1. According to Knight, we should distinguish between risk expectations, whose evidence includes the theoretical value of the probability of the predicted event, and uncertainty expectations, whose evidence includes no information about the predicted event.[1] The advocates of probability as a unifying concept of all expectations claim that uncertainty expectations can be reduced to risk expectations. Also, they claim that since risk expectations are ordinarily measurable, so are all expectations. However, as Georgescu-Roegen argues,

[t]he difficulty of formulating an objective criterion for decisions involving Knightian uncertainty stems precisely from the fact that the corresponding expectations are not ordinarily measurable, probably not even comparable, and have only a distant and debatable connection with past observations. Under such circumstances there seems to be no other recommendation for dealing with Knightian uncertainty than the common advice: “get all the facts and use good judgment.” But what is “good judgment”? The concept seems to resist any attempt at an objective definition that also would be operational ex ante. […] As difficult as it may be to justify this principle from a strictly logical viewpoint, inductively there is much to be said in its support. Indeed, when people have to entrust statesmen, generals, business executives, trustees of all sorts, with making crucial decisions, they ordinarily look for individuals of “proven” good judgment. Together with gathering, presenting, and analyzing in a logical fashion as many facts as possible, to detect and to use good judgment constitute the only means by which we can respond to living without divine knowledge in an uncertain worlds.[2]

Georgescu-Roegen, himself a champion of analytical economics, thus concludes: “To many this may sound very discouraging, but the opposite view, that good judgment is an obsolete concept in an era of panlogistic models is patently delusive.”[3]

2. Although the concept of “good judgment” is indeed vague, it nevertheless points at the substance of judgment in general. The Greek root krinein, to judge, brings us to the bifurcation between critique and crisis. Their common root defines the subject. Kritikos is able to judge in order to resolve crises—to discern, decide, or to cut. Furthermore, crisis is understood in terms of uncertainty, where the indifferent procedures of the calculus of choice cannot be applied. Critique reaches beyond theory. An active involvement is implied here: critique is engaged, or it is nothing.

3. Theory, on the other hand, leads us to the Greek theorein, to look at. Theory has a different root, and a different telos. Theoros is a spectator—a passive participant at best. Theoros and kritikos part company precisely in a crisis, that is, when Knightian uncertainty prevails. The very notion of “critical theory” is thus a contradictio in adjecto, a testimony to the fundamental ambiguity of that strain of thought that chose it as it banner.

The split between critique and theory is consequently not a matter of mere choice. It is rather a matter of epistemology—of the objective limits to knowledge. In other words, critique has no choice but to reach beyond theory in times of crisis. Ironically, this is unwittingly argued by a positivist, while the critics unwittingly aspire to become spectators.

Footnotes

1. Knight, F.H., Risk, Uncertainty, and Profit, London: London School of Economics (reissue), 1948.

2. Georgescu-Roegen, N., Analytical Economics: Issues and Problems, Cambridge, Massachusetts: Harvard University Press, 1967, pp. 274-275.

3. Op. cit., p. 275.