The following text has been automatically reproduced by an Optical Character Recognition (OCR) algorithm. It may not have been checked over by human eyes. For matters of precision please consult the original pdf.

In partial praise of a positivist

In partial praise of a
The work of Otto Neurath
John O’Neill

There is a tradition in socialist writing of rediscovering
neglected socialist thinkers and showing how the
recovery of their memory can contribute to the solution
of contemporary problems in socialist theory and
practice. This paper belongs to this genre of rediscovery.

The theorist with whom I am concerned was an
Austrian Marxist. He played an active part in the German
revolution that followed the First World War: some of
his best work on socialist planning was written as
addresses to the workers’ councils of Germany, and he
acted as director of the agency responsible for
socialization during the soviet phase of the Bavarian
revolution. Following the defeat of the Bavarian
revolution he was brought to trial, during which Max
Weber testified in his defence. His work on socialist
economics formed one of the starting points of the
socialist calculation debate, being an object of criticism
in the arguments against socialism developed by Mises
and Weber, and later Hayek. In the 1920s, ’30s and ’40s
he continued to develop a defence of socialism which
examined problems of socialist planning in a way that
took seriously the ecological dimension of socialist
thought, and recognized the problems of reconciling
individual freedom and economic planning. He was also
a major philosopher of science whose work in that area
has continued to be influential. However, in the history
of socialist thought he rarely gets a mention, and no
standard survey of Western Marxism discusses his
socialist ideas. Indeed, he tends to be known by two
quotations, which might give the impression that his
main concerns were boat-building at sea and the role of
coffee drinking in the development of sociological
thought. I The reason for the neglect is that my socialist
philosopher, Otto Neurath, was a positivist and a leading
member of the Vienna Circle.

Positivism has become a term of abuse in academic
circles generally, and particularly amongst socialists.

From the Frankfurt School the story has emerged that
positivism is a conservative doctrine necessarily
committed to existing social institutions and to a
technocratic conception of politics. 2 Even the most
scientistic orthodox Marxist is unlikely to announce that
she is a positivist. Such is the disrepute into which
positivism has fallen that to accept the title of positivist
would amount to an admission that one’s position was
untenable. The picture of positivism that informs its use
as a term of academic abuse is a caricature. Positivist
philosophy was much more heterogeneous than recent
thumbnail versions allow, and many of the doctrines
ascribed to it were explicitly rejected by many of its
proponents. 3 Neurath himself was unhappy with the term
for the very reason that it suggested a systematic set of
doctrines incompatible with the methodological
pluralism he defended, although ‘not being a pedant’ he
was willing to ‘bear it’ .4
It is a feature of fashion in intellectual history that one
generation reduces the previous generation’s orthodoxy
to a few simple and easily criticizable slogans, only for
the next generation to rediscover their grandparents’

genius. No doubt the time for the positivist movement to
be rescued from simplification will come. However, it is
not my purpose to defend positivism, still less to resurrect
it. Many of the epistemological, ontological and ethical
doctrines Neurath himself defends are ones that I would

My purpose in this paper is narrower: to show why
his social and political thought remains worthy of
consideration. It is of value for at least three reasons.

First, while it was at the heart of the socialist calculation
debate, Neurath’s contribution to the defence of
socialism in that debate has been forgotten. That it has
been lost to memory has weakened the socialist case, and
rendered that for the market stronger than it is. Second,
his work on unified science, which also indirectly

Radical Philosophy 74 (Nov/Oec



addresses problems of planning and markets, raises
problems of importance that have been lost in the
caricature of the project for unified science. Third,
Neurath made a contribution to the development of an
associational account of socialism that took seriously the
problems of reconciling socialism with both ecological
ends and individual freedom. All three features of his
work render it of particular relevance to current problems
for socialist and Green theory and practice.

The socialist calculation debates
There is a received story about the socialist calculation
debate that goes like this. In a paper published in 1920,5
later incorporated into a book in 1922, 6 Mises presented
an argument which asserted that rational economic
calculation would not be possible within socialism.

Hence, a socialist economy was not a real practical
possibility. The core of Mises’ argument was refined by
Hayek as an epistemic argument against the possibility
of a planned economy. 7 The socialist response to this
position was articulated by Lange and Taylor. 8 The
socialist calculation debate is then presented primarily
as a conflict between Mises and Hayek on the one hand,
and Lange and Taylor on the other, different sides being
accorded the laurels of victory. 9 This story is
unsatisfactory. First, this account of the debate is focused
on the later English-speaking phase of the debate and
underplays the initial German-speaking chapter which
involved Neurath, Polanyi and others. Second, and more
significant, it has given a mythical unity to the debate.

There was no socialist calculation debate. There were at
least two debates that concerned two independent
objections to the possibility of socialism. The first debate,
to which Mises’ article was a contribution, was an
argument about rational choice and commensurability specifically the possibility of rational economic action in
the absence of a single unit of comparison between
alternati ve economic activities. N eurath’ s direct
contribution to the socialist calculation debates was to
this phase, and his arguments formed I believe a
convincing response to Mises’ objection to socialism.

The second debate, instigated by Hayek’s epistemic
objection to socialism, concerned the possibility of
planning, given the dispersal of knowledge amongst
different actors in an economy. While Hayek presents
his epistemic argument as a continuation of the first
debate, it in fact forms a departure from it. Neurath made
no direct contribution to the second debate. However,
his writings in the 1930s on the unity of the sciences do
have a bearing on that argument.

Mises’ arguments against a socialist economy,
understood as an economy without a market in


production goods, turned on a point about
commensurability: rational economic decision-making,
beyond the most simple individual decisions, requires a
single measure on the basis of which the worth of
alternati ve states of affairs could be calculated and
compared. Thus, for example, given the choice ‘whether
we shall use a waterfall to produce electricity or extend
coal and better utilize the energy contained in coal’ ,10 we
need some way of calculating the advantages and benefits
of alternatives, and this in turn requires a common unit
of measurement. The ‘subjective use-value of
commodities’ provide no units for computation ‘judgements of value do not measure: they arrange, they
grade’ .11 Hence such values cannot enter directly into
comparisons between options. A common unit of
measurement is provided by monetary prices in the
market: ‘calculations based upon exchange values enable
us to reduce values to a common unit’ .12 It forms, indeed,
the only adequate unit of comparison. Comparability
between options requires monetary prices that measure
exchange values such that one is able to have a
determinate answer to the advantages of alternatives by
way of simple rules.

In developing this objection to socialism, one of
Mises’ central targets was Neurath’s 1919 report to the
Munich Workers’ Council, ‘Through War Economy to
Economy in Kind’.13 Neurath had there argued that a
socialist economy, since it was to consider the use·value
of goods only, would have to be a non-market ‘economy
in kind’ , in which there would exist no role for monetary
units to compare options:

We must at last free ourselves from outmoded
prejudices and regard a large-scale economy in
kind as a fully valid form of economy which is the
more important today in that any completely
planned economy amounts to an economy in kind.

To hold on to the split and uncontrollable monetary
order and at the same time to want to socialize is an
inner contradiction. 14
In such an economy, while physical statistics about
energy use, material use and so on would be required,
there would be no need for a single unit of comparison.

There are no units that can be used as the basis of a
decision, neither units of money nor hours of work.

One must directly judge the desirability of the two
possibilities. 15
There is no simple unit for decision-making. Rather, one
requires direct comparisons of alternatives, and hence
there is no possibility of excluding political and ethical
judgements from even ‘technical’ decisions. In making
this claim, Neurath is not only criticizing the market, but

also socialist alternatives to the market that employ
single units in making decisions, be these labour hours
or the energy units of earlier ecological economists such
as Popper-Lynkeus and Ballod-Atlanticus.1t is a position
Neurath reaffirms in his later contributions to the socialist
calculation debate. Thus in his 1928 article ‘Personal Life
and Class Struggle’, Neurath takes up Mises’ examples
of choosing between alternative sources of energy and
responds thus:

The question might arise, should one protect coal
mines or put greater strain on men? The answer
depends for example on whether one thinks that
hydraulic power may be sufficiently developed or
that solar heat might come to be better used, etc. If
one believes the latter, one may ‘spend’ coal more
freely and will hardly waste human effort where
coal can be used. If however one is afraid that when
one generation uses too much coal thousands will
freeze to death in the future, one might use more
human power and save coal. Such and many other
non-technical matters determine the choice of a
technically calculable plan … we can see no
possibility of reducing the production plan to some
kind of unit and then to compare the various plans
in terms of such units … 16
Rational practical thinking need not involve any single
unit that reduces decision-making to a purely technical
procedure. It requires ethical and political judgement.

The debate between Mises and Neurath turns on
differences concerning the nature of practical rationality.

For Mises any rational decision, beyond the most simple,
requires the commensurability of different values. There
needs to be a single common unit which reduces the
choice between different options to a matter of
calculation. Mises assumes an algorithmic conception of
practical reason. Rational decision-making requires the
application of mechanical procedures of calculation to
arrive at a determinate answer to any question. 17 Neurath
rejects this account of rational choice for both practice
and theory. It exhibits what he calls ‘pseudorationalism’.

The basis of Neurath’s objections to this view are to be
found in two of his earliest papers, ‘The Lost Wanderers
of Descartes and the Auxiliary Motive’ (1913) and ‘The
Problem of the Pleasure Maximum’ (1912).

In ‘The Lost Wanderers of Descartes and the
Auxiliary Motive’ Neurath criticizes the algorithmic
view of reason, that one can give a set of rules that
determine unequivocally a particular decision: ‘in many
cases, by considering different possibilities of action, a
man cannot reach a result.’ 18 Our knowledge that informs
decision-making is uncertain and the rules of rationality
rarely determine a unique answer given what is known.

A rationalist who believes in reason must recognize the
boundaries of the power of reason in arriving at
decisions: ‘Rationalism sees its chieftriumph in the clear
recognition of the limits of actual insight.’ 19 It is a mark
of the pseudorationalist to believe that there exist rules
of insight that determine answers to all decisions.

Pseudorationalism exists not only in the domain of action
but also of thought, in the belief that there exist rules for
the scientific method which if followed eliminate
falsehood and lead to ever nearer approximations to the
truth. What marks the philosophy of Descartes is a
realization of the limits of rules of reason in action, but a
failure to recognize similar limits in the rules for the
direction of the mind. Just as in action, so in theoretical
matters, reason underdetermines our theories.

It is on this basis that Neurath later criticizes Popper’s
philosophy of science in ‘Pseudorationalism of
Falsification’ (1935b), and his arguments in that paper
develop clearly his rejection of an algorithmic
conception of reason. Popper’s falsificationism exhibits
pseudorationalism in the domain of thought: it is driven
by the belief that valid scientific argument is fully
capturable in a set of deductive rules that unequivocally
eliminate candidates for the truth. Against that view
Neurath notes the now much repeated observation that
the historical development of the sciences required that
some statements be regarded unfalsifiable. 20 Against both
‘the absolutism of falsificationism … and the absolutism
of verificationism’21 (and it is notable that he does not
take verificationism to define positivism) Neurath
defends a principle of methodological pluralism and

We believe we are doing the most justice to
scientific work if, in our model construction, we
set out from the assumption that always the whole
mass of statements and all methods can come
under discussion …. Various factors determine the
methodical scientist in his choice of a model. We
deny that the encyclopedia preferred by the
scientist can be logically selected by using a
method that can only be generally outlined.

Together with this we not only deny that there
could be general methods of ‘induction’ for the
factual sciences, but also that there could be
general methods of ‘testing’ – however, Popper
advocates just such general methods of ‘testing’ .22
N eurath’ s critique of Popper deserves to be better
known. It anticipates much of more recent post-Kuhnian
philosophies of science. The rejection of falsificationism
on the ground that it is inadequate to the history of
science predates Kuhn. The belief in the impossibility of
a simple general method for science predates


Feyerabend. Indeed, like Feyerabend, N eurath’ s
tolerance extends even to the rules of deduction and the
demands of consistency:

I knew very well that in any consistent system of
statements a single contradiction would ‘infect’,
as it were, the whole body and would enable one to
infer anything one pleased. I also knew that in the
practice of scientific behaviour, occasional
contradictions did not destroy the work. … Our
scientific practice is based on local systematizations only, not on the overstraining bow of
deduction. 23


However, unlike Feyerabend, Neurath recognizes that
this position is quite compatible with a proper
rationalism, since such rationalism is consistent with the
recognition of the limits of rules of reason. It is
pseudorationalism to believe that the rules of reason
determine unique answers for us in the domains either of
action or of thought. There are no algorithms that
substitute for judgement applied in local contexts.

In his paper ‘The Problem of the Pleasure Maximum’

Neurath criticizes a second assumption that Mises was to
make in his opening salvo against socialism, that values
are commensurable – that is, that there is a scale of values
according to which options can be uniquely ordered.

Neurath rejects that assumption, somewhat surprisingly,
from within a hedonist perspective. Neurath throughout
his writings defends a utilitarian and Epicurean position
which takes the good of social policy to be the
maximization of happiness understood as pleasure.

However, he rightly rejects the possibility of units of
pleasure on which calculations could be made. 24 Even
given the aim of pleasure maximization, there is no
possibility of a purely technical ordering of states of
affairs: pleasures are themselves incommensurable. In
his work on planning this point has a more general
significance. It follows that even on the simplifying
assumption of a single evaluative category, no planner
could ignore substantial value questions and treat a
decision in ethically neutral technical terms.

The rejection of the pseudorationalism of algorithmic
rules and of the assumption of value commensurability
informs Neurath’s conception of non-market socialism
as an economy of kind, and lies at the basis of his
arguments in the socialist calculation debate. Mises’

attack on the possibility of socialism exhibits precisely
the kind of pseudorationalism in the domain of practical
reason that Neurath had attacked in his earlier writings.

In so far as this first phase of the socialist calculation
debate is concerned, Neurath’s position is the stronger.

He rightly allows that comparability need not assume
commensurability, that there is not any rule that can be


mechanically applied to produce a determinate decision
as to which plan to adopt, and that there is an ineliminable
role for non-technical judgement in the most technical of
decisions. 25 Here the myth that there was a single socialist
calculation debate gets in the way of seeing the strengths
of socialist positions like Neurath’ s. Contrary to Hayek’ s
accepted definition of the debate, Neurath’s account of
social and economic planning allowed for a role for nontechnical judgements, which his Austrian opponent in
the socialist calculation debate denied. It was not
positivists like Neurath who had an algorithmic
conception of practical rationality but their opponents.

The myth that there was a single debate has had other
unfortunate consequences for socialists. The absence of
subsequent discussion of Mises’ original argument has
meant that the commensurability assumption Mises
defends has become an unquestioned dogma. The
assumption that rational economic activity requires
commensurability was accepted without question by
Lange and Taylor in their contribution to the debate. It
has also been largely assumed elsewhere in both Austrian
and neo-classical economics and has re-emerged as an
explicit assumption in environmental economics. Thus,
for example, the attempt to shadow price all
environmental goods for the purposes of cost-benefit
analysis is founded on the claim that rational decisionmaking requires a single monetary unit of calculation.

Typical is the comment in the influential Pearce mport:

‘CBA is the only [approach] which explicitly makes the
effort to compare like with like using a single measuring
rod of benefits and costs, money’ .26 Neurath’s response
to Mises has continuing significance for both socialists
and Greens and deserves to be rescued from its
undeserved obscurity.

The unity of the sciences
An examination of Neurath’s papers of the 1930s
suggests that in terms of quantity alone, the project of a
unified science dominated all others. 27 The concern was
related to his treatment of the problems of socialist
planning. Planning raises not only problems of value
commensurability, but also problems concerning the
division of knowledge between different groups and
individuals in society. Is rational social planning possible
given the division of knowledge? This question forms
the basis of the second phase of the socialist calculation
debate instigated by Hayek. While Hayek presents his
position as a development ofMises’ position, it is so only
if one puts heavy emphasis on some of Mises’ passing
comments and ignores the central arguments about
commensurability. Hayek’s position was a departure in
the debate. His argument against planning is epistemic –

that the social division of knowledge, the dispersal of
knowledge amongst different actors, rules out the
possibility of rational plans in a socialist economy. While
Neurath does not address that argument directly, the
project of a unified science aims in part at showing that
the divisions of knowledge could be overcome for the
purposes of rational socialist planning.

Subsequent accounts of the project of a unified
science have tended to oversimplify the programme. The
project took one of four forms: (1) a reductionist project
in which all the sciences would be logically derivable via
bridge-laws from physics;28 (2) a programme for a
unified method which would be followed by all sciences;
(3) a project for a unified language of science; and (4) a
project that would integrate the different sciences, such
that, on any specific problem, all relevant sciences could
be called upon – a project for the ‘orchestration of the
sciences’ .29 All four doctrines were defended by
positivists in different stages of its history. However, in
subsequent critical accounts of the doctrine, the first has
tended to be taken to define the project, and the last has
tended to be ignored, or at least has been taken as a
project of integration through reduction. This picture
again ignores the heterogeneity of the positivist
movement and of the arguments that it generated.

Neurath rejects the first reductionist project
completely: ‘would it not be preferable to treat all
statements and all sciences as coordinated and to
abandon for good the traditional hierarchy: physical
sciences, biological sciences, social sciences and similar
types of “scientific pyramidism”?’30 Opposition to
pyramidism runs through Neurath’s work on unified
science. So also does a rejection of the second doctrine,
the possibility of a unified method for the sciences. As
we have seen, that search Neurath took to form a part of
pseudorationalism: hence his opposition to the
absolutism both of falsificationism and of
verificationism. On method Neurath was a pluralist.

Indeed, the denial of pyramidism is based in part on this
pluralism and his rejection of pseudorationalism. Given
that there is no set of rules that determine a unique answer
to either practical or theoretical matters, one cannot
rationally expect to arrive at a deductively closed and
consistent set of statements, in which statements in the
‘higher’ sciences are deduced from those below them.

Hence his advocacy of ‘encyclopedism’: ‘I thought it in
accord with the historically given situation to
acknowledge … “localized” contradictions, and to think
of an “encyclopedia as a model” as intentionally opposed
to the “system as a model”.’31
In defending the programme for a unified science,
Neurath was concerned to defend the third and fourth

projects, that of unifying the language of science and that
of the coordination of the sciences. In the project of a
universal language or ‘jargon’ for the sciences, Neurath
appears in the guise that the later images of positivism
have constructed. Neurath believed it basic to the
movement for a unified science: ‘The fundamental thesis
of our movement is that terms similar to those employed
in physics and everyday language are sufficient for
constructing all sciences.’ 32 In defending this thesis,
Neurath expresses a more familiar positivist commitment
to the elimination of ‘metaphysical’ terms for unified
science. This physicalist version of the eliminative
project is less dogmatic than others: for example, it
makes no attempt to reduce either scientific or ordinary
language terms that have a physical reference to some
more basic observation language in the manner of earlier
positivists. It includes in the universal slang the
unreduced physical language of everyday folk. However,
it remains true that it does entail that terms without a
physicalist interpretation are to play no role in the
sciences. The familiar consequences of the doctrine are
apparent in Neurath’s discussion of the social sciences,
which advocates the elimination of intentional, ethical
and metaphysical terms: hence the abolition of terms
such as ‘ “existence”, “entity”, “reality”, “thing”, “fact”,
“concept”, “mind”, “mental world”, “physical world”,
“meaning”, “progress”, “the beautiful”, “the good” ‘.33
Hence also Neurath’s rejection of any interpretative
component in the social sciences.

The unified language version of the unified science
project is that which is most clearly positivistic in the
sense in which the term has been used in later accounts
of the movement. It is also, I believe, indefensible. It is
not insignificant that Neurath continually uses in his
papers vocabulary that fails his own physicalist
sanitization programme. To state and defend that project
he requires the use of such terms. As I noted at the outset,
it is not my purpose in this paper to rescue positivism nor
to defend its physicalist offspring. Moreover, while
Neurath himself took the eliminative project to be at the
core of the unified science programme, and while it might
form a (misguided) route to the orchestration of the
sciences, it is not a necessary condition for the project of
orchestration. Just as the programme of orchestration
does not require a unified method, nor does it require a
sanitized language.

The aim of orchestrating the sciences was the most
important but least discussed component of the
programme for unified science. The intent of the other
projects was the realization of the coordination of
different disciplines.

The purpose of this work [the International


Encyclopedia of Unified Science] is to explore the
foundations of the various sciences and to aid the
integration of scientific knowledge. The universe
does not follow the division of departments of a

The aim was an encyclopedia in which all the different
sciences would be coordinated and incompatibilities
addressed, a project that represents a modern form of the
Enlightenment’s encyclopedic ambitions. The problem
that it addresses is the way that questions about particular
states of affairs draw on different sciences. This problem
is central to any possibility of social planning that calls
on a variety of forms of knowledge. Moreover, the
raising of this problem need not have any scientistic or
technocratic consequences. Rather, it is important in
critically addressing the limits of the authority of

Consider, for example, the biochemist who claims
that, since all biological processes are ultimately
chemical, there can be no difference between the use of
artificial and natural chemicals in farming, nor between
inorganic and organic agriculture. The judgement
appears to be one that the biochemist is able to make – it
calls on knowledge of his field – and it is true that such
knowledge is relevant to the merits of different forms of
agriculture. However, the judgement he makes about
their respective merits calls on fields beyond his
authority. The abstract and general principles of
biochemistry cannot of themselves deliver the more
specific knowledge required to answer questions about
different kinds of agriculture. It fails to allow that
judgements about particular kinds of agriculture need to
appeal to other disciplines – to biology and ecology, for
example. It also needs to call on judgements which are
not about the soil at all, but about the institutional and
social context in which agriculture takes place. The
introduction of fertilizers has economic and social
implications on which no natural science would provide
judgement. A fault of purely ‘technical’ solutions to
economic problems in the past has been a kind of
‘technical utopianism’

which ignores
considerations. The programme of orchestrating the
sciences, at least in the hands of Neurath, aimed at
resolving just such problems.

The programme of orchestration is independent of the
other projects that typically are taken to define the unityof-science project. The version of orchestration that
Neurath defends is explicitly distanced from the
pyramidical programme of reducing social to biological
to physical science. Moreover, as noted above, it is
compatible with pluralism not just at the level of method,
but also, against Neurath, at the level of languages. A



physicalist language is not required to realize the
coordination of disciplines that Neurath defends. The
orchestration of human and physical sciences, for
example, does not require elimination of the intentional
language required to characterize human institutions

The project of orchestrated science is neither
technocratic nor scientistic as such. Rather, it highlights
the limits of the authoritative judgements of any
particular ‘expert’ in a single discipline about particular
matters. However, it could not in itself solve the
problems of moving from universal scientific principle
to particular applications. Not all knowledge can be
articulated in encyclopedic form; and, even in the case of
a unified body of articulated knowledge, there is no
reason to suppose that it can deliver ‘authoritative’

judgements on any particular case – some knowledge is
practical knowledge embodied in skills and know-how
that cannot be articulated in propositional form. For these
reasons it fails to meet Hayek’s challenge to socialism.

The claim that not all knowledge can be articulated in
propositional form is the central assumption in Hayek’s
epistemic case against central planning. While Hayek
frames the argument in terms of the division of
knowledge in society, that does not form the key to his
argument. It is rather the dispersal through society of that
local know ledge that cannot be articulated or vocalized,
and hence necessarily could not form an item that €ould
be passed on to a central-planning body. Hence, any
attempt to centralize economic planning decisionmaking reduces the amount of knowledge that is
available. Hayek’s argument against the possibility of
complete centralized planning is sound and one need not
accept his positive argument for the market to accept it. 34
The positive argument depends on the assumption that
only the market can coordinate dispersed nonvocalizable
knowledge. This is false: even the centralized firm of
existing society must make use of local knowledge that
is distributed within the institutional. Indeed Hayek’ s
point is one that has been articulated within the history of
socialist planning as an argument for democratic and
decentralized decision-making and for a proper
appreciation of the limits of scientific expertise. 35
Individuals often have local practical knowledge
relevant to the application of general principles – workers
and peasants are sometimes quite properly sceptical
about the self-confidence of the advice offered by newly
trained university graduates; their own everyday
knowledge of the materials and soils they work with
often provides a useful corrective to scientific authority.

For similar reasons, parents are properly wary of the
latest manuals for child care; and teachers, nurses and

others in practical professions ought to hold a degree of
scepticism about the latest theoretical offerings of
academic disciplines that inform their practices. This is
not to say that theory has no role, but that not all
knowledge is theoretical and even the application of
abstract theoretical knowledge is concrete contexts
requires good practical judgement and needs to be
corrected by practical experience.

Neurath’s encyclopedic account of unified science
does not address these problems. Moreover, his
scientistic view of knowledge renders him blind to them.

And here another concession to later critics of positivism
must be made. While Neurath’s position recognizes that
there is no purely technical solution to economic choices,
his own earlier writings nevertheless have a distinct
technicist flavour. The orchestration of knowledge is
seen in terms of the knowledge of experts. Thus he writes
in his 1919 report to the Munich Workers’ Council:

Socialization … is to be regarded as a trend
towards technicism. Engineers, doctors and
economists will have to collaborate and directly
use all achievements of technology, medicine, and
social organization, in order to further the
happiness of all. 36
Neurath’s early technicism is also expressed in his
reservations about the involvement of workers’ councils
in technical decisions.

A democratization of firms that goes so far that
technical direction is given by workers’ councils
and the administration of whole groups of firms by
boards of higher rank, entails from the social
engineering point of view a paralysis of
production. 37
Whatever the virtues of Neurath’ s encyclopedic version
of the unified thesis, it was blind to the dispersal of
practical knowledge and hostile to democratic
participation in ‘technical’ decision-making.

It needs to be added, however, that this strand of his
early work sits uneasily with his critique of a purely
technical account of decision-making. If, as he claims,
even the most technical of decisions involves nontechnical considerations, the rationale for leaving such
decisions to technical experts is considerably weakened.

This theme is taken up in his later work in the 1940s
which is more sensitive to the limitations of coordinated
scientific expertise and an acknowledgement of the
importance of democratic participation and
decentralization. Thus the following memorandum on a
visit of Neurath to the Borough of Bilston in 1945:

of administration…. With reference to the
decentralization of administration, Dr Neurath
stressed that participation is vital. The whole
success of any plan involving the lives of human
beings depends upon obtaining the assent,
encouragement, and co-operation of those human
beings. 38
This move in his later work to a more decentralized and
democratic image of the socialist commonwealth does in
part reflect an awareness of the dispersal of local
knowledge and the consequent limits of the competence
and authority of scientific experts. It is also rooted in his
own earlier critique of pseudorationalism, which, as
noted in the last section, denies that unique solutions to
practical problems can be arrived at by way of purely
technical procedures. Thus Neurath opposes ‘what is
called the “technocratic” movement’ which assumes
there exists ‘one best solution with its “optimum
happiness”, with its “optimum population”, with its
“optimum health”, with its “optimum working week”,
with its “optimum productivity” or something else of this
kind’ and which ‘asks for a particular authority which
should be exercised by technicians and other experts in
selecting “big plans”.’39 Against this scientific expert
Neurath appeals to common knowledge shared by all

Let us take an uncontroversial example. ~ssume
the scientists tell the English people that their
fireplaces waste calories – of course they do so
enormously. But fireplaces as an element of our
environment are not ‘happiness-neutral’ as it were,
as is e.g., the cable below the surface of the street.

Fireplaces are related to homely comfort. This and
other conditions of happiness would be the subject
of discussion and, finally, decisions would be taken
based on common sense and influenced by the
scientists’ information. 40
A decentralized and participatory politics is to be
preferred to the pseudorationalism of technocratic

Associational socialism, ecology
and freedom
The later Neurath’s advocacy of a decentralized and
participatory account of socialist planning had its source
not only in criticism of the technocratic account of

Dr Neurath stated that … within reasonable limits,

politics, but also in concerns about freedom and
ecological problems. In response to these Neurath was
led to defend an associational conception of socialism.

Given the revival of associational socialism,41 his

there must be the greatest possible decentralization

account deserves to be better known.


The existence of a variety of associations with power
and functions distributed amongst them is defended as
an institutional condition for freedom:

[T]he ‘freedom’ of a democratic country might be
described by the fact that each member is permitted
to have more than one loyalty, e.g. to his family, to
his local community, to his position, to his political
party, to his church, to his lodge, to an international
movement and to his country. One expects, in a
democratic country, that a citizen shows how to
handle these various loyalties and to assemble
them in one way or another. 42
The basis of dictatorial or totalitarian regimes lies in the
‘tendency for one, and only one loyalty to “devour” all
the others, and various loyalties are not permitted to grow
up side by side’ .43 Recently that familiar account of
totalitarianism which lies at the basis of the case for the
associational model of socialism has often been stated in
the language of civil society. What is significant about
Neurath’s version of the associational model is that it
remains strongly anti-market and makes a clear
distinction between the flourishing of associations and
the flourishing of exchange relations, a distinction that
many recent uses of the term ‘civil society’ have
blurred. 44 He avoids the assumption, which has dogged
much twentieth-century political thought and action, that
we must choose either state planning of the
‘internationalism of the “money-order'” .45
Neurath develops a picture of a socialist society as a
‘societas societum’ – a view in which economic life is
not governed by market principles, but in which ‘civil
society’ in the sense of thriving public association exists.

Thus he rejects the centralization of powers and functions
in the state in favour of dispersed overlapping planning
authorities. While this is independent of the guild
socialist model which forms the intellectual heritage of
recent associational socialism, it shares the appeal to the
structures, if not the content, of medieval Europe:

We know from the Middle Ages how
‘overlapping’ authorities can work. There could be
international organizations which would be
responsible for the administration of the main
natural resources, e.g. an organization dealing with
iron, others with coffee, rubber, foodstuffs which
could act as members of an international planning
board – such organizations could be in action
before a world commonwealth would be
organized. 46
Similarly, ‘big rivers with their banks could be
“internationalized'” .47 More local units of selfgovernment with powers of regional planning might exist


alongside such larger functional units.48
A significant feature of the international functional
units of planning that Neurath describes here is that they
are of the kind required if global resources are to be used
in an ecologically rational way, in particular to overcome
international ‘tragedy of the commons’ problems.

Neurath’s associational model of socialism has a clear
ecological dimension. Moreover, it is one that manages
to avoid the narrow localism of some Green thinkers 49
and the authoritarian statism of others.50 This ecological
dimension runs throughout his work. His sensitivity to
the issue is founded on his early familiarity with the
tradition of ecological economics developed early in this
century, which has been largely forgotten. 51 The work of
theorists like Ballod-Atlanticus and Popper-Lynkeus
attempted to base economic planning on the use of
energy units. As I noted in the first section, the claim that
one could rely entirely on energy units was one that
Neurath rejected. No single unit of evaluation existed to
order different alternatives. The use of the fireplace
example against the technocratic conception of politics
reflects the same critical attitude to any approach to
planning that relied solely on energy units. However, the
influence of these theorists on his work did mean that
Neurath had an awareness of problems of sustainable
development which has until recently been largely absent
from socialist work on economic planning. As
Bottomore notes in his survey of work on the socialist
economy, ‘Neurath’s conception of “ca1culation in kind”
… in principle enables economic planning to take into
account the use, as between generations, of nonrenewable natural resources (raw materials and
energy).52 Hence the discussion ofthe impact of different
uses of energy on future generations already quoted in
the first section.

This sensitivity to ecological issues was heightened
by Neurath’s commitment to the orchestration of the
sciences and his rejection of scientific pyramidism.

Against the standard forms of reductionism, in which the
social is reduced to more basic physical sciences,
Neurath argues that sociological knowledge about
human institutions needs to appear in full-blooded form
in the putatively prior physical science. One consequence
was an awareness of the relationships between human
institutions and the physical and biological environment.

We may say: ‘Men are connected with alterations
of geological structure like rain and rivers’, and
therefore we may get statements which speak of
correlations between alterations of human
institutions (connected with the construction of
dams, plowing etc.) and the alteration of the
surface of the earth and the climate. This implies

that sociological statements enter the geological
and perhaps also the astronomical departments in
full dress. Let me anticipatively say that difficulties
in making predictions on human institutions
therefore enter the geological sphere, which does
not remain watertightly separated from
N eurath’ s ecologically informed account of the nature of
the functional units for economic planning just noted
reflects this appreciation of the relationship between
human institutions and environmental change.

The central components of Neurath’s social and
political thought have a relevance today at least as
significant as when they were originally written. The
collapse of the dictatorships of Eastern Europe has
highlighted the need to escape statist models of
socialism. Neurath’s associational model of socialism
avoids statism without making the move, which has
unfortunately become standard in recent socialist
thought, of simply embracing the market. More than any
other socialist theorist of this century, Neurath offers an
account of socialism that is sensitive to ecological
problems. He offers a vision of an ecologically rational
society that allows for the representation of the interests
of future generations in current decisions and that offers
the basis for resistance to the attempt to resolve
environmental problems by putting prices on
environmental goods and harms. Given the particular
relevance of Neurath’s work for the contemporary
problems facing socialists and Greens, it deserves more
attention than it has received. It would be a tragedy if the
positivist label attached to Neurath should mean that his
contributions to socialist theory were to be forgotten. 54

Bibliography of the works of Otto
(1912) ‘The Problem of the Pleasure Maximum’, in (1973).

(1913) ‘The Lost Wanderers of Descartes and the Auxiliary
Motive’, in (1983).

(1919) ‘Through War Economy to Economy in Kind’, in

(1928) ‘Personal Life and Class Struggle’, in (1973).

(1935a) ‘The Unity of Science as a Task’, in (1983).

(1935b) ‘Pseudorationalism of Falsification’, in (1983).

(1936a) ‘Individual Sciences, Unified Science, Pseudorationalism’,
in (1983)
(1936b) ‘An International Encyclopedia of Unified Science’,
in (1983).

(1936c) ‘Encyclopedia as “Model'” , in (1983).

(1937a) ‘Unified Science and its Encyclopedia’, in (1983).

(1937b) ‘The New Encyclopedia of Scientific Empiricism’, in

(1937c) ‘The Departmentalization of Unified Science’, in

(1938) ‘Unified Science as Encyclopedic Integration’, in
Neurath, O. et al. (1938) Encyclopedia and Unified Science,
vol. 1, no. 1, Chicago, Chicago University Press.

(1939) ‘The Social Sciences and Unified Science’, in (1983).

(1941) ‘Universal Jargon and Terminology Science’, in (1983).

(1942) ‘International Planning for Freedom’, in (1973).

(1944) Foundations of the Social Sciences, Chicago, University
of Chicago Press.

(1946),The Orchestration of the Sciences by
Encyclopedism of Logical Empiricism’, in (1983).


(1973) Empiricism and Sociology, Dordrecht, Reidel.

(1983) Philosophical Papers 1913-1946, edited by R. S. Cohen
and M. Neurath, Dordrecht, Reidel.

‘Memorandum of the Visit by Dr Otto Neurath to the Borough
of Bilston; 24 July 1945’, in (1973).

1. The two quotations I have in mind are:

We are like sailors who must rebuild their ship on the
open sea, never able to dismantle it in dry-dock and
reconstruct it there out of the best materials.

Empathy, understanding, and the like may help the
researcher, but it enters into a system of science as little
as does a good cup of coffee, which helped the
researcher do his work.

The first quotation has even made it into the Oxford
Dictionary of Quotations; the second appears in a number
of standard discussions of interpretation in social science
(see, for example, R. Keat and 1. Urry, Social Theory as
Science, London, Routledge and Kegan Paul, 1982, p.


2. See, for example, H. Marcuse, One Dimensional Man,
London, Routledge and Kegan Paul, 1964; and J.

Habermas, Knowledge and Human 1nterests, London,
Heinemann, 1972. For a similar view from a very different
tradition, see F. Hayek, Counter Revolution in Science,
Indianapolis, Liberty Press, 1979. An account of the
political radicalism of the positivists is to be found in J.

Ravetz, ‘Ideological Commitments in the Philosophy of
Science’, RP 37,1984, pp. 5-10.

3. In the English-speaking world the reduction of the
movement to a set of simple doctrines is in part the
responsibility of A. 1. Ayers’ account of the doctrine in
Language, Truth and Logic (London, Gollancz, 1936).

4. Neurath, 1946, p. 235.

5. L. von Mises, ‘Die Wirtschaftrechung im Sozialistischen
Gemeinwesen’, Archiv fur Sozialwissenschaften, 47,
1920, translated as ‘Economic Calculation in the Socialist
Commonwealth’, in F. Hayek, ed., Collectivist Economic
Planning, London, Routledge and Kegan Paul, 1935.

6. L. von Mises, Die Gemeinwirtschaft: Untersuchungen
uber den Sozialismus, Jena, Gustav Fischer, 1922. A later
edition was translated as Socialism: An Economic and
Sociological Analysis (Indianapolis, Liberty Press, 1981).

7. F. Hayek, ‘The Uses of Knowledge in Society’ and
‘Economics and Knowledge’, in Individualism and
Economic Order, London, Routledge and Kegan Paul,

8. O. Lange and F. Taylor, On the Economics of Socialism,
edited by B. Lippincott, New York, McGraw Hill, 1956.

9. See Hayek, ed., Collectivist Economic Planning and
Lange and Taylor, On the Economics of Socialism. For
different appraisals of the outcome of the debate, see A.

Buchanan, Ethics, Efficiency and the Market, Oxford,
Clarendon Press, 1985, ch. 4; D. Lavoie, Rivalry and
Central Planning: The Socialist Calculation Debate



Reconsidered, Cambridge, Cambridge University Press,
1985; and D. Shapiro, ‘Reviving the Socialist Calculation
Debate: A Defense of Hayek against Lange’, Social
Philosophy and Policy, 6, 1989, pp. 139-59. D. Steele,
From Marx to Mises: Post-Capitalist Society and the
Challenge of Economic Calculation, La Salle, Open
Court, 1992, and R. Blackburn, ‘Fin de Siecle: Socialism
after the Crash’, New Left Review 185, 1991, pp. 5-67
both show more awareness of the earlier stages of the
debate, although with different appraisals.


10. L. von Mises, Socialism, translated by 1. Kahane,
Indianapolis, Liberty Press, 1981, p. 98. The part ofMises’

text from which I quote here is a reproduction of his first
essay in the socialist calculation debate, ‘Economic
Calculation in the Socialist Commonwealth’ of 1920. A
translation of this essay by S. Adler appears in Hayek, ed.,
Collectivist Economic Planning.

11. Ibid., p. 98.

12. Ibid., p. 99.

13. Neurath’s work also formed the main target of Weber’s
contribution to the prehistory of the socialist calculation
debate. See the extended discussion of Neurath’ s work in
M. Weber, Economy and Society, Berkeley, University of
California Press, 1978, ch. 2, section 12, pp. 100-107.

Weber’s argument is more careful than that of Mises. He
argues that some ‘value indicators’ (plural) must take the
place of prices for rational planning, and he expresses
some doubt as to what they might be. The objection is
weaker than Mises’ in the sense that it does not rule out
the possibility of such indicators. However, it has more
clout in that, unlike Mises, he makes no simple
commensurability assumption.

14. Neurath, 1919, p. 145.

15. Neurath, 1919, p. 146.

16. Neurath, 1928, p. 263.

17. See L. von Mises, Human Action, London, William
Hodge, 1949,p.209.

18. Neurath, 1913, p. 4.

19. Neurath, 1913, p. 8.

20. Neurath, 1935b, p. 125.






1935b, p. 131.

1935b, pp. 122-3.

1946, p. 232.

1912, p. 119.

25. To appeal to the necessary role for practical judgements in
decision-making is not to deny any role for general
principles or technical rules. Neurath does not make either
of these claims. One of the mistakes of defenders of
practical judgement is to set up an opposition between
moral and aesthetic judgement and the ‘technical’ rulegoverned rationality of science. There is a necessary role
for rules of thumb, standard procedures, and institutional
arrangements that can be followed unreflectively and
which reduce the scope for explicit judgements comparing
different states of affairs. We cannot be exercising ethical
and political judgements in a reflective way all the time.

Rules and institutions can free us space and time for
reflective judgements where they matter most. Such rules
and institutions need, however, to be open themselves to
critical appraisal.

26. D. Pearce, A. Markaandya and E. Barbier, Blueprintfor a
Green Economy, London, Earthscan, 1989, p. 57. For
critical discussion of that assumption, see 1. O’Neill,
Ecology, Policy and Politics: Human Well-Being and the
Natural World, London, Routledge, 1993, ch. 7.

27. See Neurath, 1935a, 1935b, 1936a, 1936b, 1936c, 1937a,
1937b, 1937c, 1938, 1939, 1941, 1944, 1946.


28. For a classic account ofthis programme, see E. Nagel, The
Structure of Science, London, Routledge and Kegan Paul,

29. Neurath, 1946.

30. Neurath, 1944, p. 8.

31. Neurath, 1946, p. 232.

32. Neurath, 1937a, p. 176.

33. Neurath, 1944, p. 18. Compare the list in Neurath, 1941,
pp. 217-18.

34. For criticism of the epistemic argument for the market, see
1. O’Neill, ‘Markets, Socialism and Information: A
Reformulation of a Marxian Objection to the Market’,
Social Philosophy and Policy 6, 1989, pp. 201-10; and
‘Property in Science and the Market’, The Monist 73,
1990, pp. 601-20.

35. See, for example, P. Cardan, Modern Capitalism and
Revolution, 2nd edn, London, Solidarity Publications,
1974, p. 45. H. Wainwright, Arguments for a New Left
(Oxford, Blackwell, 1994) examines some of these
parallels with Hayek’s work.







1919, p. 147.

1919, p. 139.

1973, pp. 75-6.

1942, pp. 426-7.

1942, p. 247. Translation amended.

41. P. Hirst, ed., The Pluralist Theory of the State: Selected
Writings of G. D. H. Cole, 1. N. Figgis and H. 1. Laski,
London, Routledge, 1989; L. Martell, ‘New Ideas of
Socialism’, Economy and Society, 21,1992, pp. 152-73;
W. Streeck and P. Schmitter, ‘Community, Market, State
– and Associations’, in W. Streeck and P. Schmitter, eds,
Private Interest Government? Beyond Markets and State,
London, Sage, 1985; S. Yeo, ‘Three Socialisms: Statism,
Collectivism, Associationalism’, in W. Outhwaite and N.

Mulkay, eds, Social Theory and Social Criticism: Es~ays
for Tom Bottomore, Oxford, Blackwell, 1987.

42. Neurath, 1942, p. 429.

43. Neurath, 1942, p. 429.

44. I develop this point at greater length in 1. O’Neill, Ecology,
Policy and Politics: Human Well-Being and the Natural
World, London, Routledge, 1993, ch. 10; and ‘Economy,
Polity, Neutrality’, Political Studies, forthcoming.

45. Neurath, 1942, p. 434.

46. Neurath, 1942, p. 433.

47. Neurath, 1942, p. 434.

48. Neurath, 1942, p. 435.

49. K. Sale, Dwellers in the Land: The Bioregional Vision,
San Francisco, Sierra Club, 1985.

50. R. Heilbroner, An Inquiry into the Human Prospect,
London, Calder and Boyars, 1975, especially ch. 4; and
W. Ophuls, ‘The Politics of a Sustainable Society’, in D.

Pirages, ed., The Sustainable Society, New York, Praeger,

51. Martinez-Alier, Ecological Economics (Oxford,
Blackwell, 1987) does an excellent job in recovering that

52. T. Bottomore, The Socialist Economy, New York,
Harvester Wheatsheaf, 1990, p. 26.

53. Neurath, 1944, p. 8.

54. My thanks to Chris Arthur, Russell Keat and the editorial
board of Radical Philosophy for their comments on an
earlier version of this paper.

Download the PDFBuy the latest issue