The huge ongoing interest in behavioral economics manifests dissatisfaction with the rational choice paradigm. At the same time, behavioral economics does not as yet constitute a new paradigm in itself. Social psychology and cognitive psychology have documented to date hundreds of biases and dispositions, some of which behavioral economics pursued. There haven’t yet emerged any clear “rules of the game” for behavioral economics that bound the framework (the analogue of “an individual is defined by her immutable preferences [over possibly very complex and stochastic objects], which she maximizes under the prevailing constraints” in rational choice), as well as the legitimate questions to ask within the framework and the ways to tackle them (the analogue of “what behavior and comparative statics/dynamics are ruled out under non-perverse preferences and a solution concept compatible with some version of common belief in rationality” in rational choice).
It is therefore interesting to compare the current state of affairs with a possibly similar episode in the history of science, namely Eighteenth-Century Chemistry. Chemists were envious with the immense success of Newtonian mechanics and optics, and were seeking to do away with alchemy, whose language tangled predictions about the world with allegorical and mystical prescriptions. Most chemists did believe in atoms, but they were reluctant to phrase any theories about them which would be no less hypothetical than alchemy itself. Textbooks of the period were still paying lip service to the four Aristotelian elements of Earth, Water, Air and Fire, but it was obvious these were not helpful for predicting the behavior of e.g. acids and bases.
So chemists began to construct Affinity Tables, which sorted known substances into categories, and specified which of these categories “attract” one another and to what extent. Soon enough they realized that these tables grow exponentially in size. The conceptual breakthrough came with Lavoisier, who suggested that one should first and foremost study the relation between elements, i.e. substances which cannot be decomposed into simpler substances.
Lavoisier’s list of elements contained, of course, the Oxygen which he discovered, and it dispensed with Phlogiston. However, it contained also Caloric which was dispensed with not-too-long afterwards, as well as other items such as Light, which we do not count today as chemical elements. Still, the concept of an element became the bedrock of chemistry, and as such enabled Mendeleev’s further breakthrough and, in the sequel, the modern atomic model. At the same time, Lavoisier himself acknowledged that his advances were made against – and in contrast with – the background of the systematic classification efforts of all those chemists who constructed and refined the affinity tables.
Are there any “elementary” motives of human behavior? Whatever the answer, it could be useful to keep the perspective that today’s behavioral economics, with its incremental efforts to classify dispositions and their implications, may eventually serve as a jump-board for a future conceptual breakthrough and a better paradigm; but that the language, approach and concepts of such a new paradigm may very well turn to be decisively different than those employed by behavioral economics today.
2 comments
June 16, 2011 at 1:54 am
Burkhard C. Schipper
Phlogiston did explain oxidation pretty well. Why was it discarded? Because “real” elements were discovered empirically. One may argue that some tendencies in behavioral economics such as neuroeconomics or endocrinological economics follow the same trend, and try to discover biological elements that effect economic behavior rather than starting explanations from the level of more or less unobservable preferences. It is yet to be seen whether this could be as useful to economics as modern chemistry was to chemistry. Personally I doubt it despite being attracted by the endeavour purely out of scientific curiosity. But in some sense, one may argue that the reductionist spirit of modern science may root much deeper in behavioral economics than in standard economics, only that it takes a more empirical form given today’s easier availability of data and increased computational power. What may be missing are workable data-organizing theories.
June 16, 2011 at 11:07 am
Aviad Heifetz
The sentence ” ‘real’ elements were discovered empirically” seems to be mixing two different things. You have to have the *concept* of an element (which is very different from the building block of the affinity tables, which was “an equivalence class of substances” defined by “having similar affinities with representatives of other equivalence classes”) in order to even conceive of the idea of checking *empirically* whether a particular substance is an element.
Lavoisier conceived of the concept of an element as fundamental, and then showed empirical evidence that what he called oxygen is an element.
My intention was to point out that we might not yet be conceiving what in the future we will be considering as the fundamental notions to conceptualize human behavior. When behavioral economics classifies and analyzes “dispositions” (e.g.”fairness concern”), the very notion of “disposition” may be the analogue of 18th-century “equivalence class of substances”, which will be replaced in the future by an analogue of “element”.
Experimental findings (in particular in neuroeconomics) may inspire us to come up with new concepts, but new concepts cannot be discovered “empirically”. And you have to have concepts in order to even phrase “workable data-organizing theories.”