Tourists in an unknown town: Remapping the social sciences
>> Andrew Abbott, AM'75, PhD'82, proposes a theory of research in the social sciences that follows a fractal pattern.

There's a curious thing about fractals. Once you recognize a pattern, you begin seeing them all over the place: in the tangled tops of crab apple trees, in the early winter waves crashing against the concrete riprap along the South Shore, and-in Andrew Abbott's case-in the way knowledge moves through the social sciences.

The Ralph Lewis professor and chair of sociology and in the College, Abbott has written a book on his observations, Chaos of Disciplines, published this December by the University of Chicago Press. Abbott maps the patterns of social science research during the last century and lays down a set of principles by which it is organized-or, he would argue, by which the research in fact seems to organize itself. Passionate disagreements among scholars, the rise and fall of schools of thought, the revolutions that occur in one discipline when scholars borrow (even "thieve") ideas from colleagues in others; all of these, Abbott says, are part of a wonderful pattern that is fractal in its nature. Fractals, complex yet self-similar geometric shapes, are capable of accounting for the irregularly shaped objects or spatially nonuniform phenomena in nature that Euclidian geometry can't digest. Once it's apparent, Abbott's fractal offers social scientists a handy set of practices to go about their work of discovering the mysteries of social life in a more efficient and complete manner.

"The real problem that drove this," says Abbott during a conversation in his office in the Social Science Research Building, "is that I kept running into people who appeared to disagree profoundly but, when you put them in another context, appeared to be arguing for exactly the same thing as the people they had been arguing against in the other context."

Abbott has argued with many a social scientist during the 18 years since earning his Ph.D. in sociology at Chicago, joining the faculty in 1991, and publishing two books, The System of Professions (University of Chicago Press, 1988) and Department and Disciplines (University of Chicago Press, 1999). But perhaps more pertinent to his views about fractal discovery is that he frequently loses these arguments. And losers take note of the points on which the winners trounce them. Indeed, as a self-described "eclectic" who would rather take the best of several points of view than live within certain "obnoxious" intellectual boundaries, Abbott says he is loath to disagree with any intellectual position at all.

A lanky, laid-back man, Abbott has deep-set eyes and thick, wavy brown hair that replicates itself in smaller scale in his bushy eyebrows and the tufts on his knobby knuckles. He is tall and angular, and he props a sprained ankle on a chair beside him-a soccer injury from the weekend past. On the wall above his head, framed black-and-white photos of his Harvard undergraduate mentors and the former occupants of his office (Chicago sociologists Ernest W. Burgess, Everett C. Hughes, Morris Janowitz, and William Julius Wilson) peer down on a garrison of filing cabinets standing shoulder to shoulder opposite the neatly indexed bookshelves that furnish Abbott's scholarly life.

It was in the course of losing arguments that Abbott began to see certain similarities between social scientists who would otherwise consider themselves to be at opposite ends of an issue. He returns to this phenomenon repeatedly in Chaos. "[I]f we take any group of sociologists," he writes, "and lock them in a room, they will argue and at once differentiate themselves into positivists [those who take an empirical approach to social life] and interpretivists [those who believe that social experience can be interpreted in many ways]. But if we separate those two groups and lock them in separate rooms, those groups will each in turn divide over exactly the same issues." Indeed, he says, keep separating the sociologists into smaller groups, and they'll keep dividing themselves in exactly the same way. This curious type of self-similarity is, Abbott realized, fractal. The idea that many basic debates of social science are organized around such "fractal distinctions" is central to Abbott's argument.

However sociologists classify themselves-positivist or interpretivist, a belief in the "constructed" rather than the "real" nature of social phenomenon, the study of individuals or emergent groups-however they draw the boundaries between themselves, they are actually just playing with these fractal distinctions at various levels and scales.

As fractal distinctions unfold in time, they make what Abbott calls a generational paradigm. A generation of young, upstart graduate students and assistant professors flips a fractal distinction to generate a radical "new" idea that appears to overturn their elders' ideas. Groundbreaking work gets published defining the parameters of the new theory. The grad students and assistant professors gain tenure, and younger academics follow in their footsteps, clarifying and debating the theory's finer points-less groundbreaking, but still important work. Our young upstarts become full professors, and suddenly their radical theory seems passé in light of the changing times and intriguing ideas being put forth in, say, economics or the humanities. A new generation of upstarts then overthrows these new elders, usually by flipping the fractal distinction back again or by changing scale, and the cycle begins once more.

But, Abbott argues, because the discipline of sociology is open to many points of view (he calls it an "interstitial" discipline, unable to rule out anything that lies between the humanities and the hard sciences), someone inevitably points out that the new theory omits certain important matters addressed by the old one. In the end the victors inevitably "take up the burdens" of their old enemies by "ingesting" the core principles of the conquered theory-albeit with a new name and vocabulary. Abbott calls this process "remapping," and it's extremely useful, he says, because "We get to keep our best concepts forever, and yet can retain our belief in perpetual intellectual progress."

Abbott goes on to demonstrate this fractal nature of sociological research, devoting several chapters to mapping certain areas of scholarly inquiry over the past half century: the theories addressing why people in Western society experience so much stress, the spiraling debates over how deviance and social problems should be understood, and the influence of historians on sociology. He compares all of these sociologists' endeavors to the wanderings of a group of tourists who set out to explore a city. They go about their sightseeing systematically, deciding at an intersection to go east or west, north or south, based on the choices that have carried them thus far. Sometimes they'll end up in a blind alley and must reverse themselves to find their way out, and sometimes they'll run into a fellow tourist on the corner of Rational Choice Street and Conflict Avenue, and they won't necessarily understand how the other tourist got there, but they'll try to agree on the merits of the sights at that intersection. When the sociologists return to the tour bus, he says, hopefully they'll bring back a fairly complete picture of the city of social life. Hidden in Abbott's observations, however, is a caveat: get lost in the fractal distinction-or, one might say, keep going around the same city block and claim that you're covering new territory because you're walking on the grass rather than the sidewalk-and you may risk never getting to know entire areas of the city at all. Or, worse, you may surrender those unexplored neighborhoods to the upstarts in another discipline. He argues that this is exactly why the field of economics is so powerful today.

"Economics," he writes, "has pushed its rigorous rational choice approach into substantial areas of political science, sociology, and history. In all of these disciplines, local thieves have been busy making their reputations by bringing the good news from Ghent to Aix, reselling simplified economic ideas to revolutionize their own disciplines back home. What is unusual about this is not that it is occurring; local thievery is common.... What is unusual is rather the happening of this pattern across several disciplines at once."

What makes the fractal model so handy, Abbott says, is that it is not only a descriptive model, but also a prescriptive one for generating new ideas and covering new ground. "One of the ways to generate new ideas is to say, 'Context! We've got to have a context for X!'" he says. "On the other hand, sometimes you can say great new things by saying, 'To hell with the context.' And depending on how you scale your idea or where it is [in the fractal cycle], that can be very much the right thing to do. I suggest we focus our attention on what these scalable ideas are." He calls on his colleagues in the social sciences to "recognize that this is the way we do things and then just roll with it a little more."

In his final two chapters, Abbott pushes his argument beyond the social sciences to academia in general and society at large. He argues that there are self-similar structures at all levels of society. He also uses the fractal analysis to investigate the politicization of academia.

But his main focus is the social sciences. "It is an emotional business, this work of ours," he admits in the concluding chapter of Chaos. "For me, the chief emotion is a sense of wonder at all the possible ways to know social life. That is what it is to be an eclectic, and that is why it has been necessary for me to think up a way of embracing all the best of it.... [Social science] is progressive, but not cumulative. It can and does forget. But its evolutions are wonderful to watch. And the ever-growing complexity that grows out of the endless permutations of its fundamental ideas is a fulfillment in itself." - S.A.S.