Incommensurability, explanation, world views

tuomas.kuronen@tkk.fi — Wed, 04 Mar 2009 00:00:00 +0100

Consider LSE's slogan: rerum cognoscere causas ("to know the causes of things", ripped from Virgil, I suppose). It makes a direct claim favouring causal explanation. Another matter, however, more rooted in the Anglo-American tradition is the tendency to embrace methodological individualism (individuals matter).

Taken these, it should not come as a surprise that a joint session of Arabic and US speakers covering contemporary issues in the Middle East became a clash of world views above anything else. Everyone agreed with the existence of roughly the same set of problems.

What is interesting, is that given the newly elected US administration, the approach seems to be: ‘micro-agents and macro-measures'. Both the measures of the ‘international community', as well as the ‘group controlling' seem somehow to have access to the solving of the problems. The field of application is huge, and as the focus of interest shifts from the unpleasant to the more generally accepted (thus more interesting), some fates become irrelevant to the nexus of policy practice. Weirdly enough, some people seem to find the flow of medical equipment relevant.

As I am in the middle of digesting about a half of Thomas Kuhn's The Structure... it appeared to me that perhaps incommensurability is fundamentally conceptual by nature, rather than simply "talking past each other" as sometimes proposed? Taking this a bit Wittgensteinian, are practitioners of different schools of thought unable to even see the problem (or ‘puzzle', as Kuhn would have it), not to speak of speaking of it?

If we believe the professionals (no matter what discipline, in this case a foreign policy professional from Washington), they have a whole toolbox full of ready-to-use means to solve all kinds of problems in the world. Belief in agency for ideological (and practical) reasons is handy. By being able to point out causes of things, one can justify pretty much any policy decision. Some people call this ‘pointing fingers'.<meta http-equiv="Content-Type" content="text/html; charset=utf-8"><meta name="ProgId" content="Word.Document"><meta name="Generator" content="Microsoft Word 11"><meta name="Originator" content="Microsoft Word 11"><link rel="File-List" href="file:///C:%5CDOCUME%7E1%5Ctkuronen%5CLOCALS%7E1%5CTemp%5Cmsohtml1%5C01%5Cclip_filelist.xml"><style> </style> <p>Consider LSE's slogan: <i>rerum conoscere causas</i> ("to know the causes of things", ripped from Virgil, I suppose). It makes a direct claim favouring causal explanation. Another matter, however, more rooted in the Anglo-American tradition is the tendency to embrace methodological individualism (individuals matter).</p> <p>Taken these, it should not come as a surprise that a joint session of Arabic and US speakers covering contemporary issues in the Middle East became a clash of world views above anything else. Everyone agreed with the existence of roughly the same set of problems.</p> <p>What is interesting, is that given the newly elected US administration, the approach seems to be: ‘micro-agents and macro-measures'. Both the measures of the ‘international community', as well as the ‘group controlling' seem somehow to have access to the solving of the problems. The field of application is huge, and as the focus of interest shifts from the unpleasant to the more generally accepted (thus more interesting), som [...]

Gigerenzer on heuristics

tuomas.kuronen@tkk.fi — Wed, 22 Oct 2008 00:00:00 +0100

The public lecture super-week at LSE was launched by Gerd Gigerenzer, a German psychologist whose main work concentrates on bounded rationality and heuristic decision-making. His main thesis was the re-evaluation of the unconscious processes taking place in human heads that are commonly seen as, at best, obscuring so called rational ones. The people that think so, let us call them ‘limitationists’, seem to be mainly the same individuals that received the bulk of criticism during the lecture held last Monday. The bottom line is that there can be something happening beyond the realms of language.

Known from Gigerenzer’s and his colleagues’ previous works, information ignoring ‘fast and frugal heuristics’ seem in many cases to outsmart the now suspect ideal of homo economicus. By the way, Markowitz himself didn’t use portfolio theory in allocating his own investments, but used a 1/N heuristic instead. What was new in this presentation, however, was the predictive power of heuristics in some contexts. The audience was shown how – at times – heuristic decision-making gives more accurate predictions. This is rather counter-intuitive.

Consequently, it is suggested that the classical accuracy-frugality –divide doesn’t necessarily hold, which has some serious implications. It means that more is actually less. Not only is arduous computing time-consuming, but it also tells us less of what is going to happen. As Gigerenzer himself put it, “too much information can hurt.” In this case, the poverty of sophisticated mathematical models can be put quite plainly: “fitting easy, prediction hard”.

In our times, we witness whole industries being built based on false assumptions of computability and episteme itself. Gigerenzer himself sees gut-feeling-based decision-making being especially prevalent in two cohorts; small children and very highly ranked directors. These two groups of people tend to have the self-esteem for it. Middle-managers, on the other hand, have the need to resort to increase in confidence derived from complex mathematical decision support models. What makes Gigerenzer’s claims strong is the fact that he’s not in the business of searching theoretical elegance a priori, but working with empirical evidence. The search of predictive power in decision-making heuristics is strong enough without assumptions of human Bayesianism.

GloStra Blog Entries tagged 'decision making'

Incommensurability, explanation, world views

Gigerenzer on heuristics