Foucault on Governmentality and Population: The Impossible Discovery
Author(s): Bruce Curtis
Source: The Canadian Journal of Sociology / Cahiers canadiens de sociologie , Autumn,
2002, Vol. 27, No. 4 (Autumn, 2002), pp. 505-533
Published by: Canadian Journal of Sociology
Stable URL: https://www.jstor.org/stable/3341588
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Canadian Journal of Sociology is collaborating with JSTOR to digitize, preserve and extend
access to The Canadian Journal of Sociology / Cahiers canadiens de sociologie
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and
Population: The Impossible Discovery*
Bruce Curtis
Abstract: For Michel Foucault's analysis of state formation, the 'discovery of population' was the
pivot on which the transition from rule based on police to rule in liberal modes of government took
place. Moder 'governmentality' takes population as its object. Foucault's usages of 'population'
are reconstructed. It is claimed that political authorities cannot 'discover' population, for population
depends on the exercise of sovereign authority.
Resume: Michel Foucault propose que la 'decouverte de population' aurait ete le point tournant
d'une transition entre une logique politique axee sur le modele de police et une logique axee sur le
respect de 'la nature des choses' dans un mode de gouverement liberal. Nous retracons les usages
foucauldiens du concept de "population." Selon notre argument, il serait impossible pour des
autorit6s politiques de "d6couvrir" ce concept, car 1'objet "population" depend d'un exercice particulier
de 1'autorit6 politique.
In the late 1970s, Michel Foucault claimed that Karl Marx had been perfectly
correct to regard the history of past societies as the history of class struggle
(Foucault 1994 (1978)). However, contemporary Marxism had reduced this
proposition to the continually asked questions 'what's the class?,' 'where's the
class?,' 'what class do you belong to?.' And the French Communist Party confronted
all social struggles with the question, 'is this struggle in the interest of
theproletariat?.' Marxists askedrepeatedly 'what's the class?' butnever 'what's
the struggle?,' 'who is engaged in struggle?,' 'what is at stake in struggle?,'
Canadian Journal of Sociology/Cahiers canadiens de sociologie 27(4) 2(X)2
* Earlier versions of this paper were presented to the Foucault SIG at AERA, Montreal, April 1999
and at the Center for History, Society, and Culture, University of California-Davis, in October
2000. For critical commentary, I wish to thank Beradette Baker, Patrick Carroll-Burke, John
R. Hall, Alan Hunt, Frank Pearce, Lorna Weir, and the anonymous reviewers from this journal.
505
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
506 Canadian Journal of Sociology
'by what means is struggle conducted?' Marxist economism was thus incapable
of making sense of the myriad struggles characteristic of contemporary
political life. To come to grips with such struggles, political analysis had to
'get rid of Marxism' as a doctrine that controlled the political imagination.
The way to get rid of Marxism, Foucault argued, was not by adopting the
typical tactic of sociology - discounting Marx's empirical predictions - nor
by adopting the tactic of Althusser - extracting the 'one true Marx' through
textual analysis. Rather, one needed to displace the analysis of social class
through the use of other categories and other tactics. Foucault proposed that
political analysis listen to all those voices subjugated by the authority of the
Party. In modern liberal government, individual conduct and comportment
have become directly implicated in the operations of power. Government
focuses on the 'conduct of conduct,' and the state has been de-centred. The
totality of subjugated voices exists as the population of the governed. Their
struggles are those to which the French media refer as 'la grogne des
categories' [the grumbling of the categories].
This article traces the emergence and deployment of the concept 'population'
in Michel Foucault's work in the period from about 1970 to 1982. The
body of the piece offers a detailed textual exposition. It is preceded and followed
by some critical commentary. The concept 'population' was central to
Foucault's attempts to write an analysis of state formation adequate for
contemporary politics. 'Population,' he argued, is the pivot on which turned the
transition from rule based on sovereign authority to a 'governmentalized' rule
which decentres the state under liberalism. It allows us to think the shifting
coalitions that constitute the new social movements and the struggles of the
governed.
Foucault maintained that there have been two major revolutions in the
technologies of power since the classical age: the development of anatomopolitical
techniques aimed at the individual body and the development of biopolitical
techniques aimed at the collective or social body. Both sorts of techniques
emerge from engagement with 'population.' Rule in modern societies,
according to Foucault, is to be understood as triangulated around practices of
sovereignty-discipline-government and to have as its essential object the population.
In an attempt to divorce the analysis of political power from what seemed
to be reductionist preoccupations with the state and class relations, some
contributors to the early governmentality literature claimed that political power
is 'beyond the state' in liberalism, because the 'terrain par excellence' of
government is 'population' (Miller and Rose 1993).
Despite the evident importance of 'population' in Foucault's intellectual
project for much of the 1970s, the concept has escaped sustained critical
attention in the Foucauldian literature, just as it has in the related fields of the
analysis of risk and the history of statistics. It is true that Mitchell Dean
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 507
observed, in relation to Foucault's claims that there was an 18th century
population 'discovery,' that 'any attempt to read the [18th century] concept of
population as an index of the modernity of political discourses is deeply
problematic.' He continued that 'the concept of population in eighteenthcentury
thought on government is strikingly different from its classical liberal
(and more recent) uses. It entails neither the formulation of policies and
political action by reference to an explicitly economic rationality which is the
characteristic of liberal governance, nor the welfarist focus on the enhancement
of the life of 'individuals' (Dean 1991:33). Dean did not pursue such criticism
in his later work, adopting instead a more charitable view towards the evident
inconsistencies in Foucault's account of modern state formation (Dean
1994:180), but I will argue his observation was well-founded.
'Populousness,' the Social Body, and 'Population'
In following out the concept 'population,' my concern is not with the word
'population' as such. As one commentator on an earlier version of this article
pointed out, it would be an elementary misunderstanding of the approach of
'historical epistemology' which shaped Foucault's work to take a word for a
concept. In historical epistemology, one distinguishes words from concepts in
that the latter 'make a difference' in theoretical discourse. As Foucault put it
in his introduction to Canguilhem's The Normal and the Pathological, a
concept 'must give access to a structure of intelligibility' (Foucault 1989
(1978):19). The modern concept of population certainly does so; or, at least,
the construction of the concept population is central to the creation of new
orders of knowledge, new objects of intervention, new forms of subjectivity
and, I argue, new state forms.
However, the development of the concept 'population,' and more especially
its deployment in rationalities of government, is to be distinguished from an
earlier concept of 'populousness' and the work that it sustained both in
political theory and practice. Both concepts in turn are to be distinguished from
a concept midway between them: the 'social body' or the 'collective body.'
Both populousness and population can sustain analyses of the 'social body,'
although that body is configured differently by them. The distinction is primarily
one between an organic and an atomistic view. An exposition of Foucault's
development of 'population' is rendered complex for the reason that he
employs the word indifferently to refer to the three concepts in question. The
word 'population' is used by Foucault to refer to the concept of populousness,
in discussions of police and mercantilism, for instance. The word refers to the
collective or social body in discussions of bio-politics. It is used to refer to
what I argue is population, properly conceived, in discussions of bio-politics
and liberal modes of government.
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
508 Canadian Journal of Sociology
Populousness has a long and convoluted history in western thought. It points
to the sense that units of government (kingdoms, empires, countries, parishes,
cities) contain greater or lesser numbers of entities - hearths, soldiers or souls,
for instance - distributed across different orders or classes. Peter Biller shows
that populousness was present already in Aristotle's Politics, and was then
taken up in the canonical literature following the translation of the Politics into
Latin. Medieval 'demographic thinking' in the canonical literature of the
thirteenth century already contained relatively sophisticated discussion of
matters ranging from the mechanics of generation, through the obligation of
pastors to know the numbers of their flocks, to the relative size of different
orders of the people in the good polity (Biller 2000). And Christian thought
clearly contained a notion of the collective body, as the communion and the
body of Christ.
Again, Jean Blondin wrote of populousness in his 16th century argument that
census-making was an essential technique of government. Through it,
...first of all, as to the people, one would now the age and the quality; and how much one could
derive from them, be it for warfare or domestic purposes, or to send them as colonists or to employ
them in the works and obligations of public maintenance and fortifications, to know the ordinary
provisions and means of livelihood necessary for the inhabitants of each city: principally in case of
surviving an enemy siege, which it is impossible to anticipate if one does not know the number of
subjects (quoted in Dupaquier and Dupaquier 1985:75-76).
The city is a collective whose parts are not equivalent, but whose survival is
dependent on relations among them. The concept of populousness makes it
possible to plan for the future and, in mercantilist doctrine, the degree of
populousness is an index of wealth and a measure of policy.
As well, Mary Poovey argues, populousness is the concept at work in many
parts of Malthus' Essay on Population, where generation depends upon the
comportment of one order of society - men of marriageable age (Poovey
1998: 286-287). Populousness commonly involves a concern with 'numbering
the people' (see Glass 1973), but because the concept implies hierarchical
differentiation of orders of the people, it does not do the work that 'population'
does. Specifically, it does not involve a substantial development of the work
of abstraction whereby regularities emerge through the creation of relations of
equivalence among heterogeneous entities.
'Population' is dependent, in the first instance, on the establishment of
practical equivalences among subjects, objects or events. In contrast to
populousness, whose logic centres on the hierarchical differentiation of
essences (knights fight, priests pray, peasants till), population depends upon the
notion of a common abstract essence. At the outer limit of abstraction,
population consists of so many undifferentiated atoms distributed through
abstract space and time (Curtis 2001: chpt. 1). As an object of knowledge,
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 509
population is primarily a statistical artefact. The establishment of practical
equivalences means that population is connected to the law of large numbers,
which causes individual variation to disappear in favour of regularity. In its
developed forms, population is bound up with the calculus of probabilities.
Population makes it possible to identify regularities, to discover 'things which
hold together,' and such things may be both analytic tools and objects of
intervention, such as birth, death, or marriage rates (Desrosieres 1991, 1992,
2001).
Moreover, while populousness sustains analyses of the collective or social
body that connect the relative size of its categories to policy intitiatives,
population sustains analyses that may centre on the categorization and recategorization,
the decomposition and recomposition, the articulation or rearticulation
of the molecular elements of the social body.
It is the political-statistical concept of population that preoccupied Foucault
especially in his 'governmentality' work: the concept whose discovery made
it possible for the Christian pastoral, with its concern for 'each and all,' to
become part of the mentality of government - and of governmental practice.
This concept, 'population,' made possible a logic in which the government of
the state came to involve individualization and totalization. The Christian
notion of a 'flock' or a 'corpus,' in which each member was the equivalent of
any other, and any member was the equivalent of the whole, was paired to
sovereign authority and, via this marriage, the state was 'governmentalized.'
The commonality of souls was replaced by common subordination to sovereign
political authority. Foucault repeatedly made reference to this process as the
'discovery' of population.
From Populousness to Population
The development of political theory and political administration in the wes
over the last several hundred years might be read in terms of a progressiv
displacement of 'populousness' by 'population.' The displacement affects how
the collective body is thought in political theory and addressed in practice.
When, how, and why this displacement takes place are theoretical and
historical questions of some moment. Among the matters at issue are the
relations between political forms and forms of knowledge; between theoretica
abstractions and empirical determinations; and between contemporary politica
power and its object. Foucault's account of population, I suggest, is characterized
by conceptual imprecision. When paired with a tendency to locate
conceptual innovation in political theory divorced from empirical practice,
such imprecision leads to the positing of an impossible discovery.
The tension between 'population' and 'populousness'- or between thei
respective logics - is certainly present in Foucault's early work. While it is
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
510 Canadian Journal of Sociology
outside the body of work I consider directly here, it is worth pointing out that
in The Birth of the Clinic Foucault examined the tension between a medical
logic of disease essences and species and a logic of induction centred on the
repeated observation of cases, where diseases presented in the complexity of
their circumstances. In the late 18th century, one sees the development of notions
which 'indicate that the medical field assumes a statistical structure,'
although later changes in the organization of the medical field were necessary
for this logic to triumph (Foucault 1973:101-102). In his analysis of the clinic,
the logic of population does not seem to have penetrated eighteenth century
medical practice because of organizational and political conditions. I suggest
that the historical and sociological caution of this analysis is missing from the
work on governmentality.
I propose that there are two sorts of interpretation and analysis of the
transition from populousness to population in the body of work under
consideration. One area of emergence and line of argumentation centred on
Foucault' s analysis of the development of social medicine. Here, I will suggest,
he came closest to a defensible historical sociology of the concept 'population.'
He related its emergence to the disciplined observation of individual
patients in the novel 18th century technology of the curing hospital. Recordkeeping
about diseased individuals in individual hospitals, he suggested, led
to comparative investigations of hospital records and hence to the awareness
on the part of doctors and others that there were disease populations. This
account was refined by an investigation of 18th century health politics, in which
the domain of health was seen to be defined and delineated more clearly
through a complex of specific struggles that mobilized many different actors
in pursuit of a variety of projects.
While Foucault thus seemed to be on solid ground in connecting disease
populations as abstractions to concrete medical determinations, his social
history of the hospital was rather cursory. Moreover, comparative statistical
investigation of the sort he described appears to have been a 19th rather than an
18th century phenomenon. In any case, this line of argumentation around
population, which approximated Foucault's own mid-1970s methodological
injunctions, was not pursued in any serious way.
A second area of emergence of population in Foucault's work is to be found
in his responses to criticism of the limited scope of the 'micro-physics of
power' presented in Discipline and Punish (Foucault 1979a) and related texts.
In the middle 1970s, Foucault began to argue that parallel to and in relation
with techniques of the individual body were bio-political techniques that aimed
at the collective body, population. The concern with population as a political
object stimulated Foucault's 1977-1978 lectures on state formation, the fourth
of which, widely known under the title 'Governmentality,' pointed to population
as the essential object of modern forms of government. The discovery
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 511
of population, known through political economy and organized through
apparatuses of security, sustained the transition to the 'governmental' state and
its analysis provided Foucault with a means of countering orthodox Marxist
accounts of capitalist state formation.
The concept of population as an artefactual abstraction of practices of
observation remained latent in those parts of Discipline and Punish (1979a)
that analysed the subjection of individuals to normalizing judgement. Mundane
techniques of writing, registering and recording attached individuals to new
kinds of administrative and hence also epistemological spaces, turning them
into 'cases,' elements in series, that could be rank-ordered and subjected to
practices of 'optimization.' Normalizing judgement implied authoritative
categorization and hence the construction of populations.
However, Foucault did not follow the explanatory tactic of relating
population as an object of political government to the development of largescale
practices of social observation and recording when he became concerned
with 'governmentality.' Instead, population was situated in the field of 'political
reason.' Foucault argued that population was implicit in the arts of
government which were anticipated in the anti-Machiavellian literature of the
17th century. However, these arts of government could not develop themselves
until propitious material circumstances appeared in the second half of the 18t'
century. When such circumstances did appear, rulers 'discovered' population
and made it the essential object of government.
I will argue that very considerable scepticism should be aroused by this
account. On its face, the account seems to be genealogical: population, like the
confessional, appears to emerge in a particular historical context in response
to particular concerns and interests, and to migrate to other contexts at other
moments in quite different circumstances. Yet if the distinction made above
between 'populousness' and 'population' is upheld, a genealogical analysis
cannot be sustained: the 'discovery' must be a moment of discontinuity, rather
than the reappearence of a theme. The discovery is not a rediscovery, but a
new construction; hence one must question its conditions of possibility. I
attempt to interrogate Foucault's account both of the conditions under which
such a discovery could take place and its substance.
The Birth of Social Medicine
Foucault began using the concept 'bio-politics' in the early 1970s, especially
in his work on the history of social medicine. Indeed, it was work on the
history of the hospital that led him to the study of institutional architecture and
thus to Bentham and the panopticon. The attempt to understand how medical
questions came to be articulated as social questions, similarly, led him to a first
engagement with bio-politics and population.
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
512 Canadian Journal of Sociology
The ideas worked out in detail in Discipline and Punish appear already to
have been formed in Foucault's course at the College de France in 1972-1973;
'bio-politics' was invoked in his course at the State University of Rio de
Janeiro in October 1974. It was this notion that would be mobilized as the link
between micro- and macro-physical forces in response to criticism of the
arguments published in Discipline and Punish. (Foucault 1989:29-51; 1994b).
As in a great many of his texts before the shift to 'governmentality' and his
later preoccupation with ethics, Foucault's free use of Marxist concepts and
arguments, and his invocation of Marx's intellectual authority, are striking in
the Brazilian course, 'La naissance de la medecine sociale,' even as he
attempted to shift the ground of argument away from the conventional
intellectual presuppositions of those close to the French Communist Party.
Here he argued that the first object that the capitalism of the late eighteenth
and early nineteenth centuries subjected to socialization was the body as a
productive force, as labour power. Individuals were not controlled by society
simply by ideological manipulation, but also by and through the body. 'For
capitalist society,' wrote Foucault, 'it is the bio-political which was of first
importance, the biological, the somatic, the bodily. The body is above all a
bio-political reality; medicine is a bio-political strategy' (Foucault 1994b:210).
But if capitalism socialized labour power, Foucault argued, this was not
medicine's object in the beginning period of capitalist development. How, he
asked, did the socialization of the body take place and how did medicine come
to be implicated in it? There were three sequential, nationally-specific
developments: the appearance of state medicine in Germany, urban medicine
in France, and, after these two historically, what we would now call occupational
medicine in England.
The examination of the development of state medicine in Germany served
as an early occasion for Foucault to rehearse some of the arguments around
cameral science that later appeared in the work on governmentality. He drew
very heavily on the work of George Rosen (Rosen 1974). Germany was the site
for the development of the Staatswissenschaft, the science of the state, whose
object is to create an inventory of resources, including the conditions of life of
the inhabitants, at least insofar as these constitute forces for the state. This
form of knowledge developed here, Foucault claimed, because of the peculiar
conditions of German state formation. A unitary German state appeared much
later than in France or England. The existence of small states, deprived of large
standing armies, but in continual conflict and contest, 'made necessary and
possible this discursive awareness of the state functioning of society [du
fonctionnement 6tatique de la societe].' Added to this was the economic
stagnation characteristic of the German states in the wake of the Thirty Years'
War, which deprived the bourgeoisie of an economic field of action and led
members of this class into an alliance with the sovereign authority. They
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 513
provided both human and material resources for projects of state formation
(Foucault 1994b:210-211). The first modern state developed in Prussia where
there was neither strong political power nor economic development.
These remarks on state formation were made solely to account for the first
emergence of state medicine. One important anticipation of later arguments
was the claim, again borrowed directly from Rosen, that in the mid-eighteenth
century there was an enormous accumulation of information, by speciallycreated
state agencies, about medical conditions, based directly on the systematized
observations of local officials. This knowledge, whose accumulation
presupposed the subordination of medicine to state authority, was about bodies
within the state territory as such and not simply about workers' bodies.
Commonality among these bodies was found in their subordination to the state,
not in their common class position.
It is not workers' bodies that interested this public administration of health, but the bodies of
individuals themselves who, collectively, constitute the State. It is not a matter of the labour force,
but of the strength of the State faced with these conflicts, undoubtedly economic but also political,
which opposed it to its neighbours. To this end, medicine must perfect and develop this state
strength. This preoccupation of state medicine includes a certain political-economic solidarity. It
would thus be misleading to seek to tie it to an immediate interest in producing a strong and freelyavailable
labour force (Foucault 1994b:214).
Foucault takes his distance from conventional Marxist accounts through the
perception that solidarities may emerge from common subordination to
administrative authority and its categorizations of social relations. The insight
is presented as an argument against the analytic privileging of class solidarities.
Social medicine emerged in Prussia as an administrative project of the
state authority. Common subordination to state medical authority fashioned a
possible grounds for solidarity, but common subordination is also one ground
of the equivalences that make human subjects appear as a population.
Common administrative subordination, for which medicine was a vehicle,
also emerged in England and France, but following different routes. In France,
argued Foucault, it was the urban question, the emergence of the city as a
large-scale centre of production and exchange, that contributed to the undermining
of multifarious feudal powers and their replacement by a common and
homogeneous administration. The development of an impoverished urban
working population, undergoing processes of proletarianisation, also propelled
the formation of a common authority capable of dealing with the political
threats potentially posed by it, threats emerging out of its conditions of life.
In the eighteenth century, the large cities, and especially Paris, were the site
of frequent social panics, which Foucault described as 'politico-sanitary' in
character. The medieval cities were already equipped with means of dealing
with extraordinary threats in the form of leprosy and plague, and did so by two
characteristic types of measures: exclusion in face of the former, and
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
514 Canadian Journal of Sociology
quarantine in face of the latter. In Foucault's account, the reaction of urban
authorities, pressured by the bourgeoisie, to the politico-sanitary menace was
the refinement and generalization of quarantine. As he put it, 'public health
was a refined declension of quarantine' (Foucault 1994b: 219).
Quarantine operates by identifying and separating out problematic
dimensions of social life and social relations and then subjecting them to
particular treatment protocols. Although Foucault was not yet invoking the
concepts of totalization and individualization, they are latent in his notion of
quarantine. For instance, he argued that towards the end of the eighteenth
century, the political-sanitary problem of dealing with the bodies of the dead
led to the individualization of the cadaver, to the individual coffin and grave.
The dead of Paris were exported to the countryside, lined up one by one in
rows, like a veritable army (Foucault 1994b:219).
Quarantine applied to many other dimensions of urban life that menaced
health and provoked unrest: garbage removal, the isolation of slaughterhouses,
or the laying out of streets to promote the circulation of air. What these
techniques share is participation in centralized administrative practices that
treat the city as a whole and that operate upon its vital processes through logics
of categorization and sub-division: they are bio-political.
It was in England, finally, in the second third of the nineteenth century, that
social medicine took the bodily existence of the developing working class as
its object. There were three principal causes of what Foucault described as a
change in the position of the poor from a necessary and largely invisible social
category to an object of politico-medical concern. First, the struggles surrounding
the French revolution and the agrarian radicalism of the 1830s made
the 'needy population' into a political force capable of revolt. Second, the
dependence of existing social organization on the services provided by the
working poor was undermined by the development of new means of communication
and transport. Such developments made them more menacing politically.
Finally, there was the fear and anxiety of the plebian and proletarian
populations intensified by epidemic cholera, especially the epidemic of 1832.
These developments collectively propelled an interest in social medicine in
England, and Foucault continued to argue that its prime vehicle was the 1834
Poor Law. However, the English case pointed to the development of multiple
forms of medical power and the articulation of politico-sanitary concerns with
social relations along different axes.
Population was both analytically and historically prior to class relations as
an object of rule in the above lecture. In a second Brazilian lecture concerned
with the related theme of the hospital as a general curative institution, Foucault
argued explicitly that the disciplined hospital-space gave rise to the perception
of collections of individuals as populations. 'With the introduction of discipline
into the space of the hospital,' he wrote, 'medicine offers as an object of
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 515
observation a vast field bounded on one side by the individual itself and on the
other by the entire population.' Hospitals began to keep daily registers and
when the registers of different hospitals were compared, pathological phenomena
came to appear as population phenomena. 'Thanks to the technology of
the hospital, the individual and the population present themselves simultaneously
as objets of knowledge and medical intervention....The medicine
which forms in the course of the 18th century is at once a medicine of the
individual and of the population' (Foucault 1994g:521). This 1974 text is one
of the first in which Foucault writes of individualizing and totalizing powers.
What is interesting in this account for my purposes is that 'population' is
an artefact of the medical administration of groups of individuals and of certain
investigative technologies. It emerges out of administrative practice that seeks
to specify and reorganize social relations. Common subordination to medical
administration creates the commonality, the conditions of equivalence, for the
bearers of disease to appear as a population. It is the increasingly complex
material determinations of medical administration that sustain the abstraction
'population,' even if this population is to be found on the terrain of the theory
of disease.
The Politics of Health in the Eighteenth Century
Foucault moved away from this first, heavily state-centred account of the birth
of medical police, and away from his analysis of social medicine as ultimately
derivative from the practices of quarantine, in his influential 'The politics of
health in the eighteenth century' (Foucault 1980a, 1994c).' For quarantine, he
began to substitute the more general notion of a consciously thought-out 'nosopolitics,'
the politics surrounding the establishment of disease classifications
and treatment protocols, accompanying a new regime of social sanitation. The
move allowed a closer attention to those aspects of medical politics best
approximating the model of positive power under development, even if a
casualty of it has been the possibility of analysing the role of the establishment
of cordons sanitaire in medical administration.2
Now, however, instead of seeing medicine as closely tied to state formation,
Foucault offered an analysis of the emergence of social medicine that stressed
the separation of medical policy from a close preoccupation with poor relief.
The analysis became more specific empirically, moving away from a consideration
of abstract structural transformations like the 'rise of capitalism' and
1. There is a second version with a bibliography (Foucault 1994d)-rare for Foucault.
2. Perhaps, however, quarantine and the idea of cordon sanitaire reappear in the later notions of
'disciplinary enclosure' and 'partitioning.'
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
516 Canadian Journal of Sociology
the formation of the modern state. At issue were the emergence of a market for
medical services, innovation in medical techniques, the professionalization of
medical practitioners, the development of benevolent associations, and of
learned societies concerned with the observation of social conditions, among
other forces. The state played a variety of roles in relation to these developments
but, collectively, the diverse ways in which health and sickness were
problematised contributed to an awareness of them as phenomena of population:
'the health of each as an urgent matter for all; the state of health of a
population as a general objective.' Health became an imperative duty for each
citizen and for the collectivity, an anticipation of Foucault's later interest in
pastoral power (Foucault 1994c:15).3
What is central to the politics of health in the 1 8th century according to Foucault,
is the emergence of the health and well-being of the population as an
essential objective of political power. Political power aims at the 'social body,'
and society comes to be managed as a domain for physical well-being and
longevity, especially through the application of the techniques of police to that
social body (Foucault 1994c:17). The treatment of the sick poor, the main
leverage point for medical policy, comes to be inscribed in the broader field
of the health of 'population.'
Foucault argued that while this transformation was certainly related to the
reproduction of labour power, it was not reducible to the analysis of class
relations. The shift towards the policing of the social body was related to the
political economic consequences of the 18th century demographic transition,
which created a need to integrate rapidly increasing numbers of people into the
apparatus of production and to control them closely. It was these forces that
made 'population,' 'with its variables of number, spatial and temporal distribution,
longevity and health,' appear not simply as a theoretical entity, but
'as an object of surveillance, of analysis, of intervention, of initiatives aimed
at modification' (Foucault 1994c: 18). Technologies of population sought to
discover the secrets of demographic processes and to render them susceptible
to management. They also caused the body, both as the individual body and the
social body, to emerge as the site of a much larger complex of determinations.
Foucault concluded this set of reflections with the observation that 'the
biological traits of a population become relevant elements for economic
management, and it is necessary to organize around them an apparatus which
3. The passage reads 'la sante de tous comme urgence pour tous; l'etat de sante d'une population
comme objectif g6enral.' The highly readable translation by Gordon et al, oddly, at 168, gives
'the problem of the health of all as a priority for all, the state of health as a general objective of
policy,' leading us to miss the pastoral overtones at first; but later they give III: 16, 'L'imperatif
de sante: devoir de chacun [my emphasis] et objectif genral,' as 'The imperative of health: at
once the duty of each and the object of all' at 170.
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 517
will not only ensure their subjection, but also the continual increase of their
usefulness' (Foucault 1994c: 18).
Having presented this historical sociological account of the emergence of
the object 'population,' closely related to new technologies of knowledge,
Foucault suggested that the concern with it shaped family policy and also
urban policy, especially through the extension of the notion of 'regime' [i.e.
regimen] of health to the social body as a collective. Social sanitation was
particularly significant in the new regime and, in the nineteenth century,
doctors increasingly infiltrated administrative structures and the machinery of
power. Indeed Foucault remarked here, as he was fond of doing, that the
systematic observation at the roots of the discipline of sociology is to be found
in the activities of doctors concerned with questions of the social body
(compare Poovey 1995). Political medicine sought purchase on population and
this quest also led to a transformation in the social role of the hospital.
The Interview
In a wide-ranging 1976 interview with Alessandro Fontana and Pasquale
Pasquino, familar to English readers as 'Truth and Power'(Foucault 1980b,
1994i),4 the question of 'population' as the link between the micro-powers of
discipline and the general administration of social relations was posed
explicitly. The interviewers suggested that disciplinary techniques did not
subsist on their own, but rather were connected to the more general phenomenon
of population, which appeared in the eighteenth century as an object of
scientific investigation. They proposed that disciplinary power was in fact linked
to two bodies: population, as an assembled mass of elements, and its component
parts, docile bodies.
Foucault responded that this was precisely the line of development of his
own work, and proceeded to draw a contrast between feudal powers, based on
exactions and levies and tied to rituals, ceremonies, and symbols of loyalty,
and new forms of power appearing in the seventeenth and eighteenth centuries
based on social production and productive services. This shift in the mode of
production and appropriation of the social surplus propelled the penetration of
techniques of power into bodies and at the same time created a need for power
to deal with population. The same techniques that led to the accumulation of
capital, Foucault observed, led to the accumulation of human beings and hence
4. The English and French versions differ in some important particulars, but I have not had access
to the original Italian. Notice that Gordon et al give 'the discovery of population as an object of
scientific investigation' where the French gives 'the taking of population as an object of
scientific investigation'; Gordon et al give 'draw support from,' the French 'connected to' etc.
etc. These are not insignificant.
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
518 Canadian Journal of Sociology
gave rise to the administrative problems of collective health, hygiene,
longevity, fertility and demographic phenomena more generally. The political
importance of sex was precisely that it was located at the intersection of
individual bodies and phenomena of population.
The Stitches of Power5
Foucault again took up the problem of the elaboration of an alternative to what
he would soon call the 'juridico-discursive' model of power in a public lecture
given at the University of Bahia in 1976. After a peculiar attempt to suggest
that his work was an elaboration of analyses contained in volume II of Marx's
Capital, Foucault argued that primitive readings of Marx saw the great
technological advance of the West in the steam engine, but neglected the
invention of the political technology that made industrial innovation operative
(Foucault 1994f).6 This political technology developed in two directions or
included two groupings of practices: those of discipline, already well-known,
and a second grouping 'which was developed...especially in England' in the
second half of the 18th century, composed of 'technologies that were not directed
at individuals as such, but rather were directed at population.'
Here Foucault stressed that the 18th century 'discovery' of population was
what distinguished monarchical power from those forms of power that followed
it. 'And what does it mean to say 'population',' he asked?
It does not mean simply a numerous human group, but living beings penetrated, compelled, ruled
by processes, by biological laws. A population has a birth rate, a death rate, an age curve, an age
5. 'Maille' literally translates as 'stitch' in English, but it is also the word from which comes the
English 'chain mail,' armour composed of a set of rings. The sense, in any case, is of an
endlessly interconnected system of relations.
6. He suggested that the elements of an alternative analysis were probably present in the work of
Jeremy Bentham, but they were obviously present in the work of Karl Marx, especially in his
Capital Volume II. 'What we can find in book II of Capital is, in the first place, that there isn't
simply a power, but rather several powers. Powers, that means forms of domination, forms of
subjection, which function locally, for example, in the workshop, in the army, in property of the
slave sort, or in property where there are servile relations.' Foucault continued at length in this
vein, crediting Marx with noting the positivity of power resulting from the division of labour and
surveillance in the workshop and with, at least implicitly, presenting an analysis of power as
technology. The privileging of the state apparatus and the juridical superstructure, according to
Foucault, was a social-democratic re-reading of Marx.
Foucault did not take up and pursue these Marxist lines of argument in any systematic way.
They are peculiar, since Capital II is mainly taken up with a lengthy and technical demonstration
of the possibilities of capitalist reproduction, with which Foucault did not engage. Perhaps in this
tumultuous period he felt compelled to hang his arguments on the hook of a purified Marxism.
Yet, the shift from feudal to disciplinary modes of power was also directly related to the 'needs
of capitalism' in this and other texts of the period.
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 519
pyramid, a degree of morbidity, a state of health, a population may perish or may, on the contrary,
expand (Foucault 1994f:193).
While Foucault' s account of population seemed to move indifferently from the
early to the late eighteenth centuries and from Germany to France to England,
the theoretical argument was increasingly refined. The discovery of population
meant that it comes to be understood that power can not operate simply on
individual bodies by means of exaction, but rather must operate on individuals
as members of a biological species, at least if this population is to be put to
productive use. 'The discovery of population is, at the same time as the
discovery of the individual and the trainable [dressable] body, the other major
technological nucleus around which the political procedures of the West were
transformed.' In parallel to 'anatomo-politics' there emerges 'bio-politics'
(Foucault 1994f:193). Bio-politics problematises elements of individual
comportment, such as how to get people to have more or fewer children, but
it also problematises structural conditions, such as the quality of housing. It
advances both through the development of observational techniques, including
statistics, and through the operations of the great administrative organs of state.
Through anatomo- and bio-politics, Foucault argued, 'power has become
materialist. It ceases to be essentially juridical.' Sex, once again, appears as the
key link between the two political technologies.7
Population was again described by Foucault in this essay both naturalistically,
as an empirical phenomenon with 'processes,' and artefactually, as the
construction of objects of government through the investigation of administered
individuals. Capitalist development was held to be responsible for
demographic transformation which in turn confronted political administration
with problems with which it must deal. 'Population' is slippery, existing in the
account as an object of police, as a collective body, and as a statistical artefact.
The statistical artefact, it is worth underlining, is seen to have been 'discovered'
in the eighteenth century.
Security, Territory and Population
The development of population as an object of political knowledge and of
regulatory techniques was the focus of Foucault' s course on state formation at
7. Foucault, at 201, comes back to his understanding of Marx in responses to questions from a
critical audience. 'Once again, here a certain kind of academic Marxism frequently makes use
of the opposition of the dominant class against the dominated class, dominant discourse against
dominated discourse. Well, first of all, this dualism will never be found in Marx ... because in
effect Marx is too clever to be able to concede any such thing; he knows perfectly well that what
solidifies relations of power is that they never end, there isn't on one side a few and on the other
side many; they are everywhere: the working class reconveys relations of power, it exercises
relations of power.'
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
520 Canadian Journal of Sociology
the College de France in 1977-8, and the concept 'population' echoed increasingly
loudly through his work in this, the period in which he first began
to use the concept 'governmentality.' Part of his interest was also with the
development of the Christian pastoral and with the migration of its concern
with 'each and all' into political administration (Foucault 1989:99-106).
In his own summary of the course, 'governmentality' was defined as 'the
manner in which the conduct of a mass of individuals comes to implicated, in
an increasingly marked manner, in the exercise of sovereign power,' and
Foucault stressed that the shift in question was not in any simple sense a
change from a preoccupation with sovereignty, but rather a shift in emphasis.
Rule as a 'mentality' of government, as a particular way of problematizing
social relations, emerged historically, in this account, with the breakup of
feudalism and the appearance of a new order of competition and conflict
among states. Such conflict made it imperative both to know and to work to
increase the internal forces of states, thus giving rise to the new rationality of
'reason of state.' In Foucault's account, quite apart from the political theories
of reason of state, this rationality took form in two major sets of knowledge
and political technology: a diplomatic-military technology involving a set of
inter-state alliances and military construction; and the technology of police,
designed to augment the forces of the state from within. These two technologies
were joined by trade and by inter-state monetary circulation. It was
enrichment through trade and the accumulation of money that made it possible
to increase population, the labour force, and hence military might. The new
reason of state focussed on the couple population-wealth, and both mercantilist
economic analysis and cameralism in political administration took its increase
as their privileged object.
The problematization of population-wealth as an object of rule, according
to Foucault, was one of the conditions for the development of political
economy as a form of investigation and analysis. Political economy emerged
in part out of the realization that the administration of population-wealth could
not be limited to detailed regulation and administrative coercion. For reasons
that remain obscure, Foucault claimed that it was with the Physiocrats that
population ceased to be 'the simple sum of the subjects who inhabit a
territory.' Instead, it came to be seen as a 'variable that depended upon a certain
number of factors' which, although themselves not facts of nature, could
be analysed rationally. As a political problem, population derived from the
experience of police and emerged in correlation with the birth of political
economy. Population
is not conceived as a collection of subjects of right, nor as a mass of hands destined to labour; it is
analysed as a mass of elements which, on the one hand, belongs to the general administration of
living beings (population then depended on the 'human species': the notion, new to the period, is
to distinguish the 'human genus' [a more general category]) and, on the other hand, may provide
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 521
a hold for concerted interventions (by the intermediary of laws, but also by changes in attitude, in
ways of doing and living that may be achieved by 'campaigns') (Foucault 1989:103-4).8
Foucault's seminar in this year focussed particularly on the techniques of
police science, and reprised some of the arguments made earlier in his work on
the birth of social medicine. Police science developed a bio-politics 'which
tends to treat the 'population' as a mass of living and co-existing beings, which
evidence biological traits and particular kinds of pathologies and which, in
consequence, give rise to specific knowledges and techniques' (Foucault
1989:106).
Governmentality
Foucault described the 1977-1978 course of lectures as being 'on the formation
of the state.' It has been the transcribed and edited version of the fourth
in the course, published under the title 'governmentality,' that has been
particularly influential for English-speaking intellectuals (Foucault 1991; I am
working from Foucault 1994a).
Foucault began this lecture with an examination of the anti-Machiavellian
literature of the 16th and 17th centuries which contested the proposition that the
object of political analysis was the sovereign authority of the prince. There
emerged here, by contrast and proto-typically, a literature concerned with the
'art of government' conceived as the 'right disposition of things.' Political
analysis moved away from a consideration of sovereignty as authority above
and outside the organization of bounded social relations, towards an analysis
of the internal management of the state. For Foucault, the shift in analysis
foreshadowed a transformation in the political form of the state itself, away
from feudal state forms based on ritualized extraction towards what he called
the 'administrative state.' The expanding literature on the art of government
itself was contextualized by the break up of feudal powers and the emergence
of a new competitive European state system, on the one hand, and by the
Reformation's subversion of official religion as the basis of social allegiance,
on the other. Shifts in the basis of allegiance problematised the matter of the
ends of government, and inter-state rivalries, once again, created an interest in
a knowledge of the internal forces of states and of the means to augment them.
8. Foucault (Foucault 1994e) is reprising early work on Cuvier and Darwin. It is interesting that
he credits the latter with the theoretical insight that there is no discontinuity between individuals
and species, a way, I think, of reflecting on 'each and all,' but Keith Tribe has pointed to the
weakness of this claim in relation to its characterization of political economy (Tribe 1978:80-
109).
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
522 Canadian Journal of Sociology
In the art of government, then, governing ceased to be seen as existing on
the external boundaries of the state; it was inside the state, inside society. And
Foucault emphasized that these arts presented governing as practices in
continuity from the individual's government of itself, through the father's
government of the household, to the prince's government of the state. The art
of government sought to introduce 'economy,' conceived as right management,
initially a concept that applied to household government, into political practice.
For Foucault, this concern, which develops more fully in the late 18th
century, marks the beginning of the conceptual shift towards our contemporary
use of the word 'economy' to designate an autonomous region of social
relations.
The art of government was opposed to the theory and practice of sovereignty;
it invoked not law or the imposition of rules, but rather 'the right
disposition of things.' This meant, first, that the ends of government were
multiple and, second, that these ends were to be found in the objects of government
themselves. This, for Foucault, represented an important break with
sovereignty. Government was a positive form of power and, indeed, Foucault's
account describes it entirely in such terms. Thus, the end of government 'is to
be sought in the perfection, the maximization, or the intensification of the
processes it directs'; or, again, 'he who governs, must only govern to the extent
that he thinks and acts as if he were in the service of those who are governed'
(Foucault 1994a: 146-147). There is no interrogation of the ethical or political
substance of this positivity in the essay, and no attention to the ways in which
positive powers are also repressive powers.
Foucault attempted to make a connection between the quite rudimentary
theoretical articulations of the art of government in the anti-Machiavelli
literature, and political practice from the 16th century. His argument acquires
heavily idealist overtones at this point, a product perhaps of its admittedly
schematic character, but something of which I think one should be extremely
wary.9 The theory of government has 'correlations' with political practice, but
Foucault argues that its realization was blocked by the limiting framework of
mercantilism and the 'reason of state' characteristic of the developing
administrative states. The changing geo-political conditions of the 16th and 1 7th
centuries propelled the development of the sciences of state, 'statistics' especially,
that would later sustain the arts of government. But the framework of
'reason of state,' with its transcendental notion of state legitimacy, prevented
the full development of the art of government, which has to operate not on the
basis of transcendental rules, but on the basis of the reality of conditions within
9. This claim in particular has been subjected to extensive and telling criticism in (Dupont and
Pearce 2001).
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 523
the state territory. 'One can say at the outset that this reason of state was a kind
of obstacle for the development of the art of government until the end of the
18th century' (Foucault 1994a: 148).
Unfortunately, Foucault spoke consistently in this part of his lecture as if
'the art of government' was fully formed and waiting in the wings of historical
development for the appropriate conditions in which to realize itself. Whether
or not this way of presenting the matter was a product of the lecture format, it
is problematic as an historical account. It breaks sharply with Foucault' s earlier
methodological injunctions that recommended studying the ways in which
particular techniques, tactics, and practices emerge out of particular localized
struggles and that recommended investigating general strategies as conjuncturally-specific
assemblages of such elements (for instance, Foucault
1994f:201). Still, he argued that the conditions blocking the 'realization' of the
arts of government were in large part material: the Thirty Years' War, peasant
and urban rioting in the mid-17th century, financial crises and so on. 'The art
of government could only unfold, be thought out, lay hold of and multiply its
dimensions during a period of expansion' (Foucault 1994a:648)."' But the
obstacles to the art of government were also 'institutional and mental' structures.
So long as politics was thought in terms of sovereign authority, the arts
of government could not develop fully. The example given is mercantilism,
which problematises the development of the forces of the state but which
cannot do so effectively within the framework and with the practices of
sovereignty. 'Mercantilism attempted to mobilize the possibilities provided by
a rational art of government from within mental and institutional structures of
sovereignty that blocked it' (Foucault 1994a:649).
Once again, it should be stressed that implicitly we are led to believe that
there is a will to govern existing outside actual governing practices, attempting
to realize itself through such practices, but finding them inadequate to itself
(cf. Dupont and Pearce, 2001:132ff). Foucault spoke at such a level of
abstraction that we never see concrete and grounded struggles over rule and
domination as motor forces in the development of governmental theory and
practice. Nor did he offer any reason for us to believe that mercantilist
investigations of the forces of the state were inadequate in practice for the
political or administrative projects they sought to address. Surely they appear
inadequate only in hindsight: there are teleological overtones to this analysis.
10. The original uses reflexive verbs: 'L'art de gouverner ne pouvait se deployer, se reflechir,
prendre et multiplier ses dimensions que pendant une p6riode d'expansion....' The translation
in Burchell et al (Foucault 1979b:97) reads 'The art of government could only spread and
develop in subtlety in an age of expansion....' How can the art of government be articulated
before the conditions of its own articulation exist, except in an idealist conception?
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
524 Canadian Journal of Sociology
So, how is the art of government freed, finally? In a context marked by the
18th century demographic expansion, the growth of the supply of money and
of agricultural production, which fed off one another, in 'keeping with circular
processes well-known to historians.' In this general context, more precisely,
the art of government was freed by the 'emergence of the problem of
population.' It appears to be the emergence of this problem that encouraged
Foucault to speak now of 'the science' rather than of the 'art' of government;
but how and where the problem of population 'emerges' or is 'discovered' was
not addressed. The argument becomes circular. It is through the science of government
that 'economy' came to be refocused on the domain we know as the
'economic' and it was through this science that it became possible to come to
grips with the problem of population. But also, the perception of population
problems and the recentring of economy made it possible to consider the
problem of government outside the framework of sovereignty. And statistics
escaped from the framework of sovereignty to act as one of the main forces in
unblocking the science of government. Foucault left it to 'the historians' to
work out the details.
Foucault's argument attempted to account for transformation in state forms
and the problematic of population was the pivot upon which transformation
turned. But how population came to be problematised in ways that were not
restricted to the narrow preoccupation of mercantilism with population-wealth
was nowhere examined. The key practical/conceptual transition, from
'populousness' under sovereign authority and police to 'population' under
liberalism was not discussed. Nonetheless, it was the problematic of population
and the investigative technology of statistics that freed the art of government
from the limitations of sovereignty and enabled it to focus on a new order of
phenomena characterized as 'economic.' Statistics demonstrated that population
is an object possessed of its own regularities and its aggregations of
population demonstrated the existence of objects that cannot be reduced to the
operations of families (Foucault 1994a:651).'1 Thus, argued Foucault, the
problematic of population provided the arts of government with a more general
object and at once reduced the family to the status of a segment of this general
object. The family came to be a relay for projects whose object is population.
Population came to be the prime object and ultimate end of government, at the
same time as it became a social subject.
The concern with 'how to get rid of Marxism' echoes loudly in this
analysis. The formation of the modern state is not to be understood as the
rising hegemony of an executive committee of the bourgeoisie preoccupied
11. The English translation is weak here and misses the emergence of phenomena through
aggregation.
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 525
with the reproduction of relations of production. The modern state is a
governmental state; its agencies are multiple; the techniques and tactics at their
command diverse. The object of government is not first and foremost the
maintenance of capitalist exploitation, and its essential subjects are not
contesting social classes. On the contrary, 'population' has come to be 'the
ultimate end of government: because fundamentally, what else could its end
be?.' At the same time, 'population comes to be a subject, as well as an object
of government' (Foucault 1994a:652).
The theoretical, political, and analytic stakes in Foucault's position are
considerable. In the era of governmentality in which we now live, he
maintained, the state acquires a new form. It is not a unitary, internally
coherent entity, charged with a few functions centred on the reproduction of
the relations of production. Hence, it is not localizable, susceptible to seizure
by a revolutionary political party. Rather, the only real political struggles and
the only real terrain of political contestation, argued Foucault, are to be found
in the techniques and tactics surrounding the government of population. These
struggles define the shifting boundaries of what is public and what is private,
what comes within the competence of the state and what lies outside it. If the
state continues to exist, it is precisely because it has been governmentalized,
imbricated in the detailed management of population, and thus both a central
instance and an ephemeral entity. While it is striking in what benignly
welfarist terms Foucault described this 'governmentalized' state's relations
with its object, more striking still for my purposes is the heavy work 'population'
is called upon to do in the analysis and the apparent lack of reflection
involved in Foucault's multiple uses of the concept. 'Population,' 'the
conditions of populations,' the 'field of population,' the 'movement of population,'
population as 'a subject of needs' and as 'an object in the hands of government,'
the 'interest of the population': all this in a single page! (Foucault
1994a:652).
To foreshadow the work necessary for the further study of the history of
governmentality, at the close of this lecture, Foucault noted that there were
three things whose conjunction required investigation: the Christian pastoral,
the techniques of military diplomacy and police. These correspond to the
points of the triangle sovereignty-discipline-government around which modern
government is oriented. To my knowledge, the 'sovereignty' point of the
triangle was not pursued by Foucault before his interest shifted to a concern
with liberalism, self-government, and finally, ethics. The pastoral, an examination
of which I take up shortly, was discussed in detail after these lectures,
and police had already figured centrally in earlier work. The investigation of
the ways in which the discipline and regulation of populations is shaped by the
military-diplomatic relations existing in the system of states has yet to be
undertaken by those interested in 'governmentality.'
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
526 Canadian Journal of Sociology
The Birth of Bio-politics
Foucault's 1978-1979 course and seminar pursued the question of bio-politics,
particularly with respect to the fate of the government of population under
'liberalism,' taking as its case public health reform in mid-nineteenth century
Britain. 'In a system concerned with respect for the rights of subjects and free
individual initiative, how was the phenomenon 'population,' with its effects
and its particular problems, taken up' (Foucault 1989:109-110)? In the seminar,
the now-familiar distinction was made between police, as a governmental
technology dominated by reason of state and concerned to regulate everything,
and liberalism as a political theory continually concerned with the possibility
of governing too much. The point of departure of reason of state was the state
itself and government derived immediately from state interests. Liberalism, by
contrast, takes as its object of government 'society' and attempts to tailor
governmental instruments and practices to the nature of social relations in
'society.' The fate of attempts to govern the market and market relations in
particular, in this account, demonstrated the inadequacy of the model of police
for the maximization of economic development. Foucault argued that the
question that called for further study was 'the way in which problems specific
to life and population have been presented' in governmental technologies
haunted, since the end of the 18th century, by the question of liberalism (Foucault
1989:109-119).12
Each and All
Although accounts of population are scattered throughout other texts, including
The History of Sexuality, two other works are worth mentioning briefly here,
'Omnes et singulatim' and 'The Subject and Power' (Foucault 1981, 1983). In
the first, Foucault gave an account of the emergence of pastoralism as a way
of conceptualizing power relations, arguing that it represented the individualizing
moment of modern power. The modern Welfare State, he argued, is the
manifestation of pastoralism in contemporary administration, spending most
of his lecture recounting once again the development of the theory of police.
'The Subject and Power' in fact contains two essays written at different
moments and with rather different foci. The first of these is the more relevant
here. In it, Foucault articulated a fuller notion of subjectification than that
present in his earlier works, one which attended to subjects as capable of
12. It is pertinent to notice that 'society' and 'population' co-exist as objects of government in
Foucault's account of liberalism. Later writers have tended to substitute 'society' or the 'social'
for population as liberalism's governmental object (Curtis 2002; Donzelot 1984).
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 527
reflection and self-formation. He outlined at length his analytics of power,
stressing that the study of power should begin with resistance. Contemporary
societies continue to contain struggles against direct domination and against
capitalist exploitation, but increasingly they manifest social struggles against
the forms of subjection themselves. And what are these forms of subjection?
They possess a totalizing moment, in which states group subjects together in
order to rule them, and an individualizing moment, in which subjects are
separated as the objects of pastoral power. As Foucault put it, in a well-known
passage,
The reason this kind of struggle tends to prevail in our society is due to the fact that since the
sixteenth century, a new form of political power has been continuously developing. This new
political structure, as everybody knows, is the state. But most of the time, the state is envisioned as
a kind of political power which ignores individuals, looking only at the interests of the totality, or
I should say, of a class or a group among the citizens.
That's quite true. But I'd like to underline the fact that the state's power (and that's one of the
reasons for its strength) is both an individualizing and a totalizing form of power. Never, I think, in
the history of human societies...has there been such a tricky combination in the same political
structures of individualization techniques, and of totalization procedures.
Such is the case because pastoralism has become a modality of state power.
We should not think of the state as an entity that exists above and beyond
individuals. Rather, it involves the organization of new forms of individuality:
'we can see the state as a modern matrix of individualization, or a new form
of pastoral power.' Pastoralism now aims not at salvation, however, but at the
government of life, health and well-being.
With pastoralism's increasing importance, suggested Foucault once again,
the tasks of political struggle have also perhaps changed. Their aim, perhaps,
should no longer be 'to try to liberate the individual from the state, and from
the state's institutions, but to liberate us both from the state and from the type
of individualization which is linked to the state' (Foucault 1983:213-6). Finally,
note one of Foucault's last definitions of 'governmentality': it is the relation
'between the technologies of domination of others and those of the self,'
that is the relation between state administration and pastoralism (Foucault
1988:19).
Conclusion
Foucault' s work was influential in re-orienting political sociology and political
theory towards a focus on the technologies and practices that are associated
with the construction of statistical populations and towards the administrative
practices that 'population thinking' makes possible. There was active interest
in and a wave of publication on the subject in France at the time he was
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
528 Canadian Journal of Sociology
writing. In the context of the debates of the 1970s, the demonstration that bases
of solidarity existed in administrative categories that were not reducible to
class relations was an important contribution. The same is true for the stress
Foucault laid on the problem of the collective body as an object of government
under liberalism. Moreover, the broadening of the concept of 'government' to
refer to all instances of the 'conduct of conduct' did indeed shake the often
extremely narrow hold economistic Marxism exerted over the political
imagination. Foucault's work made it possible to carry forward attempts to decentre
the state.
That being said, it is important to stress that Foucault does not offer a
cogent account of 'population.' Sorting out his analysis is made complex by
the conflation of the three concepts of populousness, the social body, and the
statistical construct 'population,' to all of which at times he applies the word
population. Most of Foucault's observations and claims about the development
of anatamo-politics and bio-politics do not require the politico-statistical
concept of population. He mistakenly locates the effective emergence of
modern demographic concepts in the eighteenth century and erroneously
suggests that they are 'discovered' by political authorities.
Analyses of anatomo-politics and body-politics are perfectly compatible
with the concept populousness, with the logic of police, and with the growth
of statistics. Yet the 'statistics' that are involved are descriptive rather than
inductive: eighteenth and early nineteenth century statistics remains an
inventory science concerned with 'the methodical and positive exposition of
the objects which compose the wealth and strength of the State' (Denis
2000:13). It employs a logic of classification in which things as expressions of
essences are put in their appropriate places.
The science and practice of police in seventeenth and eighteenth century
European states gave rise to vast stores of information about conditions within
the relevant territories and laid some of the groundwork for the shift to the use
of inductive logic in statistical practice. Municipal, royal, religious and later
departmental authorities in eighteenth century France conducted an extensive
array of inquiries into a variety of conditions and their efforts made it possible
to know such things as the numbers of births, marriages and deaths. The
practice of police and inventory statistics makes it possible to determine how
many people die, when and where, within limited territories but there are no
18th century national population registers. Such things remain in their local
singularity. They are not paired to an inductive logic that would have permitted
the emergence of conceptions of 'rates.' The concern of police with populousness
could sustain campaigns directed at the individual body which sought to
affect the social body, as debates over the effectiveness of smallpox vaccination
show (Cole 2000; Coleman 1982).
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 529
What police, populousness, and inventory statistics could not do was to
sustain the kinds of practices that make it possible for social relations, events,
and conditions to appear in the politico-statistical form of population. It is this
political knowledge-object whose molecular elements can be combined and
recombined in pursuit of political projects. Its construction is also the construction
of the uniformity of the governed. Yet the core of Foucault's
argument is that there was an 18th century discovery of population. At times it
is medical authorities who become aware of this object; at times the object
bursts into political theory; at times it appears in France in the middle of the
eighteenth century; at times it appears in England later in that century. The key
claim for state formation, however, is that political authorities became aware
that the domain they sought to govern was characterized by intractable
regularities, about which Foucault writes naturalistically as demographic facts:
birth and death rates, age pyramids, rates of death and disease, and so on. The
shift to liberal government is propelled in his account by the political discovery
of the intractability of things: such intractability demonstrates the limitations
of police and pushes authorities to seek to govern liberally in keeping with the
nature of things.
The absence of population as a developed abstraction in eighteenth century
political thought and, more especially, practice is not simply a matter of
empirical interest. Rather, it strikes at the substance of the claim Foucault
makes about the origins of the 'governmentality.' Population cannot be 'discovered'
by political authorities, for its existence as a political abstraction
depends upon the work of a particular kind of sovereign political authority
itself. Population depends upon the establishment of equivalences among the
subjects within a particular territory. Political-scientific knowledge depends on
the discipline of potential objects of knowledge. It is only on the grounds of
constructed and enforced equivalences that one body comes to equal another,
that each death, birth, marriage, divorce, and so on, comes to be the equivalent
of any other. It is only on the grounds of such constructed equivalences that it
is possible for statistical objects to emerge in the form of regularities and to
become the objects of political practice. Population is coincident with the
effective capacity of sovereign authority to discipline social relations. The
intractability of things is inextricably tied up with the forms in which things are
known and the work done to make them knowable.
Recent scholarship on the history of statistics in France shows repeatedly
that it was the destruction of the status differences of the ancien regime that
made it possible first for the dream and then for the practice of population to
emerge. A further threshold of liberal governmentality is crossed when 'every
person identifiers' make it possible reliably to link observable regularities to
known individuals (Caplan and Torpey 2001). Joshua Cole, for instance, points
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
530 Canadian Journal of Sociology
to the significance of the French revolution's abolition of status differences
through the establishment of the etat civil: 'the principle of equality of membership,
once established in the etat civil opened the way for population
researchers to search for a new evaluation of every individual's function and
value to society' (Cole 2000:40). Again, one can contrast the persistent and
enduring problem faced by police in distinguishing individuals through
technologies of bodily singularities and status differences with the schemes
floated in the wake of the revolution for the construction of universal identifiers.
Thus, before the revolution, Bentham, for instance, mooted schemes for
tattooing individuals' names on their arms, for insisting people be called by
standard names, or that members of different occupational groups be required
to wear different costumes as a means of police (Bentham 1931). By contrast,
Vincent Denis discusses the detailed scheme of a certain Ducrest who, in 1804,
presented the Ministere de la Police G6nerale with a detailed plan for dividing
the national territory up into small units in which all houses would be numbered,
all individuals would be required during a census on a common day to
declare residence in a numbered house and thence to carry a numbered identity
card at all times (Denis 2000). The revolution made it possible for number to
take the place of status.
The generalization of every-person identifiers, such as national systems of
civil registration and nominal census enumerations, developed unevenly
internationally, but generally speaking, it is a comparatively recent phenomenon.
For instance, the first nominal census enumeration in England was in
1841; in Belgium in 1846; in the United States in 1850; in Canada in 1852; in
Italy after 1860. Canada had no national system of civil registration until 1921.
Political and administrative projects that embody individualizing and totalizing
initiatives tended to follow the mathematization of statistical investigation,
which is primarily a twentieth century phenomenon. While the political
purchase on social relations offered by every-person identifiers was already
evident in mid-nineteenth century experiments, such as the English use of civil
registers to demonstrate the aquatic transmission of the cholera, both insurantial
and prudential governmental technologies depend upon highly sophisticated
statistical modelling. The population of the governed depends on the
work of sovereign authority in creating equivalences.
Still, populations do not have to be composed of citizens and the authority
that creates equivalences does not have to be that of the state. One might
attempt to sustain Foucault's claims about a population 'discovery' on the part
of political authorities by arguing that the logic of population was worked out
at some local site or in some sector of practice and subsequently seized upon
by political theorists or administrators as a model. Perhaps his temptation in his
work on social medicine was to argue that medical authority in the institution
of the hospital constructed forms of equivalence that caused disease to emerge
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 531
as an entity and reality apart from the diseased body. Disease populations may
have shaped his thinking about other populations. On the other hand, especially
from the mid-1970s, Foucault wrote of population in increasingly naturalistic
terms, as a real entity characterized by resistant qualities. He seemed to
participate in the dominant usage in the social sciences which treats population
as a thing which awaits discovery and management. As Remi Lenoir has
fittingly remarked:
In effect, what appears in the form of a 'fact of population' or of a 'demographic structure' is the
result of a double social construction: demographic data are constructed in keeping with the
bureaucratic categories of moder states which aim to identify individuals and to define groups...and
the principle of construction of these categories is itself constructed as based in nature (Lenoir
2000:96).
If population is such a doubly constructed object, one whose naturalness
naturalizes the state, it certainly makes little sense to argue that the state
discovers it, and equally little sense to argue that political sociology can move
beyond the state by focussing on population.
References
Bentham, Jeremy
1931 The Theory of Legislation. London: Routledge & Kegan Paul.
Biller, Peter
2000 The Measure of Multitude: Population in Medieval Thought. Oxford: Oxford University
Press.
Caplan, Jane, and John Torpey, eds.
2001 Documenting Individual Identity. The Development of State Practices in the Modern
World. Princeton: Princeton University Press.
Cole, Joshua
2000 The Power of Large Numbers: Population, Politics, and Gender in Nineteenth-Century
France. Ithaca: Cornell University Press.
Coleman, William
1982 Death is a Social Disease: Public Health and Political Economy in Early Industrial
France. Madison: University of Wisconsin Press.
Curtis, Bruce
2001 The Politics of Population: Statistics, State Formation, and the Census of Canada,
1840-1875. Toronto: University of Toronto Press.
2002 "Surveying the social: Techniques, practices, power. Introduction." Histoire sociale/Social
History 35:in press.
Dean, Mitchell
1991 The Constitution of Poverty. Toward a Genealogy of Liberal Governance. London:
Routledge.
1994 Critical and Effctive Histories: Foucault's Method and Historical Sociology. London:
Routledge.
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
532 Canadian Journal of Sociology
Denis, Vincent
2000 "Entre Police et Demographie." Actes de la recherche en sciences sociales : 72-78.
Desrosieres, Alain
1991 "How to Make Things Which Hold Together: Social Science, Statistics and the State."
In Discourses on Society. The Shaping of the Social Science Disciplines, ed. P. Wagner,
B. Wittrock, and R. Whitley. 195-218. Dordrecht.
1992 "Discuter l'indiscutable. Raison statistique et espace public." In Pouvoir et legitimite.
figures de l'espace public, ed. Alain Cottereau, and Paul Ladriere. 131-54. Paris: Ecole
des Hautes Etudes en Sciences Sociales.
2001 "How Real are Statistics? Four Possible Attitudes." Social Research 68:339-355.
Donzelot, Jacques
1984 L'invention du social: Essai sur le ddclin des passions politiques. Paris: Fayard.
Dupaquier, Jacques, and Michel Dupaquier
1985 Histoire de la Demographie. La statistique de la population des origines a 1914. Paris:
Librairie Academique Perrin.
Dupont, Danica, and Frank Pearce
2000 "Foucault contra Foucault: Rereading the 'Governmentality' Papers." Theoretical
Criminology 5:123-158.
Foucault, Michel
1973 The Birth of the Clinic. An Archaeology of Medical Perception. New York: Vintage.
1979a Discipline and Punish: The Birth of the Prison. New York: Pantheon.
1979b "Governmentality." Ideology and Consciousness 6:5-21.
1980a "The politics of health in the eighteenth century." In Power/Knowledge: Selected
Interviews & Other Writings, 1972-1977, ed. Colin Gordon. 166-182. New York:
Pantheon.
1980b "Truth and Power." In Power/Knowledge: Selected Interviews & Other Writings,
1972-1977, ed. Colin Gordon. 109-133. New York: Pantheon.
1981 "'Omnes et singulatim': Towards a Criticism of Political Reason." In The Tanner
Lectures on Human Values, ed. S. McMurrin. 223-254. Salt Lake City: University of
Utah Press.
1983 "The Subject and Power." In Michel Foucault: BeyondStructuralism and Hermeneutics,
ed. Dreyfus H.L., and P. Rabinow. 208-226. Chicago: University of Chicago Press.
1988 "Technologies of the Self." In Technologies of the Self: A Seminar with MichelFoucault,
ed. Luther H. et al Martin. 16-49. Amherst: University of Massachusetts Press.
1989 Resume des cours, 1970-1982. Paris: Julliard.
1989
(1978) "Introduction." In TheNormalandthe Pathologicalby Georges Canguilhem. 7-24. New
York: Zone Books.
1991 "Governmentality." In The Foucault EfJect: Studies in Governmentality, ed. Graham
Bruchell, and et al. 87-104. Chicago: University of Chicago Press.
1994a "La <gouvemementalite>." In Dits et ecrits. 635-657. Paris: Gallimard.
1994b "La naissance de la medecine sociale." In Dits et ecrits, 1954-1988. 207-228. Paris:
Gallimard.
1994c "La politique de la sante au XVIIIe siecle." In Dits et ecrits, 1954-1988. 13-27. Paris:
Gallimard.
1994d "La politique de la sante au XVIIIe siecle." In Dits et ecrits, 1954-1988. 725-742. Paris:
Gallimard.
1994e "La situation de Cuvier dans l'histoire de la biologie." In Dits et ecrits, 1954-1988.
595-618. Paris: Gallimard.
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms
Foucault on Governmentality and Population 533
1994f "Les mailles du pouvoir." In Dits et ecrits, 1954-1988. 182-201. Paris: Gallimard.
1994g "L'incorporationde l'hpitaldans latechnologie moderne." In Ditset ecrits, 1954-1988.
508-521. Paris: Gallimard.
1994h "L'oeil du pouvoir." In Dits et ecrits, 1954-1988. 190-207. Paris: Gallimard.
1994i "Verite et pouvoir." In Dits et ecrits, 1954-1988. 140-160. Paris: Gallimard.
1994
(1978) "M6thodologie pour la connaissance du monde: comment se d6barrasser du marxisme."
In Dits et ecrits. 595-624. Paris: Gallimard.
Glass, D.V.
1973 Numbering the people. The eighteenth-century population controversy and the
development of census and vital statistics in Britain. Farnborough, Hants.: Saxon House.
Lenoir, Remi
2000 "Savoirs et sciences d'Etat: g6enalogie et d6mographie." Actes de la recherche en
sciences sociales: 96-97.
Miller, Peter, and Nikolas Rose
1993 "Governing Economic Life." Economy and Society 19:1-31.
Poovey, Mary
1995 Making a Social Body: British Cultural Formation, 1830-1864. Chicago: University of
Chicago Press.
1998 A History of the Modern Fact: Problems of Knowledge in the Sciences of Wealth and
Society. Chicago: University of Chicago Press.
Rosen, George
1974 From medicalpolice to socialmedicine: Essays on the history of health care. New York:
Science History Publications.
Tribe, Keith
1978 Land, Labour and Economic Discourse. London: Routledge and Kegan Paul.
This content downloaded from
86.49.251.249 on Mon, 21 Oct 2024 09:37:27 UTC
All use subject to https://about.jstor.org/terms