In general, we we would expect that factors will be perceived as causal to the degree that their magnitudes resemble the magnitude of the effects they are adduced to explain. In the development of causal schemata, both the notion that large causes can produce large effects and the notion that small causes can produce small effects probably precede the development of the notion that large causes can produce small effects. The notion that small causes can produce large effects probably develops very late and never attains very great stability.This dry statement comes near the end of a relatively well known paper by the psychologists Richard Nisbett and Timothy Wilson, written in 1977 and published in the Psychological Review (Vol 84, No. 3) , called 'Telling More than We Can Know' (An allusion to and inversion of Michael Polanyi's famous aphorism about tacit knowledge that 'we can know more than we can tell').
The paper was written as a summary of, and attempt at explaining the results of, a large number of experiments involving subjects' verbal reports of explanations for their behaviours in a variety of carefully controlled situations. (Or 'formalised anecdotes', as I would call them, if I am to be consistent with myself last week when I referred to ethnographies as anecdotes.) The counter-intuitive discovery of many of these experiments seemed to be that people often don't know why they do the things that they do: we often act for one reason, but think we act that way for another.
Nisbett and Wilson's explanation - a meta-explanation, as it were - is that we possess a limited range of cognitive archetypes for causal schemas, intuitive patterns we apply to events in order to explain their occurance. Following from this, causal explanations which fit our causal schema more closely appear more intuitively plausible than those which don't, even if the plausible causal explanations can't really explain (or predict) anything, and the implausible causal explanations can.
The entire article is well worth reading. Nisbett and Wilson describe a number of qualities of intuitive causal schema, each of which have profound implications that, even thirty years after publication, still seem underexplored. As I said at the start of this entry, however, I'm going to concentrate on the magnitude-matching aspect of the causal schema, as described in the quoted passage. Nisbett and Wilson seemed to believe what they were writing isn't completely inconsequential to everyday life, culture, history, and common understanding, as the passage immediately following the one quoted suggests:
It is likely that conspiracy theories often feed on the discrepancy between officially provided causal explanations and the large effect they are invoked to explain. It is outrageous that a single, pathetic, weak figure like Lee Harvey Oswald should alter world history. When confronted with large effects, it is to comparably large causes that we turn for explanations. [When some insomniacs were asked] why they slept so little, both on particular occasions and in general, they were inclined to explain their insomnia in terms of the stress of their current life situation or even in terms of neurosis of chronic anxiety. Smaller causes, such as an overheated room, a tendency to work or exercise or smoke just before going to bed, or a tendency to keep irregular hours, were overlooked.
My guess is that this tendency to link magnitude of effect to magnitude of cause causes us to miss, or greatly under-value, a large number of seemingly inconsequential ,is-it-plugged-in-and-switched-on, explanations to some important social issues, and through this to miss the obvious. As a now relatively well-known and widely-publicised example, consider the theory that there is a link between abortion rates and crime. Another, slightly less well-known candidate for the unduly-and-implausibly simple cause-effect schema is provided by the (somewhate eccentric) experimental psychologist Seth Roberts in the form of the Shangri-la Diet. "Of course it won't work: it's too simple," seems to be the monotone chorus of response to Roberts' suggestion. "It doesn't add up." (As an example of this response, see the end of this generally excerable video about the diet, by Simon Jenkins, a nutritional scientist: a battle seems to start to rage between the 'trained' part of him which knows it should be tested and evaluated;and the latent, 'intuitive' part of him which simply can't find the idea credible.)
Within physical systems, perhaps it's easier to see how small causes can lead to big effects: within systems engineering, control theory (formally known by the much more intriguing name of cybernetics) teaches one to think about feedback loops. Positive feedback means small causes can have big effects; negative feedback means the converse: it all depends on how the causal structure in between is 'wired'.
Unfortunately, without being taught to think unnaturally, our minds tend not to consider the amplifiers and inhibitors which may lie between A and B. By default, our minds just tie A and B together with Causal String: if I feel a big tug at my end of the Causal String, I'm going to guess there's something big at the other end, even though it may just be an Imp-with-a-Winch (or an operational amplifier).
Additional Viewing:
No comments:
Post a Comment