Is Re-modernization Occurring ­
And If So, How to Prove It?
A Commentary on Ulrich Beck
Bruno Latour
I
HAVE been asked to test the project of reflexive modernization in the
light of the anthropology of science and technology my colleagues and I
have been developing over the years ­ and which bears the rather silly
name of actor-network theory or ANT (Law and Hassard, 1999). I take this
not only as an honour but also as a great opportunity since the social theory
advocated by Ulrich Beck and his assistants is one of the most lively,
creative and politically relevant forms of sociology developed in recent years
­ the only one, at any rate, which dares to present itself as a general theory,
going all the way from the organization of the firm to the `normal chaos of
love', through the present conditions of scientific research and the turmoil
of ecological crisis, without missing the locus classicus of state, sovereignty
and nationalism, and so on, renewing, one after the other, all the chapters
of sociology textbooks.
I will use the term ANT (actor-network theory) as if it was a real branch of
social theory. This is largely an exaggeration, but indispensable in order to
carry out my little acid test. To put it simply, ANT is an argument not about
the `social' but about the associations which allow connections to be made
between non-social elements. ANT started with research into the history
and sociology of science, tried first to provide a `social' explanation of scien-
tific facts, failed to do so, and then, from this failure, it drew the conclusion
that it was the project of a social explanation of anything that was itself
wanting (Latour, 1999b). It has since, from this new vantage point, pro-
liferated somewhat erratically in many fields from management studies to
religion, from economics to law.
Theory, Culture & Society 2003 (SAGE, London, Thousand Oaks and New Delhi),
Vol. 20(2): 35­48
[0263-2764(200304)20:2;35­48;032616]
The first point of contact with ANT and what I will call from now on
re-modernization (to avoid the misleading adjective `reflexive' and keep
some of the meaning of `second' modernity) is to be found in the very notion
of risk (Beck, 1992). From this early work on, Beck's notion of risk has been,
in my view, misinterpreted, and so has his notion of what `reflexive' means.
Let me first clarify this difficulty by translating those two concepts in terms
which are ANT-compatible. `Risk' does not mean that people are nowadays
leading a more dangerous life than before ­ this would fly in the face of all
the life expectancy tables and would be an insult to the masses still dying
in the throes of typically modern engineered famines. And of course
`reflexive' is not to be taken for `reflexivity', as if people in the time of re-
modernization were more `aware' or `conscious' than at earlier periods. Quite
the opposite: `reflexive' means, in my reading of it, that the unintended
consequences of actions reverberate throughout the whole of society in such
a way that they have become intractable. Thus, `reflexive' does not signal
an increase in mastery and consciousness, but only a heightened awareness
that mastery is impossible and that control over actions is now seen as a
complete modernist fiction. In second modernity, we become conscious that
consciousness does not mean full control.
As to `risk', it does not mean that we run more dangers than before,
but that we are now entangled, whereas the modernist dream was to disen-
tangle us from the morass of the past. A perfect translation of `risk' is the
word network in the ANT sense, referring to whatever deviates from the
straight path of reason and of control to trace a labyrinth, a maze of unex-
pected associations between heterogeneous elements, each of which acts as
a mediator and no longer as a mere compliant intermediary. For instance,
you begin with a T-bone steak on your plate and you end up in the labora-
tory of a protein specialist showing you the tertiary structure of the now
infamous prion, one of the possible causes of the so-called `mad cow'
disease. But in the mean time you have visited European Commission
bureaucracies, the cattle farmers' unions, quite a few hospitals, and partici-
pated in a lot of scientific meetings (Barry, 2001). In brief, you have traced
a network ­ a network, to prevent any objection from people not familiar
with our use of the word, being not a thing in the world but the path traced
by the researcher equipped with an ANT methodology during his explo-
rations (Latour, in preparation).
I take the notions of risk or network as two similar ways of exploring
the new type of complexities typical of late-, second or non-modernity once
the black boxes of science and technology, so typical of the first moderni-
ties (I maintain a plural here for reasons developed below), have began to
leak hopelessly (Latour, 1999a). To put it differently, objects have become
quasi-objects, bald facts have become dishevelled objects, or, to go back to
the old German terms, dings have again become things, that is, complicated
cases ­ causa in Latin ­ in need of some sort of judiciary or political
assemblies, and no longer of any use for settling disputes or closing down
36 Theory, Culture & Society 20(2)
discussions (Thomas, 1980). Or, to use still other terms, matters of fact have
become states of affairs.
A striking example of the shift from objects to quasi-objects, from
matters of fact to states of affairs was provided to me while a storm, itself
possibly due to global warming, delayed my flight from Paris to Munich by
a day ­ while my luggage began tracing one of those unexpected dances so
typical of `risk society'. While waiting, I read in the International Herald
Tribune (30 October 2000) the following incredible piece of news. Judges
in the Microsoft anti-trust case, who had requested a crash course in the
basics of computer science, were denied such a briefing because of the
impossibility of defining the computer as a black box:
Among the many problems that the bitterly divided antitrust combatants
agreed on was the likelihood that Mr Hite's briefing [the `teacher' in the
course] would stray into issues in dispute. For example, Mr Hite said he had
planned to discuss the function of operating systems. But the very question
­ What is an operating system? ­ was the focus of extensive testimony during
the trial.
William Kovacik, an antitrust professor at George Washington University who
followed the trial, said the idea of a briefing was a good one, but that judges
had `miscalculated the extent to which there are bitter disputes over the tech-
nology itself'.
`As we have seen', Mr Kovacik said, `the minute he would begin to define the
term "browser", there would have been a fistfight in the courtroom.'
That's it! It is done and the question settled: we are no longer in the modern
world when even the computer, that absolute icon of control and calculation,
cannot be defined impartially in front of judges without generating fistfights
in the courtroom! Modernists have certainly `miscalculated the extent to
which there are bitter disputes over the technology itself'. We are entangled
for good, which, of course, great novelists had realized all along (Powers,
2000).
Another way of stating what is common to ANT and re-modernization,
is to use the economists' notion of externality. Externalities are produced
by any calculation as whatever cannot be taken into account (Callon, 1998).
Externalities can be positive or negative. Put quite simply, second
modernity is first modernity plus its externalities: everything that had been
externalized as irrelevant or impossible to calculate is back in ­ with a
vengeance. This is nowhere clearer than in the ecological crisis: there is no
longer any outside that can be considered as irrelevant ­ literally anything
has to be taken into consideration. There is still work needed to produce
the separation between inside and outside ­ for instance to renew economic
calculations, to absorb new values and exclude others ­ but this work is
now visible as one type of work among a lot of others, not as what defines
Latour ­ Is Re-modernization Occurring? 37
the tasks at hand. Instead of the surfaces so typical of first modernities ­
the `domains' of science, of economy, of society, the `spheres' of politics,
values, norms, the `fields' of symbolic capital, the separate and intercon-
nected `systems' so familiar to readers of Luhmann, where homogeneity and
control could be calmly considered ­ we are now faced with the rather
horrible melting pots so vividly described by historians and sociologists of
science (Haraway, 1997). But contrary to what social scientists believe,
although rather monstrous at first sight, those melting pots are neither
beyond description nor beyond political action: they just require other defi-
nitions of what empirical enquiries and representations mean. One could
say that, 70 years further on, we are led finally to what John Dewey focused
on: social sciences have to create the public, defined by him as what cannot
be mastered by anyone but that can be represented, over and over again, by
the social sciences and the humanities (Dewey, 1954 [1927]). The prefix
`re' in re-modernization takes on the same meaning as the one in re-
presentation. This is where the contrast between the methods and expec-
tations of ANT and reflexive modernization should be most interesting. How
can the public, in Dewey's sense, be generated anew?
Empirical enquiries are of course dependent on the theory one is trying to
prove. This is all the more true for such a wide-ranging theory as that of re-
modernization. Many objectors have argued against Beck and his colleagues
that side-effects are just that: side-effects, and that the grand project of
modernity will soon take care of them to resume its straight march forward.
It is very difficult to prove that side-effects are so numerous that their pro-
liferation has eaten up projects to the point where the projects themselves
have been transformed beyond recognition. What will convince a die-hard
modernist that the first modernity has ended?
My own solution ­ not shared, I hasten to say, by anyone in ANT ­ is
even more strange since I have argued that `we have never been modern',
claiming that there never was a time when modernity's interpretation of itself
described its deeds adequately. I have even argued that this discrepancy
between self-representation and practice was not due to the banal distance
between words and things, theory and practice, consciousness and life, but
that it was fully functional: it was because the moderns took themselves to
be so thoroughly disentangled from the shackles of the past that they were
so efficient at entangling themselves, and all the other nations, with every-
thing on earth and beyond, in ways that other civilizations could never have
dared to do. It is only if you are absolutely convinced that science and
society do not mix that you can mix them so thoroughly as to produce the
mess in which we are stewing today. Thus, for me, the originality of the
moderns never relied on some ideal of science, some definition of freedom
or the individual, some peculiar geographic or civic or legal infrastructure,
some religion or philosophy, but in one very efficacious estrangement from
their own practice which allowed them to do the exact opposite of what they
were saying. While all the other civilizations we know were ­ with great
38 Theory, Culture & Society 20(2)
care and caution, or at least with great misgivings ­ connecting the state of
their societies and the state of the cosmos, those who declared themselves
moderns could simultaneously affirm that there was no connection whatso-
ever between cosmos and society, and in the same breath begin to experi-
ment on a scale hitherto impossible with new and frightfully dangerous
connections between cosmos and society ­ the atomic bomb being, of
course, the culmination of this strange mixture before global warming took
over as the ultimate learning experience of a really explicit mix between
cosmos and society. Is a modern person someone who says `I am just doing
science' ­ and who produces, as a sort of aside, a mere afterthought, a
completely new society . . .?
This difference in approach between my version of ANT and re-
modernization has important consequences for the empirical tests of the two
theories one might expect. Beck and his group work on the assumption that
there has been something like a first modernity, which ­ 30, 20, 10 years
ago ­ began to give way to a second modernity with very different charac-
teristics (Beck et al., 1994). For my part, I only have to show that some-
thing has changed in the relation between what the moderns had been doing
without saying and what ­ for the last 10, 20 or 30 years ­ they now are
explicitly saying. To put it too simply, for Beck and his group the proofs have
to be in the substance of the phenomena they study, for me only in the collec-
tive interpretation given to phenomena which, all along, have never been
modern.
This difference is especially clear in the history of facts and machines
science students have developed over the years: one after another the
modernist heroes of science have been turned into thoroughly non-modern
entanglers. They have been, so to speak, anthropologized. This is true of my
Pasteur as well of Steven Shapin's Boyle (Shapin, 1994), Simon Schaffer's
Newton, Norton Wise's Kelvin, Peter Galison's Einstein (Galison, 2000). For
each one of those heroic figures one can say that the modernist explanation
of the science they were doing at the time bears at best an uncertain relation
to what they were doing. And this is not due simply to the usual revisions
of historical interpretations: Newton was at once purifying his science and
linking it to the whole cosmos, from the alloys of golden pounds at the Mint
he was directing to the inner thoughts of God (Schaffer, 1997). And I add
again: it is because of this distance between their gesture of purification and
their practice of entanglement that they were such good and creative scien-
tists, such effective modernizers. So, for me, historians of science have retro-
spectively constructed a non-modern history of Western science, but the
same could be said of all the other symbols of the modern period from
markets (Polanyi, 1945) to technologies, to what is called modern art (Clark,
1999).
This difference in the anthropological perspective introduces a nuance
in the empirical test one should look out for. The danger for Beck and his
group is that there are so many contradictory definitions of modernities that
it would be almost impossible to distinguish the traits that could be said to
Latour ­ Is Re-modernization Occurring? 39
be typical of `second modernity'. Objectors could always say: `Yes, but this
new definition of individualism, science, company, labour, etc. was already
present in the 18th, 19th or early 20th century.' It spoils Beck's argument
but it feeds mine since my point is that modernity has always been an
interpretation thrown on to an anthropological puzzle triggered by an
entirely different dynamic, much as Noah's son threw his coat to avoid
seeing the stark nakedness of his father. On the other hand, where my test
fails is when the shift from implicit to explicit makes no difference at all.
This is the acid test for my theory: since I argue that the incredible freedom
and creativity of the moderns was due to nothing but the ability for their
right hand to ignore what their left hand was doing: when this connection
becomes visible, they should lose freedom and creativity. When all the exter-
nalities are brought visibly back in, when the side-effects are forever
attached to the projects, when unwanted consequences proliferate even
before the causes are triggered, then the life of the moderns should become
miserable, brutal and short. Disentangling has become impossible and thus
the feeling of time will be entirely different ­ even though the moderns are
simply falling back on `business' or at least `anthropology as usual'.
Another proviso should be introduced before we can begin to envision
the right tests for the two parallel arguments. This is more tricky since here
the gap between Beck's enterprise and ANT is much wider. To put it simply,
ANT is a direct descendant of Garfinkel's ethnomethodology. One could say
that it is a hybridization of Garfinkel for humans and Greimas for non-
humans ­ in fact, I discovered recently that it is an illegitimate grandchild
of Gabriel Tarde (Tarde, 1999). But nothing could be further from Beck's
theory than ethnomethodology since he remains, on that ground, an unre-
constructed social theorist, roaming freely through the `whole' of society
without showing the least interest in the practical and local conditions
making this `whole' visible (Latour and Hermant, 1998). This introduces a
greater difference between the two theories: in a Tarde­Garfinkel­ANT sort
of framework, any sociology, no matter how powerful, is just one
ethnomethod among others, competing on the ground with all the other
ethnomethods, most of them more powerful, produced by the `actors them-
selves'. It does not mean that social theories are not useful, respectable and
accurate; it just means that they add their influence, detours, interpretations
to the plot which they can in no way explain or replace. Beck, as far as I
can judge from his own writing, does not balk at placing himself firmly in
the God's-eye privileged position occupied before him by Durkheim or
Weber or Bourdieu. His interpretation of what the actors are doing aims at
explaining what society is and how it functions. On his own account, it is
not one of many `master narratives' competing in the daily intricacies of
practices for accountability with many other interpretations generated by
the actors. It is an accurate description of what is happening out there.
This is not a defect, of course, since social theory in this classical form
has always occupied such a quasi-judicial position. But it means that the
test for proving the development of re-modernization will be rather different:
40 Theory, Culture & Society 20(2)
for me, one can test whether or not the theme of re-modernization, as it is
articulated in the general press, in academia, in bar conversations, in the
bedrooms, at 10 Downing Street between Mr Blair and Mr Giddens, does
actually modify the interpretations of what people are doing compared to,
say, a Bourdieu-type definition of what late symbolic capitalism is. Actors
equipped with different master narratives of what modernity or second
modernity is will indeed take different tacks, make different choices, will
present themselves differently. But those shifts will be tiny inflections which
have to be considered against a background of practices of an entirely
different nature ­ which only an anthropological enquiry can retrace.
For Beck, however, the proof of the existence of re-modernization is
not only a change in master narrative but a real modification of the stuff out
of which practice itself is produced. Here we find again the distinction
between a change in substance (that Beck favours) and a change in interpre-
tation (that seems to me much more plausible and, above all, provable). I
have no special sympathy for the notion of master narratives but neither
have I any reason to reject them since in the social theory I work with, there
is no other way to build a society than thrashing out, through a sort of lasso-
like movement, constantly new interpretations of what gathers us together.
This is the only way to build Dewey's public, since society does not exist
as a sui generis entity but as what has to be locally achieved. This is the
Garfinkelian watershed which, it seems to me, it is difficult to ignore. But
for those like Beck who believe there is a society out there that can be
described, if not completely objectively at least with a great degree of confi-
dence that does not depend on the actors' own reflexive achievement, this
reduction of re-modernization to a new master narrative might seem unfair
­ or even offensive. Yet my overall enthusiasm for Beck's project cannot
balance my feeling that there exists a great difference between BG and AG
sociologies (meaning of course `before' or `after' Garfinkel!).
Once the parallels and the discrepancies between the two arguments have
been sketched, it is possible to work out some of the common sites where
the phenomena of re-modernization can best be observed ­ and observed
in such a way that they could count in the eyes of the vast majority of social
scientists who still believe modernity is the unsurpassable horizon of our
times, that the aim of the social science is still to emancipate actors from
their obscurantist past. I said `count' rather than `convince' since sociolo-
gists of science know better than anyone else that to convince a colleague
one needs to change worlds rather than just accumulating proofs, since each
of these can be reinterpreted or pushed aside. The social sciences have
always been connected to a political project, and of course this applies to
our social sciences too. The pragmatic test is to see whether talking in terms
of re-modernization does make a difference or not for the emergence of a
public assembled along different lines. But this is a bit premature since, for
the time being, we don't know if we are talking about a phenomenon that
has taken place or not! Maybe second modernization has never occurred ­
Latour ­ Is Re-modernization Occurring? 41
and maybe I am entirely wrong to say that we have never been modern. We
are a bit like the specialist in global warming 10 years ago, uncertain
whether they were warning about an impending disaster or registering
statistical flukes in the overall stable patterns of climates.
The key question is to move from anecdotal evidence to data. Another
pressing question is to differentiate our data from those that prove post-
rather than re-modernization. Beck is not very much interested in post-
modernity since he has no place in his own periodization between first and
second modernity. But for me, who considers sociologies to be more or less
creative interpretations thrown upon times that tick because of other
mechanisms, postmodernism is very important as a competitor to Beck's
sociology or my anthropology. Postmodernity has many definitions but it
offers, at its core, a theory of the way time flows and how connections are
established between heterogeneous phenomena. Since it relies so much on
what it claims to replace ­ I have defined it as `disappointed modernism' ­
it is close to many of the attitudes re-modernization is interested in tracking.
But since it remains so thoroughly modernist, its proliferation, if proven,
will always be at the expense of re-modernization, especially the latter's
positive and political efforts. Hence the crucial importance of making the
test fine enough to discriminate between the two interpretations of post- and
re-modernization.
We are now prepared to tackle the hard questions: supposing a scholar
is intent on checking whether or not something has changed in the way
modernity has progressed in the last 20 years, what should he or she look
for? And, if possible, what sort of quantum should be gathered? The quan-
titative dimension of social-scientific data cannot be escaped if one wishes
to avoid the essayistic or prophetic tone, or if one is not satisfied with case
studies.
Here are some of the items we might want to discuss and for which I
of course play the devil's advocate since my goal here ­ indeed my brief ­
is to provide an adequate test for our common undertaking.
Is `modern' becoming a negative adjective? What is missing is a back-
ground that could be used as a baseline to detect whether or not there has
been a shift in people's feeling about modernization. It would be tremen-
dously useful to have opinion polls of a very classical sort to see with what
other adjectives the adjective `modern' is associated in 2000, compared to,
say, 1990 and 1960. This could be done retrospectively through socio-
linguistic tools. And it would be fascinating to do cross-cultural compari-
sons among European nations. My hunch here is that modern has lost two
of its features: it no longer means something is incontrovertibly good and,
more importantly, it is no longer associated with a coherent set of values
and objects. For instance my neighbours in Bourbonnais call `traditional'
the `modernist' agriculture of the 1980s. But this is anecdotal and any study
of re-modernization requires a map of the usage of words it replaces.
Everyone, in France at least, is trying to fiddle around with the accusation
42 Theory, Culture & Society 20(2)
of `reactionary', except no one can agree any longer who is progressive, and
which attitudes are progressive . . .
Are objects slowly replaced by quasi-objects? Like Beck and his group
I have a strong feeling that the more we go on, the less we read about incon-
trovertible, bounded, mastered, black-boxed, expert-ruled innovations. The
little example above about the Microsoft trial is typical of many tell-tale
signs. But we are all strongly biased to detect precisely those events which
signal, for us, a shift away from hypes. Are we right in claiming that those
sparrows are indicating a change of season? One look at the Internet litera-
ture, for instance, or at the human genome flood of news, shows strong signs
of an unreconstructed, unrepentant hype promising exactly the same sort of
glorious tomorrows as in the 1950s with nuclear energy, or in the 1960s with
cybernetics, or in the 1970s with computers, and so on. How do we prove,
quantitatively, that objects are now presented in a more `risky' fashion, that
is, with their unwanted consequences and with their uncertain and puzzled
makers and users attached to them? Can we calculate in the general press
for instance the ratio between the number of hyped technological news and
the number of `risky' and cautionary tales? And check that between 1980
and today this ratio has been modified? Then, we would have a proof of a
sea change ­ provided of course we can operationalize the notion of risky
description (Rogers, 2000).
Is the difference between nature and society breaking down? Here
again, one familiar with controversies over national parks, rivers, water, air,
landscapes, etc. will be struck by the diminishing efficacy of the pure
nature/pure culture boundary. More exactly the hard labour of boundary
making between the two will become highly visible ­ as visible as the
building of a fence around a park to make it `wild' (Western et al., 1994).
And of course, there is no better sign of the dismantling of a distinction than
when it becomes visible as such ­ same thing with race, gender, science
and non-science, etc. But here again, are we not terribly biased in favour
of those signs, whereas modernists could very well argue that bringing the
peace and consensus of natural laws still remains the only safe way to obtain
a liveable political arena? Each example of what we consider a breakdown
of the nature/culture distinction could be taken by modernists as a
provisional hiccup. Sociolinguistic tools would be very useful here to settle
the matter by showing whether nature is used nowadays in conjunction with
words that indicate disputes, uncertainties and controversies. But has the
work been systematically done (Macnaghten and Urry, 1998)? Is a new trend
over the last decade or so really detectable?
Another problem for this indicator would be to differentiate it from the
postmodern ways of breaking boundaries by ironizing them without putting
them into a new positive political and institutional mould (Cronon, 1996).
How is the distinction made, for instance, between cyborgs ­ the complacent
eulogizing of techno and natural mixtures ­ and what we would call hybrids,
or risky, or dishevelled quasi-objects? What are the tell-tale signs that
Latour ­ Is Re-modernization Occurring? 43
the breaking of the nature/culture boundary is taken as a chance to assemble
`parliaments of things' ­ at the level of sub-politics or at the level of grand
politics? My own study of water is highly ambiguous on that score. A
systematic comparison of all the cases might shed some light on the trends
there.
Is time flowing differently in first, post- and re-modernization? Time's
directional flow is at once a very collective and a very intimate experience,
which can certainly be mapped out in some sort of detail since, after all, a
great deal of work has been done by historians to account for modernist
feelings about progress (Yack, 1992). The period after the fall of the Berlin
wall is certainly different from that of the Cold War, or the post-war, inter-
war or pre-1914 periods, or turn of the century, the 1848 revolutions, and
so on. But how are we to operationalize those fine differences? Yet, has the
feeling of progress really diminished? Is it not the only great narrative that
is proposed in countless political platforms and editorials? Can we prove
that the Great Leap Forward is no longer the way the future is presented to
us? Even of postmodernism ­ which is above all a different way of record-
ing the succession of epochs ­ can we empirically say that it has entirely
overcome the modernist ethos? Even in architecture, the birthplace of
postmodernism, we still get very ambiguous signals. But maybe it is this
very ambiguity that can be used as a handle to get good data on this
question. Maybe it is the case that there has been a detectable shift from a
time of succession to a time of coexistence, all the times ­ now in the plural
­ being simultaneously at work and represented without the past being abol-
ished. Hence the feeling of `reaction', `neo-archaism' and so on (Sahlins,
2000).
Are subjects becoming quasi-subjects? If the status of objects is being
deeply modified in re-modernization, there should be a concomitant change
in the status of subjects. It would be really odd if mastery was slowly being
eroded on the side of science and technology while a new subject was
regaining force and becoming even more master of itself. Odd but perfectly
possible; history moves in strange disjointed ways. Here again we need an
empirical way to test the two possibilities, and here again the difference
with postmodern subjects will be very difficult to detect. What Luc Boltan-
ski calls the `city by projects' or the `connexionist human' has both char-
acters: from the modern it has kept the call for unlimited freedom, but it is
thoroughly postmodern in recognizing no durable attachment except the
maximization of its own movement through connectionist space (Boltanski
and Chiapello, 1999). How do we differentiate this self-employed, self-
exploited, self-made `roi-moi', to use Pierre Legendre's expression, from a
re-modernized subject? The key difference should appear, quite logically,
in the relation between the former subjects and the quasi-objects now
offered to them. This is the great interest of the work on genetics (Strath-
ern, 1992), the brain, emotions, physiology and so on, because the destiny
of the subject is entirely different when seized by the urge for naturaliza-
tion ­ in the modernist way ­ and the attachments to `cycles of objectivity'
44 Theory, Culture & Society 20(2)
(Berg and Mol, 1998). Objectification, reification should appear, in our data,
no longer as the only way for a subject to `have a body'.
But how much of this is wishful thinking, in spite of the work done by
historians and sociologists of medicine and genetics? Can we really prove
that quasi-objectivation has indeed generated new kinds of attachments that
would look really different from modernist and postmodernist subjectivi-
ties? Again there is anecdotal evidence, but where are the masses of data?
It is also perfectly possible that re-modernization has touched the objects
deeply ­ because of the new doubts about science and technology ­ but not
yet the subjects, who are still carried forward by the great scenario of
modernism ­ postmodernism adding simply a new spin, a new acceleration
to the search for freedom. If this was the case, we would not find any proof
of re-modernization in the realm of subjectivities.
Is the economy shifting from infra- to supra-structure? Objects and
subjects in modernism were connected by a largely naturalized economic
infrastructure studied scientifically by economics. Normally, the shift to re-
modernization should be even more vivid in the domain of the relations
entertained by new quasi-subjects attached to quasi-objects. But this is
where the data are even more uncertain than with changes in subjectivities.
The last 20 years have seen an unprecedented re-naturalization of econ-
omics from the right ­ freed from the competition of socialism ­ and from
the left ­ deeply entrenched in economic reasoning (see Bourdieu, 2000
and his critic Favereau, 2001). Together with the theme of globalization and
neo-liberalism, it seems that all the indicators, far from showing a second
modernization, bear witness to an expansion of modernization with a
vengeance (Callon, 1998). Or else, re-modernization means nothing but `re-
engineering' and the advent of the `network society'.
Yet the work of socio-economists and anthropologists of goods and
accounting could be interpreted as registering a deep shift away from natu-
ralization towards a re-socialization of the economic infrastructure slowly
brought back to a superstructural level (Cochoy, 1999). Such is the crucial
distinction introduced by Callon between `economics' and `economy', the
latter being absorbed as the performative activity of the former (Miller,
1992). Marx walked on his head and is put back on his feet . . . But this is
an academic undertaking: does it prove that in the public consciousness,
the economy has undergone such a gigantic sea change? The indicators are
clearly not there. It does not prove that the change has not occurred, but
that the tell-tale signs will not count as incontrovertible data in the eyes of,
for instance, traditional economists.
To conclude this article and answer the question raised at the beginning of
this section, one can see that the data to be gathered in order to prove the
advent of a substantial phenomenon called re-modernization are not easy to
come by and, so far, are not thoroughly convincing. If it is easier to demon-
strate that things are changing fast for quasi-objects, there is as yet no clear
indication that subjects have stopped being moderns and become
Latour ­ Is Re-modernization Occurring? 45
quasi-subjects; still less are there clear signals that something has changed
in the economic connection between quasi-objects and (quasi-?) subjects.
This is where the difference between re-modernization as a master-
narrative and re-modernization as the accurate description of a substantial
sea change in contemporary societies should be highlighted again (I put it
aside momentarily to test the strength of various potential indicators). The
power of a social theory is different in both cases and should be evaluated
differently. Re-modernization might not describe what has already
happened, but it can offer a powerful lever to make new things happen. This
is where the difference between descriptive and normative theories breaks
down. It makes perfect sense, for me, to propose an interpretation of science,
subjectivities and industry that builds Dewey's public, even though the proof
is not all there. After all, this is exactly what the modernist thinkers, from
Rousseau to Weber, have always done. Where was the proof that rationaliz-
ation was occurring in Weber's time? It was just as ambiguous and discon-
nected as the proof for re-modernization today. To be sure, social scientists
are not demiurges, nor do they occupy a vanguard position, but it is also
their duty, in the situation of common ignorance so typical of public space,
to offer alternatives to earlier versions of the social link. On that score, re-
modernization is a powerful proposition because it shifts attention from the
mainstream ­ so visibly modern or postmodern ­ to the discrepancies,
cracks, failures and side-effects, and it claims that, with all those bizarre
and disconnected phenomena, barely visible to the majority, it can produce
a coherent picture of a European world which has outgrown progress.
Note
This article was initially a comment on the work done by Ulrich Beck and his group
in Munich, during a site visit he asked me to do in November 2000.
References
Barry, Andrew (2001) Political Machines: Governing a Technological Society.
London: Athlone Press.
Beck, Ulrich (1992) Risk Society: Towards a New Modernity. London: Sage.
Beck, Ulrich, Anthony Giddens et al. (1994) Reflexive Modernization: Politics,
Tradition and Aesthetics in the Modern Social Order. Stanford, CA: Stanford
University Press.
Berg, Marc and Anne-Marie Mol (1998) Differences in Medicine: Unraveling Prac-
tices, Techniques and Bodies. Durham, NC: Duke University Press.
Boltanski, Luc and Eve Chiapello (1999) Le Nouvel Esprit du capitalisme. Paris:
Gallimard.
Bourdieu, Pierre (2000) Les Structures sociales de l'économie. Paris: Le Seuil.
Callon, Michel (ed.) (1998) The Laws of the Market. Oxford: Blackwell.
Clark, T.J. (1999) Farewell to an Idea: Episodes from a History of Modernism. New
Haven, CT: Yale University Press.
Cochoy, Franck (1999) Une histoire du marketing: discipliner l'économie de marché.
Paris: La Découverte.
46 Theory, Culture & Society 20(2)
Cronon, William (ed.) (1996) Uncommon Ground: Rethinking the Human Place in
Nature. New York: Norton.
Dewey, John (1954 [1927]) The Public and Its Problems. Athens: Ohio University
Press.
Favereau, Olivier (2001) `L'économie du sociologue ou penser (l'orthodoxie)  partir
de Pierre Bourdieu', pp. 255­314 in Bernard Lahire (ed.) Le Travail sociologique
de Pierre Bourdieu: dettes et critiques, revised and expanded edn. Paris: La Décou-
verte.
Galison, Peter (2000) `Einstein's Clocks: The Place of Time', Critical Enquiry
Winter: 355­89.
Haraway, Donna (1997) Modest Witness@Second Millennium: The FemaleMan
Meets OncoMouseTM. London: Routledge.
Latour, Bruno (1999a) Pandora's Hope: Essays on the Reality of Science Studies.
Cambridge, MA: Harvard University Press.
Latour, Bruno (1999b) `When Things Strike Back: A Possible Contribution of
Science Studies to the Social Sciences', British Journal of Sociology 51(1): 105­23.
Latour, Bruno (in preparation) ANT: A Travel Guide. Oxford: Oxford University
Press.
Latour, Bruno and Emilie Hermant (1998) Paris, ville invisible. Paris: La
Découverte­Les Empcheurs de penser en rond.
Law, John and John Hassard (eds) (1999) Actor Network and After. Oxford: Black-
well.
Macnaghten, P. and J. Urry (1998) Contested Natures. London: Sage.
Miller, Peter (1992) `Accounting and Objectivity: The Invention of Calculating
Selves and Calculable Spaces', Annals of Scholarship 9(1­2): 61­86.
Polanyi, Karl (1945) The Great Transformation. Boston, MA: Beacon Press.
Powers, Richard (2000) Plowing the Dark. New York: Farrar, Straus and Giroux.
Rogers, Richard (ed.) (2000) Preferred Placement: Knowledge Politics on the Web.
Maastricht: Jan Van Eyck Editions.
Sahlins, Marshall (2000) Culture in Practice. New York: Zone Books.
Schaffer, Simon (1997) `Forgers and Authors in the Baroque Economy', paper
presented at the meeting `What is an Author?', Harvard University, March.
Shapin, Steven (1994) A Social History of Truth: Gentility, Civility and Science in
XVIIth-century England. Chicago, IL: University of Chicago Press.
Strathern, Marilyn (1992) After Nature: English Kinship in the Late 20th Century.
Cambridge: Cambridge University Press.
Tarde, Gabriel (1999) Monadologie et sociologie, réédn. Paris: Les Empcheurs de
penser en rond.
Thomas, Yan (1980) `Res, chose et patrimoine (note sur le rapport sujet­objet en
droit romain)', Archives de Philosophie du Droit 25: 413­26.
Western, D., R.M. Wright et al. (eds) (1994) Natural Connections: Perspectives in
Community-based Conservation. Washington, DC: Island Press.
Yack, Bernard (1992) The Longing for Total Revolution: Philosophic Sources of
Social Discontent from Rousseau to Marx and Nietzsche. Berkeley: University of
California Press.
Latour ­ Is Re-modernization Occurring? 47
Bruno Latour is professor at the École des Mines in Paris. He has
published extensively on science and society matters and also in social
theory. Complete references and most articles can be found at
<http://www.ensmp.fr/~latour/>.
48 Theory, Culture & Society 20(2)