ȃȇȃ Partʺʺ: Politics and Philosophycan be deemed good utterly apart from some consideration of its likely
results.
As Scott Gordon observed in reviewingĻe Limits of Liberty, Buchan-
an was trying to derive “moral principles without the aid of any moral
premise” by carrying over positive analysis of collective decisionmaking
into fundamental political philosophy. Buchanan might reply that he does
have a moral premise, a weak one favoring individualism. Anyway, in
Gordon’s view (ȀȈȆȅ, p.ȄȇȂ), Buchanan’s attempt to get normative con-
clusions from analysis that is entirely (or almost entirely) positive “cannot
succeed.... [T]ry as one will, that troublesome word ‘ought’ cannot be
excised from political philosophy and no degree of sophisticated ‘is’ can
take its place.”
Striving for clarity may justify some harshness. Except in brief and
untypical passages, Buchanan obscures his employment of and shirks his
responsibility for values that he, like anyone,mustbe employing when
he recommends anything, even when what he recommends is process
rather than substance as the criterion for appraising institutions and
policies.
Viktor Vanberg (n.d.) also raises apt questions about a supposedly
purely procedural criterion. Can we appraise a process or a set of rules
solely by the procedure of establishing and changing it, and so on? How
do we avoid infinite regress? Isn’t some substantive criterion needed some-
where? Furthermore, doesn’t an “agreement test” unavoidably muddle
together observations of people’s preferences and of their theories about
how alternative institutions would work? (Vanberg attributes the latter
point to Karen Vaughn. As Vaughn has also said somewhere, or so I
am told, choosing constitutional principles is not as easy as choosing a
toaster.) One implication, I should think, is that economists and social
philosophers have a legitimate role explaining and preaching what social
arrangements they consider best, and why.
“Contractarianism gets its name,” says one commentator (PettitȀȈȇǿ,
pp.ȀȃȆ–Ȁȃȇ), “from the device which it uses to filter out people’s enlight-
ened preferences.” It identifies just social arrangements as those that
answer to people’s preferences—not the sort recorded by ordinary voting,
however, but rather the preferences that people would have if they were
not influenced by narrow self-interest and bias.
But on what grounds would people—people in real life and even
or especially people stripped of distinctive self-interests, as by being
placed behind a Rawlsian “veil of ignorance”—prefer one set of social