea that small causes can have very large effects, what has now become the idea in chaos theory that outcomes can be very sensitively dependent on initial conditions, was popularized as the Butterfly Effect after a talk Lorenz gave in 1972 at the American Association for the Advancement of Science. Although his original metaphor for all this relied on seagulls and not butterflies, the talk’s title – Predictability: Does the flap of a butterfly’s wings in Brazil set off a tornado in Texas? – has become a popular way to make sense of micro-causes.
The comprehension of all the little causes that might set history into another dramatically different direction leads some to the view that Science can ever yield certain knowledge, although chaos theory is not typically defending the proposition that the world is unknowable. To the contrary, in some of its more influential articulations, including the popular and widely noticed work done at the Santa Fe Institute, the idea was that scientists might learn more from very simply computational models than from efforts to fully replicate nature inside the supercomputer.
But the determining confidence of Big Science did seem for a time to be thrown on the defensive by the emergence of chaos, and then network, theories. Partly this reflected the sheer hubris of some scientists. Ironically (because his point was actually to explore probabilities, not certainties), Leplace is often trotted out as Exhibit A:
An intelligence that, at a given instant, could comprehend all the forces by which nature is animated and the respective situation of the beings that make it up, if moreover it were vast enough to submit these data to analysis, would encompass in the same formula the movements of the greatest bodies of the universe and those of the lightest atoms. For such an intelligence nothing would be uncertain, and the future, like the past, would be open to its eyes.
Edward Lorenz was not the first to defend the small cause/large effect idea, of course – Henri Poincaré came to similar conclusions in laying out his work on differential equations and astrological motion, and one might say that Ray Bradbury anticipated Lorenz with his short story, A Sound of Thunder, which considers the world changing consequence triggered when a time traveler does nothing more than accidentally kill off a prehistoric mouse – but perhaps it is the power of the metaphor that has given it such currency in both popular culture (you’ll recall a fairly popular thought pretty bad movie premised on the idea just a few years ago) and in the humanities.
The small insight often credited to Lorenz, that because of unknowable small causes the weather might be fundamentally impossible to predict even three weeks into the future, itself set in motion major changes in almost every area of science. One his colleagues at MIT eulogized him by noting that, “by showing that certain deterministic systems have formal predictability limits, Ed put the last nail in the coffin of the Cartesian universe and fomented what some have called the third scientific revolution of the 20th century, following on the heels of relativity and quantum physics.” In the citation he received in winning the Kyoto Prize, the committee said his insights “brought about one of the most dramatic changes in mankind’s view of nature since Sir Isaac Newton.”
What does all this have to do with humanistic inquiry, or the disciplines that aim to track and discern underlying historical cause and effect relationships? One pernicious distortion, I think, is enacted by Malcolm Gladwell’s little book Blink, which lays out some of the research showing that gut instinct and immediate reactions can be just as reliable as those informed by years of study and scholarship. The upshot seems to be a certain anti-intellectualism that disparages the careful detail work of the academy and which elevates the immediate intuitions of students soaking up the first five minutes of a seminar (who famously are able to judge professors about as well as those with much longer interactions) and voters who don’t seem to actually know anything about their government but are defended as making instinctively right decisions by accurately reading the cues given off by politicians. David Brooks’ in my view absurd defense of the media frenzy over whether Obama should wear a flag lapel pin or not (Brooks sees it as a small but reliable indicator of the extent to which the Senator shares voters values) might be said to traffic in the same fallacies.
But neither Gladwell nor Brooks claims to make serious use of the Butterfly Effect at all; instead of arguing that these small indicators and assessments might take history in radically different directions than we imagine, they claim instead that small pieces of data give reliable insight into large scale questions that interests them. But they shed light on what a recent commentator (the science journalist Peter Dizikes) bemoaned in remembering Lorenz’ lifework: “Pop culture references to the butterfly effect may be bad physics, but they’re a good barometer of how the public thinks about science. They expose the growing chasm between what the public expects from scientific research – that is, a series of more precise answers about the world we live in – and the realms of uncertainty into which modern science is taking us.”
In the academy, serious application of the insights of chaos theory does more serious and defensible work, but may also perniciously subvert the work done in the humanistic disciplines, or at least the impulse that underwrites the enterprise. All sorts of interesting applications of chaos theory have been attempted to better explain human and social behavior. Just to list a few:
A. Two scientists, Andrzej Buller and Katsonuri Shimohara, both working in Japan, seeking to explain why human beings often hesitate to act in the moment even when the information available to them remains perfectly constant, have hypothesized that the brain, itself a vast neural network, may be generating small micro-reactions that trigger feelings of being wholly conflicted about a subject, whipsawing human subjects into otherwise inexplicable and mostly short-lived switches between, say, love and hate.
B. Katherine Hayles’ remarkably interesting work in the early 1990’s (two books especially, her Chaos Bound: Orderly Disorder in Contemporary Literature and Science [Cornell UP 1990] and the collection she edited, Chaos and Order: Complex Dynamics in Literature and Science [U Chicago P 1991] applied chaos theory to literary works in an effort to complicate the field’s understanding of textuality and intertextuality. Her claim at that time was that chaos theory bore an interesting resonating affinity to with poststructuralist critical work, which was to say that chaos theory and poststructuralism shared something of a common etiology (an example of this would be the potential affinity between Jacques Derrida and chaos theory’s mutual interest in ideas like iterability and folds). Although some of her early enthusiasm in pursuing these connections gave rise to criticisms that her argument had something of what Matheson and Kirchhoff called a “gee whiz” quality (31), Hayles was always, it seemed to me, careful not to argue for an overdetermined conceptual affinity. The result has been an evocative use of the language of chaos (even if only at times imported as metaphor) that has recast some of the most important questions of literary analysis.
C. Margaret Ward has explored the extent to which network and chaos analytical frameworks might be usefully deployed to account for complex family dynamics, an idea not altogether implausible given our common sense notions of how successful families operate in a sort of healthy equilibrium, and the connected sense of how healthy relationships can be set awry by apparently minor interactions that devolve in unhealthy ways, creating a kind of awful cascade effect. Tracking the conceptual affinities between chaos and family systems theory, Ward has argued that chaos theory adds important new insight to accounts of family interaction, explaining (for example) more adequately the relationship between order and disorder as they emerge in traditional families: “Chaos theory, far from being merely a source of new jargon, can provide a powerful tool for understanding families that are transitional or disordered” (637).
So what might be the potential problems of importing chaos terminology into humanistic work given these potentially productive provocations? One is the conceptual slippage that the very term chaos seems often to enable: at times critics exhibit a tendency to use chaos as a cover to argue for the wholesale indeterminacy of a text. Making the case for this against Gillespie’s The Aesthetics of Chaos, Danielle Follett observed that:
[T]he essential problem in these examples is whether there is anything not present in the text or in the spectrum of simultaneous (and equally valid) characterizations, that is, whether this spectrum is infinite…. For, it hardly needs to be added, there are world-views not represented in Beowulf, and in its fear of ‘exclusionary logic,’ this way of reading may indeed end up excluding Beowulf… Similarly, critics do not need the theory of a butterfly effect to emphasize the pivotal importance of seemingly insignificant details.
The pragmatic consequences of imported chaos theory may also be playing out in the historiographical debates over the relevance of histories imagined through counter-factuals. A number of popular books (e.g., Niall Ferguson’s edited collection Virtual History: Alternatives and Counterfactuals, appearing in 1999) consider the radically different futures that might have arisen had very small but key alternative outcomes have been realized. Counterfactual histories have been received with some difficulty in the traditional domains of historical scholarship – as Richard Ned Lebow put it, “They are flights of fancy, fun over a beer or two in the faculty club, but not the stuff of serious research,” reporting the conventional wisdom to problematize it.
Lebow’s main move is to interrogate the so-called bright line distinction between factual and counterfactual arguments (which leads him to insist that they differ in degree and not in kind), which then enables him to try out some potential standards for allowing what he calls “plausible” world counterfactuals and ruling out (as methodologically suspect) what he calls “miracle” world counterfactuals. He is led to make such distinctions because of some interesting examples where counterfactual analysis led to important insights in public decision making. Following the Aldrich Ames spy scandal the CIA ran a task force exercise where alternative spy catching procedures were imagined and then set against the case history for Ames to see whether (were they actually in place) he might have been caught earlier. The exercise produced interesting and valuable insights into their methods. And in spite of Lebow’s rendition of the disciplinary common sense, he points to seriously taken analyses essentially predicated on counterfactual analysis (such as the best postmortems of the Cuban Missile Crisis).
Lebow argues that “counterfactuals can combat the deeply rooted human propensity to see the future as more contingent than the past, reveal contradictions in belief systems, and highlight double standards in our moral judgments. Counterfactuals are an essential ingredient of scholarship” (557-558). This benefit can be alternatively articulated as offering a methodological corrective to the tendency of structuralist accounts to overdetermine human history; that is, attention to the small unseen turning points that may set history into radically different directions can offset the subsequent tendency to see unforeseen turning points as structurally overdetermined and somehow inevitable.
I site Lebow because it strikes me that his attempt to make a place in historical research (and in this particular instance the work of international relations scholars) suggests a useful path between the overdetermined histories of structuralism and the sometimes wholly contigent accounts of poststructuralist indeterminacy. Taking into account the mechanisms of chaotic unpredictability, one might say, sometimes makes the scholarly account more rigorous.
Is that a butterfly I see?
SOURCES: Andrzej Buller and Katsunori Shimohara, “On the dynamics of judgment: Does the butterfly effect take place in human working memory?,” Artificial Life & Robotics 5 (2001): 88-92; Peter Dizikes, “The meaning of the butterfly: Why pop culture loves the butterfly effect and gets it totally wrong,” Boston Globe, 8 June 2008; Carl Matheson and Evan Kirchhoff, “Chaos and literature,” Philosophy and Literature 21.1 (1997): 28-45; Michael Patrick Gillespie, The Aesthetics of Chaos: Nonlinear Thinking and Contemporary Literary Criticism (Gainesville: University of Florida Press, 2003), reviewed by Danielle Follett in MLN; Margaret Ward, “Butterflies and Bifurcations: Can Chaos Theory Contribute to Our Understanding of Family Systems?” Journal of Marriage and the Family 57 (August 1995): 629-638; Richard Ned Lebow, “What’s So Different About a Counterfactual?” World Politics 52 (July 2000): 550-585.