Est. May 2008

23 November, 2012

Humans Are Basically Good?

(click to enlarge)
Well, that’s what the good folks over at Scientific American would have us believe: they’re reporting on some recent studies done by ‘a diverse group of researchers from Harvard and Yale—a developmental psychologist with a background in evolutionary game theory, a moral philosopher-turned-psychologist, and a biologist-cum-mathematician’ to test ‘whether our automatic impulse—our first instinct—is to act selfishly or cooperatively’.

See the problems?


Defining ‘good’ and ‘bad’ (setting aside for the moment the biblical explanations of both behaviors) is a complicated business.  In our relativist modern society, what’s ‘good’ for some may not be ‘good’ for others.  Additionally, ‘goodness’ and ‘badness’ can’t be determined based on one set of behaviors (selfishness versus cooperativeness) – labeling even an individual as ‘good’ or ‘bad’ has to take into consideration whether ‘good’ or ‘bad’ behavior predominates in the behavior patterns of the subjects of the test.  It’s a real stretch of the data to take a single aspect of ‘goodness’ and ‘badness’ (selfishness versus cooperativeness) and try to make it fit the population en masse.

The third experiment should have clued the researchers in to the idea that their hypothesis was too narrow to make such a broad conclusion.  That third experiment – interviews with people – showed that ‘cooperation is the intuitive response only for those who routinely engage in interactions where this behavior is rewarded—that human “goodness” may result from the acquisition of a regularly rewarded trait’ (emphasis mine).  In other words, people are more likely to cooperate if they have been and expect to be rewarded for it.

Which leads to this complaint: the two experiments used – the ‘‘prisoner’s dilemma’ and the ‘public goods game’ – offer nothing but the idea of a reward, since the subjects used ‘money’ provided them by the researchers.  With no ‘skin in the game’, so to speak, whether the subjects were selfish or cooperative had zero meaning: regardless which way they went, they stood to gain or lose absolutely nothing.  And if they had even the barest inkling of what the tests were looking to find, they might have been self-persuaded to give the researchers what they were looking for, rather than acting as they would ‘normally’.

Having once done a great deal of scientific experimentation, I can offer a few suggestions for determining whether human beings are generally ‘good’ or ‘bad’.  First, define your terms: what do you mean by ‘good’ and ‘bad’?  Secondly, don’t narrow your data-collection to one or even two types of behavior: broaden them as widely as possible to take into consideration as many variables as is manageable.  Third, get out of the lab and investigate human beings ‘in the wild’: you may discover that people behave in diametrically-opposite manners to the way they do within the controlled confines of a lab.

No comments: