November 25, 2006

Social science and the effects of small changes…

Filed under: General — Mercuda @ 8:53 am

? ? ? ? ? ? ? ? ? ? ? Both movies It’s a Wonderful Life and Run Lola Run deal with the idea that small changes can make a big difference over time, but each approaches the matter in a different way. Taking this into account, what are the implications for explanations in social science?

? ? ? ? ? ? ? ? ? ? ? To answer this, we must first review the parts of each movie that deal with small changes. Only then will we see that in It’s a Wonderful Life we can predict, almost completely, what the effects of these changes would be, whereas in Run Lola Run we cannot. The former movie plays out almost as predictably (and unrealistically, at least in the social world) as mathematics, while the latter incorporates luck. The former tries to establish cause-effect, while the latter is content with grasping the importance of timing, cumulative effects, and chaos. In short, It’s a Wonderful Life encapsulates the traditional methodology of social science, and Run Lola Run shows us how this methodology is inadequate. The implication for our explanations in social sciences is simply this: We must consider doing away with cause-effect formulations, deterministic covering laws, and mimicry of the hard sciences (physics, biology, chemistry) and devise a new methodology better suited to our complicated reality. Observing how small changes play out in these movies can help us see this.

? ? ? ? ? ? ? ? ? ? ? In It’s a Wonderful Life, George Bailey gets to see what reality would be like if he were never born. The results are predictable: Since George was not there to save his little brother from drowning in the pond, his brother died. And since his brother was later to become a fighter pilot who won the medal of honour for defending a transport ship, this never happened either, and all the men aboard the transport perished. As well, since George was not there to run the building and loans company, Uncle Billy goes insane from trying to go it alone, and nobody builds a home for the Martinis. Lastly, although there are surely more instances, since George was not there to compete with Mr. Potter, Potter now owns the town, which now resembles a cheap imitation of Las Vegas and has of course been renamed Pottersville.

? ? ? ? ? ? ? ? ? ? ? In all of these instances something very simple has happened: George Bailey was removed from the picture. Or, in scientific terms, the independent variable has been omitted, and the variables (the people of Bedford Falls) simply react as we would expect them to in his absence; that is they remain the same.? Never mind that George’s brother might never have gone to the pond without a big brother in the first place, and forget that the transport ship would most likely have been assigned another fighter-escort, or perhaps never left the port at all for reasons anyone can guess. Forget that Uncle Billy might not have thought himself fit enough for the job of managing the building and loan company, or that the Martinis might have found another means to have their home. Forget that a Mr. Potter without real competition might have stuffed himself on gourmet cupcakes and choked or had a heart attack long before founding Pottersville. No, the people of Pottersville react as we would have hypothesized; as dependent variables in a scientific experiment.

? ? ? ? ? ? ? ? ? ? ? The ridiculousness of seeing the world this way is made clearer with a personal example: When I was younger I had to baby-sit my younger brother while my parents were at work. Should we assume that if I had not been born he would have been left home alone, and become a delinquent? Surely not, but in Pottersville and in the social sciences, this is the way the world works.

? ? ? ? ? ? ? ? ? ? ? In Run Lola Run, the main character, Lola, runs to get $100, 000 from her father so that she can deliver it to her boyfriend, Manni, who has become embroiled in a mob transaction gone wrong. But this storyline is played out three times, each with its unique variations and outcomes depending on the small changes that happen early in her run. Specifically, every time she left her apartment she had to pass a man and his dog while going down the stairs. She either ran right past them, was tripped by the man, or jumped around them. Each of these possibilities changed where Lola was on her run and when, which affected every single interaction she had with people and things after that. For example, in the first sequence of events Lola accidentally bumped into a woman who was pushing her baby in a stroller. The woman cursed Lola, and we are taken into a freeze-frame preview of what happens to the woman. Her hostile reaction somehow cascaded into a situation where her child was taken away by social workers. In the following sequences, depending on the woman’s reaction and whether Lola bumped the woman and how hard she bumped her, the woman either joined a church group or won the lottery. None of the outcomes could have been predicted. Other freeze-frame previews involving different characters play out much the same way; that is, unpredictably.

? ? ? ? ? ? ? ? ? ? ? In Run Lola Run luck matters and “shit happens”. During the final sequence, a long series of events and interactions between Lola, a man on a bike, a homeless man, and various others, resulted in Manni retrieving his money. Meanwhile, Lola walked into a casino, and by betting on the number “20” twice for no knowable reason, she wins $100,000. A cause-effect analysis might have said “Due to Manni seeing the homeless man on a bike, he retrieved his money”, without giving credit to every small apparently insignificant event that preceded the retrieval. And of course, there is no way to explain the dumb luck involved in winning $100,000 in the span of 10 minutes or less. As well, in the first sequence, when the small differences cumulate into Lola and Manni robbing a store, a nervous police officer accidentally shoots Lola dead. Small changes early on, and especially those that affect timing in Run Lola Run, make a big difference later, but not necessarily a difference we can predict.

? ? ? ? ? ? ? ? ? ? ? The realities and possibilities in Run Lola Run come much closer to helping us understand the real world of human action. The implication for our explanations in social science is that we must stop trying to explain using the methodology of “hard science”, cause-effect, and covering laws. It is futile to cram the round peg of human action into the isosceles tetrahedron-shaped hole of scientific methodology and expect a fit.

? ? ? ? ? ? ? ? ? ? ? Social science needs its own philosophy of science; one that examines processes. This means paying closer attention to initial conditions, cumulative causation and feedback effects. Or perhaps we need mechanisms – plausible, frequently observed ways in which things happen (Elster, 1). We should be satisfied with establishing probabilities, making quality descriptions, and focusing on how things came to be. After all, it is becoming too obviously dishonest to omit variables that do not fit, or to pile theories on top of theories to explain away those variables.

? ? ? ? ? ? ? ? ? ? ? By comparing the treatment of how small changes affect outcomes in the movies It’s a Wonderful Life and Run Lola Run, we can see why we should consider doing away with cause-effect formulations and covering laws and devise a new methodology better suited to the study of complex human behaviour.