Sei sulla pagina 1di 21

Learning for Democracy, Vol. 1, No.

2, 2005 7

Is an Open Society a Just Society?:


Popper and Rawls1

ALAIN BOYER
Université de Paris IV (Sorbonne), France

A caste system, for example, tends to divide society into separate


biological populations, while an open society encourages the widest
genetic diversity.
John Rawls2
A Theory of Justice, § 17, p. 107

[S]ocial change can be arrested only by a rigid caste system.


Karl Popper
The Open Society and Its Enemies, Vol. I, chapter 6, § VI

ABSTRACT In this paper the author addresses the problem: Are Karl
Popper’s and John Rawls’s liberal philosophies contradictory or
complementary? He argues that Rawlsian methodology is rather
Popperian, and in some respects Popper’s political philosophy is close to that
of Rawls. Unlike Rawls, Popper’s incursion into moral philosophy is not
equivalent to a systematic theory, but both philosophers were concerned with
equalitarianism, democracy, justice, and personalism. They share a concern
for the idea of a plan of life. Rawls is somewhat naïve on the innocuousness
of the institutions of a just society; but he is concerned with the promotion of
‘democracy of owners’, not a Welfare Leviathan. Popper was more sensitive
to Hayekian arguments against ‘constructivism’. In Rawls’s theory, liberty
is ‘lexically’ prior to social justice: this is a strong affirmation of liberalism.
The young Popper was a liberal social democrat. Still, even the later Popper
was not a libertarian; his so-called ‘negative utilitarianism’ has something to
do with the intuitive appeal of Rawls’s ‘difference principle’.

INTRODUCTION
There is only one piece of textual evidence that Karl Popper read A Theory of
Justice3 (TJ), and only one that John Rawls read The Open Society (OS).4 My
problem is: Are these liberal political philosophies contradictory or
complementary? My answer is that they are more or less complementary. This
8 Is an Open Society a Just Society?

has some consequences for the status of liberalism vis à vis its traditional
criticisms, from communitarianism to republicanism. I would argue that
Rawlsian methodology is rather Popperian, and also that in some respects
Popper’s moral and political philosophy can be interpreted as more or less
close to that of Rawls, with some restrictions. But it is necessarily less
Rawlsian than Rawlsians could wish, because it is more general.
First, the process of reflexive equilibrium, that Rawls borrowed from
Nelson Goodman (1955),5 can be shown to have been anticipated by Popper
(1979[1932]). Rawls methodology is neither foundationalist nor akin to any
‘analytic meta-philosophy’, but is a substantial, fallibilist and even
falsificationist one, the empirical basis being replaced by our intuitive,
normative, basic judgements (‘If slavery is not wrong, nothing is wrong’ –
Lincoln quoted by Rawls, 2001, § 10, p. 29). Rawls’s criticism of utilitarianism
is also ‘methodologically individualist’ and ‘personalist’.
Second, Popper’s incursion into moral philosophy is dominated by the
problem of freedom versus totalitarianism, and by his criticism of the
‘philosophies of history’, problems not touched on by Rawls who regarded
them, perhaps wrongly, as already behind us. (Rawlsians should read
Popper.) But they were both greatly concerned with the idea of
equalitarianism, justice, and personalism. They shared a concern for the idea
of a ‘plan of life’, with perhaps an important difference.
Third, Popper’s thesis of ‘the ambivalence of institutions’ is remarkable.
Wasn’t Rawls a bit naïve about the innocuousness of the coercive institutions of a
‘just society’ (in his sense of the term)? Perhaps not so much, as he was quite
concerned about the necessity of a ‘sense of justice’ (see what Popper would call
‘traditions’), and by the promotion not of a Welfare Leviathan but of a ‘democracy
of owners’. Clearly, Popper was more sensitive than Rawls to Hayekian arguments
against ‘constructivism’ and state bureaucracy. Still, note that in Rawls’s theory
equal liberty is ‘lexically’ prior to social justice, and that this is a strong affirmation
of liberalism. The young Popper was a liberal social democrat of sorts. But even
the later Popper was not a libertarian, and his so-called ‘negative utilitarianism’ has
something to do with the intuitive appeal of the ‘difference principle’. As a
regulative ideal, that principle is not anti-Popperian in spirit.

JUSTIFICATION
This question seems to oppose Popper and Rawls: Is not the former famous
(or infamous) for having proposed that we should eliminate the obsession with
justification? Guess as imaginatively as you can, and criticize your guesses
through the analysis of their deductive consequences! Accept provisionally, as
a good candidate to truth, what resists your best criticism (see, also, Miller,
1994). Do not try to make your guesses inaccessible to criticism by searching
for a so-called foundation of them. Justificationism is a dream, and a deceitful
one, a kind of religious quest for the absolute, for security and infallibility. Let
it be so. Anyway, Rawls seems much more concerned with justification (TJ, §
Learning for Democracy, Vol. 1, No. 2, 2005 9

87, ‘Concluding remarks on justification’, and also TJ, § 4). But if we read TJ
carefully, we find that his conception is quite fallibilist: his reference being not
Popper, but Quine. And it would be difficult to classify Quine as a
‘foundationalist’ and an ‘infallibilist’. His epistemology is more classically
empiricist, behaviouristic, naturalist and less realistic than Popper’s, for sure,
but he was a staunch revisionist, even in logic (in principle). He even accepted
what he called ‘Popper’s negative methodology’, in relation to Hempel’s
paradox (Quine, 1990, chapter 1, § 5, pp. 12-13).
Let me now quote a few sentences from TJ concerning justification: ‘The
question of justification is settled by working out a problem of deliberation’,
‘Moral theory is Socratic’, ‘There is a definite if limited class of facts against
which conjectured principles can be checked, namely, our considered
judgments in reflective equilibrium’, ‘Justification is argument addressed to
those who disagree with us, or to ourselves when we are of two minds. It
presumes a clash of views’. The similarities with Popper appear to me to be
more important than one would have thought. Do not quarrel about words!
Now, let me remind you how high a value Popper attached to impartiality.
(The more or less Rawlsian philosopher, Brian Barry, had chosen a quotation
from OS as the motto of his book Justice as Impartiality, 1995). And the
construction of a procedure of justification of the principles of justice, according
to Rawls, aims to produce impartiality as a result of the conditions of an ideal
moral discussion, without presuming that the individuals in the original position,
who have to prefer a theory of justice to others, are themselves of a perfect
impartial mind: objectivity, so to speak, is not a personal, but an institutional
affair, as it was according to Popper (‘the institutional theory of objectivity and of
progress’ – OS, II, chapter 23; 1961[1957], § 32).
Rawls’s argument is as follows: As opposed to average utilitarianism,
classical utilitarianism is not a decent principle of (individual) rationality, so
why on earth should I care about the global welfare of society? In the very
peculiar situation of uncertainty that is represented by the ‘original position’,
beneath the veil of ignorance (an impartiality producing device), average
utilitarianism has to be contested by sophisticated arguments, in favour of a
not very popular principle of rationality, the maximim principle, so we should
prefer Rawls’s principles to average utilitarianism. But, classical utilitarianism
is in a much worse position: it is a non-starter. Being ‘pious’ with the
philosophical tradition, Rawls thought one cannot attribute to Bentham or
Sidgwick the crazy idea that it would be rational for a normal person, with a
‘limited altruism’, to choose a society that maximizes the net balance of global
welfare. That shows, as Rawls argued, that the classical utilitarian theorists’
way of justifying their principle, as opposed to John Harsanyi’s, could not
possibly have been some sort of original position, but rather the Humean and
Smithian idea of an altruist impartial spectator, in love or in sympathy with
the whole (‘conflation of persons into one’– TJ, § 30). Like Popper, Rawls had
nothing against love per se, but he rightly thought – again, like Popper – that
10 Is an Open Society a Just Society?

universal love is not to be demanded of individuals, and also that it cannot


replace critical deliberation, and compromises (OS, II, chapter 24, § III).
Popper’s remarks on love and other passions, in chapter 24 of OS, II, have not
been sufficiently acknowledged.
Incidentally, let me remark that Harsanyi was an opponent both of Rawls
and of Popper, and his achievements in expected utility theory and in welfare
theory are not insignificant, if only because he achieves a systematic view
(including morals in decision theory – a thesis that Rawls himself eventually
abandoned). Anyway, John Watkins once argued that Harsanyi was a ‘friend
of Modus Ponens’, while Popperians (except in pure maths) are rather ‘Modus
Tollens’s friends’. I would propose that Rawls, compared to Harsanyi, was a
‘Modus Tollens’s friend’ of sorts: he definitely rejected the idea that we have
access to a realm of axiomatic self-evident moral truths from which one can
deduce all the true consequences of the principles. The human situation is
simply not like that. For example, we have to introduce the (Humean) idea of
the ‘circumstances of justice’ in our reasoning in the ‘original position’
(something semi a priori, belonging perhaps to the Popperian ‘situational
analysis’), as well as some anthropological or economical theory that has stood
up to tests: but these theories are conjectural, even if corroborated (the
importance, for instance, of incentives in economic life), and could be
replaced. Our conception of justice can therefore change. And we have to test
our principles against our considered judgements.
Popper himself made the analogy, pointing out the difference:
The rational and imaginative analysis of the consequences of a moral
theory has a certain analogy in scientific method. For in science, too, we
do not accept an abstract theory because it is convincing in itself; we
rather decide to accept or reject it after we have investigated those
concrete and practical consequences which can be more directly tested
by experiment. But there is a fundamental difference. In the case of a
scientific theory, our decision depends upon the results of experiments.
If these confirm the theory, we may accept it until we find a better one.
If they contradict the theory, we reject it. But in the case of a moral
theory, we can only confront its consequences with our conscience. And
while the verdict of experiments does not depend upon ourselves, the
verdict of our conscience does.
(Ibid., p. 233)

TRUTH
The first sentence of the first section of A Theory of Justice reads:
Justice is the first virtue of social institutions, as truth is of systems of
thought. A theory however elegant and economical must be rejected or
revised if it is untrue; likewise laws and institutions no matter how efficient
and well-arranged must be reformed or abolished if they are unjust.
Learning for Democracy, Vol. 1, No. 2, 2005 11

Rawls’s target is utilitarianism, as is shown unambiguously by the


following sentences, which are about the moral impossibility of the
‘sacrifices’ imposed on a few, in the name of the spurious fact that they could
be ‘outweighed by the larger sum of advantages enjoyed by many’. More on
this later. Let me remark that this objection to utilitarianism is similar to the
one made by Popper with regard to Platonism (collectivism), as Jeremy
Shearmur aptly remarked, following a suggestion by Popper (Shearmur,
1996). Anyway, it seems clear that such a conception of truth does not depart
from Popper’s, even if it departs a bit from Quine’s, who was more of a
pragmatist, and that it is quite anti-relativist. Truth, like justice, has a
‘lexical’ priority over all other epistemological qualities: no amount of
simplicity or mathematical elegance could compensate for falsehood.6 In
politics, then, justice should be our main aim, as truth is in knowledge
(according to Popperians). Popper himself suggested the same analogy
between truth and ‘rightness’ in an (insufficiently) famous addendum to OS
(II, addendum 1, ‘Facts, standards, and truth’).

INTUITION
During the 1960s, intuition acquired a rather bad reputation in ethics. In
1967, Philippa Foot wrote the following: ‘He would be a brave man who would
assert, like Ross in the middle thirties, that “every ethical system admits
intuition at some point”’ (Foot, 1967, p. 2). In 1971, Rawls claimed to be just
such a brave man; after discussing Ross, he wrote: ‘Any ethical view is bound
to rely on intuition to some degree at many points’ (TJ, § 2, p. 7). Intuition is
necessary for testing moral principles, but also for building the ‘intuitive
notion’ of an ‘objective’ original position. Note that in arguing that point,
Rawls quoted Henri Poincaré (in French): ‘Il nous faut une faculté qui nous
fasse voir le but de loin, et, cette faculté, c’est l’intuition’ (Poincaré,
1970[1905], p. 36).7
In Objective Knowledge, in a chapter written in 1967, Popper describes
quite nicely Brouwer‘s philosophy of intuitionism, and he ends his discussion
with a critical view of intuition in general:
Everything is welcome as a source of inspiration, including ‘intuition’,
especially if it suggests new problems to us. But nothing is secure […]
‘Intuition’ […] is largely the product of our cultural development, and of
our efforts in discursive thinking.
There is a give-and-take between construction, criticism, ‘intuition’, and
even tradition …
(Popper, 1979[1972], pp. 134-5, 137)

This last sentence must strike a serious reader of TJ as quasi Rawlsian


(reflexive equilibrium). Rawls did not claim that intuition is per se infallible
and that it is sufficient for erecting theories: each ‘well considered judgement’
12 Is an Open Society a Just Society?

is presented as a conjecture about our (educated) ‘sense of justice’, and the


reader has to check by himself, so to speak, whether he agrees, after critical
deliberation, with those conjectures proposed by the author. One has to
meditate with him, in a Cartesian way. Rawls’s main aim was to produce a
better (approximate) theory of social justice than utilitarianism(s), than
perfectionism (a word already used by Popper in OS, I, chapter 9), and also
better than libertarianism (‘the system of natural liberty’), in exhibiting the
diverse ways in which those conceptions are contradicted by at least one of our
moral well-considered ‘fixed points’. In the other direction, the argument he
gave against intuitionism (in ethics) is not that it would be a ‘teleological’
theory (Ross’s theory is not), but that it leads us astray when we have to
choose, or when we prefer one institution rather than another: intuitionism
denies that one can find a ‘priority rule’ in the domain of distributive justice.
Rawls’s challenge, then, was to build a non-utilitarian theory that is able to
give us reasons to prefer, after public deliberation, the better reform we have
been able, fallibly, to imagine. The theory of justice would claim to be, in
Popperian terms, at least one part of a methodology of ‘piecemeal social
engineering’, which can only be used as a trial and error method. (Rawls,
unfortunately, preferred in general to stay at the level of the ‘ideal theory’.) We
are confronted with social problems and we need regulative ideas to guide our
way (presumably, a ‘third way’ – see Tom Settle, 1976). In the methodology of
empirical sciences, it is necessary to have a real sense of the ‘intuitive
judgements’ of the best scientists, and a provisional consensus on statements
describing empirical intuitions (in a Kantian sense) is necessary to stop the
infinite regress of ‘Fries’s trilemma’ (Popper, 1959, chapter 5). But nothing
stops even a seemingly unassailable empirical statement from being rationally
examined in its contention to be absolutely true, if one has a good critical
argument against it. The same is true in morals. Of course, we are not ready
to abandon some very basic judgements: Auschwitz is bad, as surely as the
Cartesian Cogito is true (I cannot doubt that I exist): but even the Cogito can
be rationally discussed, not really about its truth but about its status, its
consequences, etc. And Rawls showed that things are more difficult with other
judgements, for example about (excessive) wealth. Why are we against it? Is
it a manifestation of envy? This deserves critical scrutiny.

THE QUESTION OF THE ‘LINGUISTIC TURN’


There is no question that Popper was an opponent of the so-called ‘linguistic
philosophy’, once founded on the ‘linguistic turn’ (Wittgenstein, Schlick). I
would dare to suggest that Popper’s polemic against Wittgenstein’s followers,
little different in spirit from that of Russell, was much too polemical (as were
his views on Hegel), and that partly explains the surprisingly low level of
consideration for Popper’s achievements in academic philosophy. Rawls was
of a different persuasion. But he was also reacting against ‘linguistic
philosophy’ in ethics and political philosophy, and in this I can only remark
Learning for Democracy, Vol. 1, No. 2, 2005 13

that he eventually won. Most of the current discussions in moral and political
philosophy are now about questions of things (‘substantial’), not of words.
Rawls’s conception of definition is quite nominalistic, like that of Popper.
There is nothing essentialist about Rawls’s questioning. Let me quote a
significant assertion:
The analysis of moral concepts and the a priori, however traditionally
understood, is too slender a basis. […] Note […] the extraordinary
deepening of our understanding of the meaning and justification of
statements in logic and mathematics made possible by developments
since Frege and Cantor. A knowledge of the fundamental structures of
logic and set theory has transformed the philosophy of these subjects in
a way that conceptual analysis and linguistic investigations never could.
One has only to observe the effect of the division of theories into those
which are decidable and complete, undecidable yet complete, and neither
complete nor decidable.
(TJ, § 9, pp. 51-2]
This is Popperian in spirit, and I would imagine that a Popperian such as
David Miller would be rather happy with it. Rawls then applied that analogy
to ‘moral conceptions’. For those who continue to regard Rawls’s theory as
‘purely procedural’ and as a formal ‘conceptual analysis’, note this: ‘I wish then
to stress the central place of the study of our substantive moral conceptions’
(TJ, § 9, p. 45). (Incidentally, I would argue that Apel’s and Habermas’s
theories are more ‘formal’ than those of Rawls and Popper.)
Rawls could nevertheless appear to be more ‘positivist’ than Popper was
(that is, not much). For the latter, the demarcation criterion was not at all a
meaning criterion: if a falsifiable theory has non-falsifiable consequences,
these cannot be meaningless (on the other hand, verifiable statements have
only verifiable consequences, but verificationism is patently false); in which
case, some metaphysical statements, particularly some ‘all-and-some
statements’, are meaningful metaphysical statements, neither analytical nor
empirical (John Watkins), like the principle of determinism, or its negation.
And the regulative role of metaphysics is quite important for Popper
(‘metaphysical programmes of research’). Rawls seemed to be of a more
positivistic mind, because he famously claimed that his theory was only
‘political, not metaphysical’ (Rawls, 1999, chapter 18, ‘Justice as fairness:
political, not metaphysical’). (But for Popper, too, methodology is in principle
not logically linked with any metaphysical thesis.)
Anyway, some people read only the titles of the papers: in fact, Rawls did
not claim that metaphysics is meaningless or without any interest, and
certainly not that ‘politics’ should be morally neutral(!), but that some of the
most important metaphysical issues, like materialism versus spiritualism,
atheism versus theism, determinism versus indeterminism, are not relevant
for the setting up of a moral consensus in a pluralistic (open) society. The
metaphysical debates are legitimate, but they are not to be confused with those
14 Is an Open Society a Just Society?

about the principles of a well-ordered society, on which we should strive for


‘an overlapping consensus’. The problematic but attractive idea proposed by
Rawls is that an ‘ideal of the person’ (or of the citizen) can be commonly held
between people with very different metaphysical and religious backgrounds
(‘tolerance’, ‘fact of pluralism’). This seems to me to be true, even if it needs
careful examination. In France we speak of ‘laïcité’. The state should be
neutral on religious matters but not on basic moral questions (equality of men
and women, possibility of ‘apostasy’). But, for example, a realistic cum
indeterministic interpretation of quantum mechanics (Popper) has nothing to
do with the principles of morality that everyone should accept and assume that
others will accept and conform to.
Popper correctly argued that traditions (or ‘social norms’), in addition to
laws, are necessary if the social world is to avoid chaos; they constitute a
network of mutual expectations which permit anticipations and action
(Popper, 1972[1963], chapter 4). But a metaphysics of what there is, deeply,
in the world, is independent of that social network. Kant would perhaps not
have accepted this, if we note the extent to which his theories of liberty and of
a moral character are linked with his non-neutral ‘transcendental idealism’
(see Boyer, 2001a). (Incidentally, neither Popper nor Rawls, despite their
common admiration for Kant, found the latter particularly appealing.) Of
course, Popper would have agreed with Rawls: in an open society,
metaphysical, epistemological and complex moral debates are among the
main sources of vitality in a society. But each metaphysical thesis is legitimate,
and none of them has to be included in the principles of liberalism which are
supposed to be consistent. See Popper’s too short but important ‘theory of
liberal discussion’ in Conjectures and Refutations (1972[1963], chapter 17, §
IV), or his paper ‘The myth of the framework’ (Popper, 1994, chapter 2). Even
what appears to be consensual can be discussed (see Boyer, 1995). There
must, however, be an agreement about the rejection of (unnecessary) violence,
and a general and definite preference for peaceful and rational modes of
dealing with conflict in a democracy (public and open critical discussions,
rules, compromise – ‘from swords to words’).

INDIVIDUALISM
The moral commitment in Popper’s Open Society is overwhelming. His
characterizations of humanitarianism, equality, justice and cosmopolitanism
should be given serious consideration. As in Rawls, the most important concept
is that of a person, free, equal to others, and inseparable from his or her rights.
One of the methodological cum moral points of the OS is ‘individualism’ or
‘personalism’. From a methodological point of view, the idea is not, as Joseph
Agassi pointed out in an important paper (1960), to deny ‘existence’ to every kind
of collective (including social systems of rules, institutions), but to deny that any
non-individual entity (history, state, nation, class, society) has any psychological
characteristics of its own, such as views, interests, ends, preferences, rationality.
Learning for Democracy, Vol. 1, No. 2, 2005 15

Methodological individualism is different from psychologism and pure ‘atomism’.


It is, rather, a criticism of the collectivist fallacy, which consists in transferring
individual properties to collectives. Usually, that fallacy takes the form of
attributing to collectives a rational ‘behaviour’: if it is rational for individuals to do
(to prefer) X, it is rational for a collective to do (to prefer) X. According to Rawls,
utilitarianism is responsible for such a holistic fallacy. It transfers to the whole an
acceptable principle of individual rationality, namely, the principle according to
which I should impartially sacrifice a minor part of my life (say, a limb, or a week
of holidays) for the sake of my overall welfare: not to do so is to give into a kind of
weakness of the will. But to treat society as an individual, able to ‘sacrifice’ a part
of itself for the sake of the whole, is not to be ‘individualistic’ – it is immoral.
Rawls concluded, as have Popperians, that ‘utilitarianism is not individualistic’
(TJ, § 6, p. 29).
Like Popper, Rawls refused to conflate ‘individualism’ and ‘egoism’:
Popper proposed to oppose the first to ‘collectivism’ and the second to
‘altruism’. This clarification helped to reveal the existence of two other
important possibilities: ‘collectivist egoism’ (tribalism), and ‘individualistic
altruism’ (humanitarianism) (OS, I, chapter 6, p. 100).

SUFFERING
Were Popper and Rawls so greatly anti-utilitarian that they could simply be
classifiable as ‘moral Kantians’ (and therefore ‘anti-consequentialists’), as it
seems they both wished to be regarded? Admittedly, they were both Kantian
in their acceptance of the primacy of the idea of a person, to which rights are
attributed, and which are not to be subject to bargaining. The very notion of
the equal dignity of every human person, and the connected idea that one
should never treat any person merely as a means to an end but always as an
end in him- or herself, are typical ‘basic moral statements’ for both of them
(Rawls, 1999, p. 167). But I would stress that their Kantianism is not pure.
They were both much more concerned than Kant was with the problem of the
practical consequences of our acts and of our rules, in particular in terms of
empirical happiness – as are utilitarians, of course.
Let us say that ‘all of our conjectures should be true’ is some sort of an
epistemological ideal. Popper never said that falsehood was a regulative ideal!
Truth is. But the human (or Humean) situation is such that one cannot
definitely establish (non-trivial) universal factual truths. Even were it true,
scientists would still never be in a position to say: ‘The quest is ended!’ And
that is a logical truth. Our sole method is to try to find contradictions in our
system of expectations, and to do our best to eliminate them, in the hope of
approaching nearer to the truth. This is also true of moral matters. Nobody
should, from non-religious premises, come to the conclusion that ‘everybody
is happy’ is not a desirable state. But apart from a formal definition – such as,
the completion of our (undetermined and changing) ‘plan of life’ (see later) –
nobody is able to give an uncontroversial characterization of happiness, as
16 Is an Open Society a Just Society?

Kant noted. But the situation is not the same with unhappiness: the idea of
suffering is easier to grasp than that of happiness. There is, as Popper said, an
asymmetry. The existence of a moral consensus on the badness of Auschwitz
is not, of course, in itself a controversial issue. Who could seriously imagine
that this moral fact is open to discussion? Charles Larmore would rightly say
that there are limits to our moral fallibilility (1987), whereas Popper would
have talked of (ultimate) ‘decisions’ of our conscience.
It seems to me that what has been called (somewhat mistakenly) Popper’s
‘negative utilitarianism’ is not very different in intention from Rawls’s
difference principle: the intuition that governs that famous and controversial
principle is nothing other than the idea that the moral point of view is not one
that takes account of all sorts of happiness, but is one centred on the ‘worst off’
(that is, those experiencing the greatest difficulties in life). Popper’s criticism
of utilitarianism seems to anticipate that of Rawls:
the Utilitarian formula ‘Maximize pleasure’ […] assumes, in principle, a
continuous pleasure-pain scale which allows us to treat degrees of pain as
negative degrees of pleasure. But, from the moral point of view, pain
cannot be outweighed by pleasure, and especially not one man’s pain by
another man’s pleasure. Instead of the greatest happiness for the greatest
number, one should demand, more modestly, the least amount of
avoidable suffering for all; and, further, that unavoidable suffering –
such as hunger in times of an unavoidable shortage of food – should be
distributed as equally as possible.
(OS, I, chapter 9, note 2).
And the following sentence seems to be more Rawlsian than utilitarian or
Nozickian:
a systematic fight against definite wrongs, against concrete injustices or
exploitation, and avoidable suffering such as poverty or unemployment,
is a very different thing from the attempt to realize a distant ideal
blueprint of society.
(Popper, 1961[1957], § 24, p. 91)
This relates to the next section of this article.

THE NOTION OF A PLAN OF LIFE


The notion of a plan of life is not unimportant in Rawls’s theory, as Popper,
interested by the Lockean problem of personal identity, noted. Rawls borrowed
the notion from Josiah Royce: ‘an individual says who he is by describing his
purposes and causes, what he intends to do in his life’ (TJ, § 63, p. 408). The
plan determines someone’s conception of the good, and its rationality (to desire
to be honest and to be rich by corrupt means is rather inconsistent). In the end,
‘someone is happy when his plans are going well, his more important aspirations
being fulfilled, and he feels sure that his good fortune will endure’ (ibid., p. 409).
(‘He feels sure’ is surely too strong). Anyway,
Learning for Democracy, Vol. 1, No. 2, 2005 17

We must not imagine that a rational plan is a detailed blueprint for action
stretching over the whole course of life. It consists of a hierarchy of
plans, the more specific subplans being filled in at the appropriate time
[…] Revisions and changes at the lower levels do not usually reverberate
through the entire structure.
(Ibid., pp. 410, 411)
This point seems to have something to do with Duhem’s problem and the
so-called ‘revision of beliefs’.
The notion occupies a strategic place in Rawls’s approach, if only because it
helps to understand in a psychologically plausible way the idea of a conception
of what is good for a person, and because it clarifies the idea according to which
the principles of justice forbid the completion of some plans of life. It has also to
do with the (Aristotelian, Humboldtian and Millian) idea that to (try to) be
happy is to (try to) realize one’s best dispositions, as one guesses they are.
Institutions must encourage individuals (even the worst off) to formulate for
themselves a realistic plan of life (with regard to family, work, culture, sports,
friends, possibly political activity), and not to be desperate about its (limited)
propensity to be actualized (‘fair equality of chances’). Not to have any plan
would be to lose one’s sense of self-respect, the most important of the ‘primary
goods’, and then give way to depression, self-abasement and, perhaps, suicide.
The future should be open for all, as Popper would have it, not, or not only, in a
cosmological sense (‘the open universe’) but in a political liberal sense (cf. Isaiah
Berlin). This is, basically, the memorable idea of an ‘open society’. An open
society is the only positive way to approach that egalitarian situation, because it
encourages the plurality and the inventiveness of different and complementary
(but not unjust) plans of life, as in an orchestra, a Rawlsian analogy (TJ, § 79, and
2001, § 21, p. 3) that Popper could not but have loved.
The American moral philosopher Charles Larmore has fiercely, and in a
subtle manner, criticized the very notion of a plan of life (Larmore, 2004, pp.
251-8). In a nutshell, his criticism is that, according to Rawls, a plan of life,
even if our idea of it is ‘revisable’, is something fixed, something that one has
to ‘discover’, something given. According to Larmore, Rawls’s ‘plan of life’
presupposes that a person’s good (the supposed object of a plan of life) is
something fixed, awaiting discovery, something that one can contemplate as a
kind of theory, as if one could see one’s life as a whole. But, Larmore
continues, one cannot be ‘impartial’ about it as if we were outside ourselves
and not anchored in the present. He insists on the unpredictability of events
that unexpectedly and radically change one’s way of life (falling in love, like
Julien Sorel; having children, even if desired): ‘La vie qui serait la meilleure
pour (l’individu) n’est pas encore définie, et c’est à mesure qu’il vit qu’elle va
se définir.’ That is, in English: ‘The life that would be the best for the individual
is not yet determined, and it will come to be determined only as he lives.’8
This seems to be an excellent argument, which also goes against the Sartrean
idea of a (unique) ‘project’, for instance Baudelaire’s project (Sartre’s Flaubert is
18 Is an Open Society a Just Society?

much more complex, perhaps too complex). But I would propose that one could
perhaps reform the notion, without rejecting it completely. What I mentioned
about the absence of any plan of life for some very poor and unlucky people
seems to me to indicate a prima facie plausibility of the notion. The pleasant
and insistent ideas of children (‘I wish to be a fireman!’) show that there is
something deeply human in this capacity to project oneself into the future, even
if life will frustrate and betray most, but not all, of one’s projects. For example,
Rawls and Popper became important philosophers, and Popper claimed to be a
‘happy philosopher’. In contrast, some life plans are really bad (to live the life of
an SS officer, to be rich because one has exploited slaves). There is something
like an idea of ‘myself’ that includes an anticipation of what I should now like to
be, and this (vague) idea of myself is not entirely unstable: the ‘conversions’ in
life are rarely momentous events (recall St Paul, Augustine).
As was shown by Michael Bratman, the ideas of plan and subplan are very
useful in a theory of action and intention (Bratman, 1987). ‘The plan of life’
seems to be a very thick notion in comparison with the more modest idea of
momentary plan(s), but this indicates that the idea of a plan of life has to be
developed. At every moment, each person will have a certain idea of himself
and his current projects; but although this idea will change, at any given
moment the plan will not be entirely novel: I can recognize (as my own)
choices I made in the past, some of which I now reject as bad. Self-criticism,
problem-solving and disappointed expectations are the central tenets of
Popper’s anthropology. He would be the last philosopher to deny the
importance of unintended events, including some effects of one’s own actions.
I would argue for a Popperian pluralistic and dynamic view of the idea of
a plan of life, one that could perhaps deal with some of Larmore’s criticisms.
Not only is our view of our plan of life revisable, as though it were an objective
theory, but one can also treat it as such by, for example, discussing an
important personal decision with a close friend. The question has to do, as
Sartre would have said, with the choice of what kind of person you wish to be.
One could argue that a plan of life is itself provisional and subject to change
and radical criticism (incidentally, something that often produces its own
anxiety).9 (And of course, it is only at the end of the life that one can ask to
oneself if one has had a good life, as Solon argued against Cresus.) This seems
to be akin to what Popper argued in The Self and Its Brain:
In his in many ways very important book A Theory of Justice, John Rawls
introduced the idea of a plan of life […] to characterize the purposes or
aims which make of a man ‘a conscious, unified moral person’. I suggest
that this idea of a man-made World 3 plan of life may be somewhat
modified: it is not the unity of one unified and perhaps unchanging plan
of life which is needed to establish the unity of the self, but rather the fact
that there is, behind every action taken, a plan, a set of expectations (or
of theories), aims and preferences, which may develop and mature, and
which at times, though infrequently, may even change radically, for
Learning for Democracy, Vol. 1, No. 2, 2005 19

example under the impact of a new theoretical insight. It is this


developing plan which – following Rawls – gives unity to the person, and
which largely determines our moral character. […] It is the possession of
such a (changing) plan, or set of theories and preferences, which make us
transcend ourselves – that is to say, transcend our instinctive desires and
inclinations (‘Neigungen’, as Kant called them).
The most widespread aim in such a plan of life is the personal task of
providing for oneself and for one’s dependants. It may be described as
the most democratic of aims: remove it, and you make life meaningless
for many. This does not mean there is no need for a Welfare State to help
those who do not succeed in this. But even more important is that the
Welfare State should not create unreasonable or insurmountable
difficulties for those who try to make this most natural and democratic of
tasks a major part of their aims in life.
There is much heroism in human life: actions which are rational, but
undertaken for aims which clash with our fears, our instincts for safety
and security.
(Popper and Eccles, 1977, § 42, pp. 145-6).
The idea of ‘a set of expectations, aims and preferences, which may develop
and mature’ seems to me to be preferable to the idea of a rather static, unique
plan of life, determining the conception of the good for one person. In that
chapter, Popper argued that what is unique about humankind is our ability to
produce theories about ourselves as individuals, to be self-conscious, and
particularly to be aware of our death. He contrasted pure ‘programmes’ in
animals with the human capacity – thanks to the (social) language (World 3) –
to add to our (plastic) programmed behaviours ‘plans’ that permit us to
transcend ourselves, to dominate our basic instincts and to produce images of
ourselves, though not necessarily exact ones (‘Man is a story-teller’, he
frequently said). In the Popperian world, where reasons can have a non-
epiphenomenal causal role, events, discoveries and deliberations can change
your (vague) plan of life, so that the ‘best life’ you aim to live is never
determined in advance, as Larmore argues against Rawls. But one has to have
some representation of the dynamic unity of a self (we ‘learn to be a self’, ibid.,
§ 31, a self which is a process, a fire), such that even the others, relatives, judges
or historians, can more or less ‘understand’ (Verstehen) a person throughout
his entire life. Not all persons are real multiple selves (Elster, 1986).
A political analogy could be of some interest here: all ‘social engineering’
is ‘the construction of social institutions according to plan’ (Popper,
1961[1957], § 22, p. 73). Popper accepted the Hayekian idea, implicit in a
passage of René Descartes quoted by both of them, that most of the
institutions are ‘the result of the action of men but not of their design’
(Ferguson, quoted by Hayek, 1967, chapter 6). But some institutions can be
designed according to a plan of our own making (Hayek did not like that
20 Is an Open Society a Just Society?

concession to ‘constructivism’). And in that case, ‘the difference between


Utopian and piecemeal social engineering turns out, in practice, to be a
difference not so much in scale and scope as in caution and preparedness for
unavoidable surprises’ (Popper, 1961[1957], § 21, p. 69).
A ‘plan of life’ could be a complex and varying set of problems, desires,
expectations, values and rules (restraining the scope of the possible ends and
means): some plans can be dogmatic (‘neurotic’) (Popper, 1972[1963], chapter
1, § VI); some more critical (rational), and those are the ones that can progress.
Nothing would be static in a Popperian concept of a plan of life. (I am
convinced that Rawls was more a ‘piecemeal’ reformer than a Utopian holist;
but, contrary to Popper, he did not offer us a real methodology of reformism.)
The advice Popper gave to young scientists and philosophers could
perhaps be generalized:
to meet a problem, to see its beauty and to fall in love with it […] even if
you do obtain a solution, you may then discover, to your delight, the
existence of a whole family of enchanting though perhaps difficult
problem children for whose welfare you may work, with a purpose, to the
end of your days.
(Popper, 1983, Preface [1956])

THE FRIEND AND THE FOE


Popper stresses the fact that emotions are important, but their role should be
controlled by reason: ‘Our “natural reaction” will be to divide mankind into the
friend and the foe’ (OS, II, chapter 24, p. 235).
This important insight seems to me not only directed particularly against
Marxists and nationalists, but also against Carl Schmitt’s ideas. And even
though Rawls does not quote that important but controversial thinker (who
eventually became a Nazi), he was equally concerned with the idea of a society
as a system of competitive co-operation for mutual advantage, and not as a
domain of pure conflict (war, class struggle, zero-sum games). This is directly
linked with the strong equalitarianism of both thinkers. Rawls’s
equalitarianism is well known, but Popper’s was just as strong:
The adoption of an anti-equalitarian attitude in political life, i.e. in the field
of problems concerned with the power of man over man, is just what I
should call criminal. For it offers a justification of the attitude that different
categories of people have different rights; that the master has the right to
enslave the slave; that some men have the right to use others as their tools.
(OS, II, chapter 24, p. 236)

LIBERTY AND THE STATE


Clearly, with regard to political matters, Popper’s ideas evolved after 1945. In
The Open Society and Its Enemies he was a liberal social democrat of sorts
(see Shearmur, 1996, and Hacohen, 2000). But in the 1950s he became more
Learning for Democracy, Vol. 1, No. 2, 2005 21

conservative. This evolution is no doubt connected with Friedrich von Hayek’s


influence, but that is not really a philosophical point. This evolution has to do
also with a progressive awareness of the problem of bureaucracy in the
modern welfare states. From a Greek understanding of ‘democracy’ (or
Roman ‘Res Publica’), it follows that one criterion of a democracy, as opposed
to a ‘tyranny’, is the fact that magistrates have ‘to be accountable’: a negative
post hoc control (possibly reinforced by an a priori control). And the big
problem with bureaucracy, as understood by Max Weber (according to
Popper), is the difficulty, not to say the impossibility, of the open control of
bureaucrats. This matter is quite serious, and it seems not to have been taken
sufficiently seriously by Rawls. But let us take a look at some underestimated
dicta from the two philosophers.
First, even the mature Popper was not a libertarian:
There are ideological worshippers of the so-called ‘free market’ (to
which we naturally owe a great deal) who think that […] legislation
limiting market freedom is a dangerous step down the road to serfdom.
But that […] is ideological nonsense. Forty-six years ago in the first
edition of The Open Society and its Enemies, I have already shown that a
free market can exist only within a legal order created and guaranteed by
the state.
(Popper, 1999, chapter 9: All life is problem solving [1991], p. 101)
The allusion to the famous book by Hayek is intriguing. It should not be
understood as a rejection of Hayek’s main ideas. In 1992 Popper declared
that his own OS took ‘a far smaller part than the books of my late friend
Friedrich von Hayek, for example […] The Road to Serfdom’ in ‘undermining
Marxism and the Soviet Empire’ (Popper, 1999, p. 155). In other words,
Popper thought that the excellent Hayekian criticism of the planning ideology
does not imply that any ‘legislation’ limiting market processes would be bad,
for this would also be an ideology. The reference to OS seems to be to the
following passage:
Liberalism and state-interference are not opposed to each other. On the
contrary, any kind of freedom is clearly impossible unless it is guaranteed
by the state […] the important and difficult question of the limitations of
freedom cannot be solved by a cut and dried formula. And the fact that
there will always be borderline cases must be welcomed, for without the
stimulus of political problems and political struggles of this kind, the
citizen’s readiness to fight for their freedom would soon disappear, and
with it, their freedom. (Viewed in this light, the alleged clash between
freedom and security, that is, a security guaranteed by the state, turns out
to be a chimera. For there is no freedom if is not secured by the state; and
conversely, only a state which is controlled by the free citizens can offer
them any reasonable security at all.)
(OS, I, chapter 6, § VI, p. 111)
22 Is an Open Society a Just Society?

Incidentally, it should be remarked that Popper held more or less the


same view as Rawls about what one can now call ‘republicanism’: it would be
quasi totalitarian to argue that the only good life is the ‘political life’ (‘civic
humanism’), but it is quite important that the free citizens give some
importance to the control of their leaders and to the defence of just
institutions (‘classical republicanism’) (Rawls, 1993, V, § 7). About
republicans, I would simply add that they adopt the same definition of liberty
that Hayek (1960) did, but without mentioning him, and that this definition
(‘not to be dominated’) comes from the Greeks, as Popper had also argued,
and not only from the Romans, as the republicans claim.10 (Popper and
Rawls belong to the classical republican tradition that emerged with the
Athenians.)
Many philosophers have sought an answer to the question: ‘What is the
origin of political obligation?’ Popper proposed to change the question to:
‘What do we demand from a state?’ That seems to me to be rather similar to
Rawls’s question: ‘What should be the principles of the basic institutions of
society?’ But Popper never said that the state was unnecessary, only that it
was a ‘necessary evil’ – a reaction against Hegelianism. But he also laid down
two important principles: (1) the Liberal Razor – the state’s powers should not
be multiplied beyond necessity; (2) the moral necessity of the state (a thesis
which seems not to be entirely in harmony with the idea of ‘necessary evil’):
Those […] who think that every person should have a right to live, and
that every person should have a legal claim to be protected against the
power of the strong, will agree that we need a state that protects the
rights of all.
(Popper, 1972[1963], p. 350)11
This thesis of a limited but ‘protectionist’ state is substantial, and Rawls
could have accepted it.

JUSTICE
What do humanitarians call ‘justice’? Popper proposed the following:
(a) an equal distribution of the burden of citizenship, i.e. of those
limitations of freedom which are necessary in social life; (b) equal treatment
of the citizens before the law, provided, of course, that (c) the laws show
neither favour nor disfavour towards individual citizens or groups or classes;
(d) impartiality of the courts of justice; and (e) an equal share in the
advantages (and not only in the burden) which membership of the state may
offer to its citizens.
(OS, I, Chapter 6, § 1, p. 89)
A few pages later, Popper claimed that justice was ‘the impartial weighing
of the contesting claims of the individuals’ (ibid., p. 106). In my opinion, not
much can be said from a Rawlsian point of view against such a nicely
formulated statement (to be considered not as a verbal definition, but still as
Learning for Democracy, Vol. 1, No. 2, 2005 23

a proposal about what should be our concern), and in which each word is
important. The same holds true of this surprisingly similar sentence by Rawls:
Justice means essentially the elimination of arbitrary distinctions and the
establishment within the structure of a practice of a proper share,
balance or equilibrium between competing claims.
(Rawls, 1999, p. 191)
Even the difference between the rational and the reasonable, so important
for Rawls,12 had been anticipated by Popper: to be rational is to act according
to one’s own analysis of the situation, to find a tentative solution that fits the
problem as seen by the agent (‘rationality principle’). The idea is morally
neutral. Besides, the ‘attitude of reasonableness’ is a moral one:
It is an attitude which tries as far as possible to transfer to the field of
opinions in general the two rules of every legal proceeding: first, that one
should always hear both sides, and secondly, that one does not make a
good judge if one is a party to the case.
(Popper, 1972[1963], p. 356)
It would take too long to compare in a systematic manner these
characterizations of justice with Rawls’s principles, which are more precise.
Nothing is said, for instance, about a possible hierarchy of principles, which is
the most original contribution of Rawls’s TJ. But it seems to me that the
notions articulated clearly by Popper are sound, and could belong to the
Rawlsian ‘overlapping consensus’.
My contention is only that Popper would not have introduced the
difference principle in the common necessary postulates of a liberal society:
discussions on desert, on the level of intervention of the state, on property, etc.
are to be left to the normal political debate, his favourite system being a
bipartite one. If one regards (a bit mistakenly) Rawls as the chief advocate of
the welfare state, and Hayek as one of his best critics, it could seem that
Popper is, in the end, more favourable to Hayek. His criticism of bureaucracy
has no equivalent in Rawls, except perhaps in one or two sentences. But Rawls
was certainly less concerned than Popper was by the problem of the
inefficiency of bureaucracy, and of the possible unexpected effects of
seemingly ‘just’ laws.

WELFARE STATE AND LIBERALISM


One thing has not been sufficiently appreciated. Admittedly, in TJ, Rawls
seems somewhat unaware of the unintended consequences of human action,
and specifically those of an important intervention of the state in the market
process. Rawls is perhaps insufficiently aware of the most important fact,
called by Popper ‘the ambivalence of the institutions’ (Popper, 1972[1963],
chapter 4). But both thinkers are very clear about the necessary role of
institutions, or ‘practices’ (Rawls), and that is another point they share.
24 Is an Open Society a Just Society?

Popper explained that only traditions can make the link between concrete
persons and abstract institutions (ibid.), and one can remark that Rawls
insisted on the great importance of the fact that persons in ‘just’ institutions
should develop a moral sense of justice, for the institutions to be stable.
Anyway, with arguments not so different from Popper’s well-balanced
criticism of ‘paternalism’, Rawls later made it plain (2001, IV) that his own
ideal was no more an omnipotent welfare state. But he criticized it with a more
equalitarian oriented idea that a welfare state is compatible with inequalities
criticizable thanks to the difference principle (I regard the principles of justice
as critical regulative principles): inequalities that do nothing for the worse off,
and leave these people at the mercy of the almighty paternalistic state. An
omnipotent even if benevolent welfare state does not ensure the capacity for all
to develop a sense of responsibility and self-respect. In view of the fact that a
pure, free (laissez-faire) market and a planned economy are both clearly unjust,
Rawls declares that the only positions that are defensible are ‘liberal socialism’
(inspired by a suggestion from Mill13) and, his favourite, the ‘democracy of
owners’ (with a large distribution of private property over the whole society and
an extension of deliberative democracy). Unfortunately, he is very brief about
that prospect, but I do not think Popper would have objected that it was, as
such, a bad, or a Utopian idea. The way towards that kind of society should not
of course be authoritarian, but ‘piecemeal’ and made possible by incentives.
The main idea, already expressed by Popper, is that an excessive concentration
of property is a danger for democracy itself.
Both philosophers are democrat political liberals, and they both
anticipated a liberal answer to the so-called ‘communitarian’ criticism of
‘abstract’ liberalism: Popper in his theory of the abstract society (OS, I, chapter
10) and of the possibility and necessity of free subgroups in an open society,
and Rawls with his idea that a pluralistic society is not a community, with a
unique common good for all, but ‘a social union of social unions’ (Rawls, TJ, §
79). Rawls borrowed the idea of ‘social union’ from Wilhelm von Humboldt
(incidentally, one of the best liberals for Mill as well as for Popper and Hayek).
Is it not then surprising that a communitarian such as Charles Taylor thinks
he has discovered an anti-liberal argument when he uses Humboldt as a
romantic (Taylor, 1989, pp. 414 and 458)? There is something wrong with his
criticism of liberalism, as if this view was a pure ‘atomism’: neither Rawls nor
Popper would accept the idea of the non-social character of man or would say
that an individual is completely determined independently of his social
situation (a point Popper condemned as ‘psychologistic’ in Mill’s
epistemological writings, but which cannot be found in On Liberty).
In an open society, debates between Rawlsians, Nozickians, and others are
to be expected, and that is a political ‘good’. Popper’s theory of an open society
asserts the minimum set of principles that have to be accepted in liberal
societies, just as Popperian meta-philosophy (critical discussion) is minimal
and the best meta-philosophy, even for anti-Popperians. Rawls’s theory of
Learning for Democracy, Vol. 1, No. 2, 2005 25

justice is less neutral (in particular, the difference principle). An open society
is not necessarily a Rawlsian perfectly ‘just’ one, but a Rawlsian just society
would be, necessarily, an open society.

CORRESPONDENCE
Prof. Alain Boyer, Université de Paris IV (Sorbonne), UFR de Philosophie, Paris,
France. Email: boyeral@wanadoo.fr

NOTES
1 Let me thank David Miller for having read my article and suggesting many
improvements. Part of the article was given as a lecture at Philosophy: Problems,
Aims, Responsibilities, a conference, organized by David Miller, to mark the 10th
anniversary of the death of Karl Popper, University of Warwick, UK, 16-18
September 2004.
2 Harvard edition (1971a), § 17. Rawls referred in a footnote to Theodosius
Dobzhansky (1962).
3 In Popper and Eccles (1977, p. 145) where he writes about the notion of a ‘plan of
life’.
4 About an even more secondary point (the relations between Glaucon’s theory in
the Republic and the Sophists): in Rawls’s Collected Papers (1999) and ‘Justice
as reciprocity’ (1971b, p. 204n), The Open Society and its Enemies, pp. 112-18
(American edition, 1950), is quoted. (In my article, page references to OS will
refer to the 1966[1945] edition.) One can remark that Plato’s city is not a model
of justice for Rawls, any more than it was for Popper. They both reject the ‘noble
lie’ (TJ, § 69), the philosopher king and the authoritative holism of the ‘greatest
of all philosophers’ (Popper). But Rawls insists more on the relevance of some
aspects of Aristotle than Popper did in an interesting but rather unfair chapter of
OS (chapter 11). Anyway, I find it strange that Rawls did not quote Popper more
often. Perhaps he judged OS and The Poverty of Historicism to be too polemical.
5 The concept was used by Goodman for dealing with ‘inductive logic’. Popperians
think, of course, that no equilibrium can possibly be found in this area. See Miller
(1994).
6 Note that a lexical ordering cannot be represented by a real-valued continuous
utility-function.
7 ‘We need a faculty which makes us see the end from afar, and intuition is this
faculty.’
8 Thanks to Charles Larmore for translating this sentence of his from French into
English! Also, for commenting on his own position to help me improve my
formulation of it.
9 If (provisional) plans of life could not change radically, one would have to be
much more pessimistic about the idea that punishment can at least sometimes
reform someone.
10 See Boyer (2001b).
11 This is a ‘republican’ point.
12 A rational Hobbesian cannot deduce the necessity of a fair distributive justice;
that is what Rawls concluded from what he regarded as David Gauthier’s failure
(Rawls, 1993, II, § 1). Rawls’s problem was to find an acceptable interpretation
of the words ‘arbitrary distinctions’ and ‘proper balance’ (TJ, § 1).
26 Is an Open Society a Just Society?

13 Mill offered a liberal and Darwinian argument in favour of self-managed firms:


the workers would be more favourable to the firm, the wages less high, etc.
Eventually, in a market system, they would win. Rawls remarked that this had
not been the case. This is a problem for sociologists (considering Mill’s conjecture
as a ‘zero model’, in the Popperian sense).

REFERENCES
AGASSI, J. (1960) Methodological individualism, British Journal of Sociology, 11, pp.
244-70. (Also in: J. O’Neill (ed.) (1973) Modes of Collectivism and
Individualism. New York: St Martin’s Press.)
BARRY, B. (1995) Justice as Impartiality. Oxford, UK: Oxford University Press.
BOYER, A. (1995) Democracy and disagreement. Ratio Juris, 8 (1), pp. 1-8.
BOYER, A. (2001a) Hors du Temps: Un Essai sur Kant. Paris: Vrin.
BOYER, A. (2001b) On the modern relevance of old republicanism. The Monist, 84,
(1), pp. 22-44.
BRATMAN, M. (1987) Intentions, Plans, and Practical Reason. Cambridge, MA:
Harvard University Press.
DOBZHANSKY, T. (1962) Mankind Evolving. New Haven, CT: Yale University Press.
ELSTER, J. (ed.) (1986) The Multiple Self. Cambridge, UK: Cambridge University
Press.
FOOT, P. (ed.) (1967) Theories of Ethics. Oxford, UK: Oxford University Press.
GOODMAN, N. (1955) Facts, Fiction and Forecast. Cambridge, MA: Harvard
University Press.
HACOHEN, M. H. (2000) Karl Popper – The Formative Years, 1902-1945: Politics and
Philosophy in Interwar Vienna. Cambridge, UK: Cambridge University Press.
HAYEK, F. A. (1960) The Constitution of Liberty. London: Routledge and Kegan Paul.
HAYEK, F. A. (1967) Studies in Philosophy, Politics and Economics. London:
Routledge and Kegan Paul.
LARMORE, C. (1987) Patterns of Moral Complexity. Cambridge, UK: Cambridge
University Press.
LARMORE, C. (2004) Les Pratiques du Moi. Paris: PUF.
MILLER, D. (1994) Critical Rationalism, A Restatement and a Defence. La Salle, IL:
Open Court.
POINCARÉ, H. (1970) La Valeur de la Science. Paris: Flammarion (first edition 1905).
POPPER, K. R. (1959) The Logic of Scientific Discovery. London: Hutchinson (first
German edition 1934).
POPPER, K. R. (1961) The Poverty of Historicism. London: Routledge and Kegan Paul
(first English edition as a book, 1957).
POPPER, K. R. (1966) The Open Society and its Enemies, Volume 1 – The Spell of
Plato. London: Routledge and Kegan Paul (first edition 1945).
POPPER, K. R. (1966) The Open Society and its Enemies, Volume 2 – The High Tide
of Prophecy: Hegel, Marx, and the Aftermath. London: Routledge and Kegan
Paul (first edition 1945).
POPPER, K. R. (1972) Conjectures and Refutations: The Growth of Scientific
Knowledge. London: Routledge (first edition 1963).
POPPER, K. R. (1979) Die beiden Grundprobleme der Erkenntnistheorie. Tübingen:
Mohr Siebeck (ed. T. E. Hansen) (unpublished version 1932).
POPPER, K. R. (1979) Objective Knowledge: An Evolutionary Approach. Oxford:
Oxford University Press (first edition 1972).
Learning for Democracy, Vol. 1, No. 2, 2005 27

POPPER, K. R. (1983) Realism and the Aim of Science. London: Hutchinson.


POPPER, K R. (1994a) The Myth of the Framework: In Defence of Science and
Rationality. London: Routledge (ed. M. A. Notturno).
POPPER, K. R. (1999) All Life is Problem Solving. London: Routledge (some chapters
trans. from German by P. Camiller).
POPPER, K. R. and ECCLES, J. C. (1977) The Self and Its Brain. London: Springer
International.
QUINE, W. V. (1990) Pursuit of Truth. Cambridge, MA: Harvard University Press.
RAWLS, J. (1971a) A Theory of Justice. Cambridge, MA: Harvard University Press.
RAWLS, J. (1971b) Justice as reciprocity. In: S. Gorovitz (ed.) Utilitarianism: John
Stuart Mill: With Critical Essays, pp. 242-68. New York: Bobbs-Merrill.
RAWLS, J. (1993) Political Liberalism: The John Dewey Essays in Philosophy, 4. New
York: Columbia University Press.
RAWLS, J. (1999) Collected Papers. Cambridge, MA: Harvard University Press (ed. S.
Freeman).
RAWLS, J. (2001) Justice as Fairness: A Restatement. Cambridge, MA: Harvard
University Press (ed. E. Kelly).
SETTLE, T. (1976) In Search of a Third Way: Is a Morally Principled Economy
Possible? Toronto: McLelland and Stewart.
SHEARMUR, J. (1996) The Political Thought of Karl Popper. London: Routledge.
TAYLOR, C. (1989) Sources of the Self. Cambridge, MA: Harvard University Press.