…

                                                                        Book Reviews


THE MARKET: ETHICS, KNOWLEDGE, AND POLITICS.
BY J OHN O'NEILL. ROUTLEDGE, 1998.



P        rofessor O'Neill raises a large number of issues in this important book,
         including the value and nature of autonomy, the merits of equality,
         and whether political institutions should be neutral between differing
conceptions of the good. (He thinks they should not be.) Although our author
sympathizes with socialism, he greatly respects the work of F.A. Hayek, and he
is particularly concerned to examine Hayek's epistemological arguments for
the market.
    I propose to confine the present examination of Professor O'Neill's book to
one central topic, likely to be one of interest to readers of the Quarterly Journal.
In these pages, and in our predecessor journal the Review of Austrian Economics,
there has been much concern with the nature of the socialist calculation debate.
Is there one calculation argument or two--one based on calculation and the
other on knowledge? Did Mises and Hayek advance the same argument, with
minor variations, or were their contentions quite different? Professors Her-
bener, Hoppe, and Salerno, on the one hand, and Yeager, on the other, have not
been in full accord on these matters.
     Professor O'Neill has an interesting and original view of the calculation
argument, and, as will be seen, he to a large extent supports the "separation-
ists," though with a very different assessment.
   Like Herbener, Hoppe, and Salerno, our author sees the reduction of
economic values to measurable units as central to Mises's case.

       For Mises any rational decision, beyond the most simple, requires the
       commensurability of different values. There needs to be a single com-
       mon unit which reduces the choice between different options to a
       matter of calculation. Mises assumes an algorithmic conception of prac-
       tical reason. Rational decisionmaking requires the application of mechani-
       cal procedures of calculation to arrive at a determinate answer to any
       question. (p. 115)

   Our author has, as it seems to me, seriously mistaken Mises's views. Mises
does indeed stress the need for reduction to a common unit when making
decisions. But what he is concerned with are decisions that involve the use of
means of production to achieve competing ends. Should I, e.g., use a supply of

THE QUARTERLY JOURNAL OF AUSTRIAN ECONOMICS VOL. 3, NO. 2 (SUMMER 2000): 81­83

                                          81
82    THE QUARTERLY JOURNAL OF AUSTRIAN ECONOMICS VOL. 3, NO. 2 (SUMMER 2000)



steel to produce dentures in Russian style or to service surgeons who require
plates to insert in their patients' skulls?
    Mises does not say that all decisions can be reduced to a common denomi-
nator of money prices. In particular, he does not claim that a person's ends can
be reduced to a common unit of measure. I should have thought it well known
that Mises believes utility is ordinal and not cardinal. So far as ends are
concerned, he denies precisely the thesis that our author attributes to him.
    Our author's misunderstanding leads him seriously to underestimate the
power of Mises's position. Thus, he correctly points out, "Mises recognizes that,
even in a market economy, there exist `noneconomic goods,' those `which are not
the subject of exchange value.' Environmental goods provide the exemplar of
these" (p. 117). In response, Mises says that those who wish noneconomic goods
can, in a system of market prices, know how much these goods cost them.
     About this answer, Professor O'Neill observes,

     Mises is assuming that every choice is implicitly an exercise in economic
     evaluation. In such hard choices, whether or not we like to admit it to
     ourselves, we are implicit accountants, putting a price on unpriced goods.
     The agent in a choice of this kind knows not only the value of everything, but
     also its price. (p. 117)

    No, no, no! Someone who chooses a noneconomic good over a good with a
price has not put a price on the good he chooses. Rather, he prefers the
noneconomic good to the good of a given price. I cannot think that our author
has fully grasped the distinction between ordinal and cardinal utility.
   He is entirely right, though, to stress in Mises's calculation argument the
importance of a reduction of production alternatives to a common set of
money prices. Now begins our author's new perspective on the argument.
Mises's most famous opponent, Oskar Lange, agrees with Mises about the
importance of a common unit of measure.

     Lange shares Mises's positions on commensurability and the nature of practical
     reason. . . . The central thrust of Lange's argument is to show that there is a
     technical solution to the problem of choice possible within a form of socialism
     which has a market in consumption goods but not in production goods. (p. 119)

    In an important sense, then, Mises and Lange were on the same side in the
debate: both accepted what Professor O'Neill terms the "algorithmic" concep-
tion of rationality.
   Opposed to them stands our author's paladin, Otto Neurath. Neurath
denied Mises's fundamental assumption that choice between values requires
BOOK REVIEWS                                                                        83



commensurability. Quite the contrary, economic choices can be made by on-
the-spot practical judgments that do not require reduction to a common unit.
True enough, we must sometimes use general principles and rules-of-thumb
in making economic decisions; but these do not support Mises's all-embracing
demand for calculability.
     In taking this view, our author thinks, Neurath adopted a position similar to
the line of thought taken by Hayek, in his contribution to the calculation debate.

       In Hayek's work the issue of calculation is largely absent. His main
       contribution to the debate is to attempt to show how, given a market,
       one can have rational decisions without a calculation procedure for
       different social options. In making this move Hayek's position is closer
       to Neurath's than it is to Mises's. (p. 120)

   Thus, with only slight exaggeration, we can say that Professor O'Neill sees
Neurath and Hayek as allies against Mises and Lange.
     What is one to make of all this? I think that our author overestimates the
plausibility of Neurath's approach and misreads Hayek. In order to resolve
questions by practical judgment, there must exist some practice in which the
judgment takes place. Sammy Sosa can judge how to hit a baseball, because he is
part of a social practice, the game of baseball, in which such assessments can be
made. But there is no social practice called "coordinating the economy without
market prices." People cannot use tacit knowledge to make decisions if there is no
set of activities in which such decisions take place. Absent such a social practice of
decisionmaking, Neurath has "solved" the calculation problem by stipulation.
     As to Hayek, our author has fallen into error by his disregard for the
context of the argument. Hayek's main contribution to the debate was his
criticism of Lange, who, it will be remembered, acknowledged that Mises was
correct about the need for commensurable units for calculation. It is only
within the framework of this shared assumption that Hayek raises his prob-
lems about gathering information and the use of tacit knowledge. Hayek does
not deny the importance of calculation: he takes it for granted.
     As suggested earlier, many other topics in this book deserve our attention.
Readers should not miss, e.g., the author's claim that John Stuart Mill's argument
for a single measure of value commits a quantifier-shift fallacy (pp. 123 ff.).

                                                                      David Gordon
                                                                          Mises Review