Bureaucratically split personalities: (re)ordering
the mentally disordered in the French state
Alex V. Barnard1
# Springer Nature B.V. 2019
Abstract
The ability to (re)classify populations is a key component of state power, but not all
new state classifications actually succeed in changing how people are categorized and
governed. This article examines the French state’s partly unsuccessful project in 2005
to use a new classification—“psychic handicap”—to ensure that people with severe
mental disorders received services and benefits from separate agencies based on a
designation of being both “mentally ill” and “disabled.” Previous research has identified
how new classifications can be impeded by cultural and cognitive barriers to their
adoption and struggles between professionals or administrators over their implementation.
Drawing on 186 interviews, archival sources, and 13 months of observations
across different French bureaucracies, I expand on this literature in two ways. First, I
use the case of psychic handicap to argue that a new classification can also fail to
achieve its intended effect when it constitutes a bureaucratically split personality—a
combination of classifications that imply that individuals belong to two, mutually
exclusive kinds of people. I show how psychic handicap embodied contradictory
expectations about the behavior, characteristics, and institutional trajectory of people
with mental disorders. Second, I identify how bureaucrats resolved these contradictions
through mechanisms of refractory looping, outsourcing expertise, and classification by
default, which in this case led to the reclassification of this population as simply
“mentally ill.” This framework calls attention to how practical inconsistencies can limit
the impact of new classifications, even absent overt resistance to their elaboration or
implementation.
Keywords Boundaryobjects.Bureaucraticallysplitpersonalities.Classificationstruggles
. The French state . Psychiatric diagnosis . Psychic handicap
https://doi.org/10.1007/s11186-019-09364-2
* Alex V. Barnard
avbarnard@nyu.edu
1
Department of Sociology, New York University, 295 Lafayette St., 4th Floor, New York,
NY 10012, USA
Theory and Society (2019) 48:753–784
Published online: 29 October 2019
In the government offices responsible for granting benefits and services to persons with
disabilities in France, people with severe mental disorders1
pose a continual classificatory
challenge. Looking at the dossier of a 26-year-old man with schizophrenia, one
evaluator remarks, “His medical certificate [written by a psychiatrist] says he has severe
symptoms, but on the next page, it’s written that he’s ‘autonomous in his activities of
daily life””—the latter being the primary basis for a disability determination. “In any
case,” another evaluator points out, “the certificate is old. He’s probably psychotic
again by now. We can’t refer him for professional retraining if he’s not stable.” A final
evaluator sighs, “He probably needs help with housing, but for some reason, he has
crossed out all those boxes on his application.” Here, an individual whom everyone
agreed was severely “mentally ill” was not recognized as “disabled,” despite the
benefits in terms of housing, employment supports, and aid in daily life such a dual
designation would bring.
These lower-level bureaucrats were attempting to apply the official classifications of
the French state, whose centrality in the “certification [and] consecration [of] legitimate
classifications” has attracted growing sociological attention (Fourcade 2009, p. 259;
see, also, Bourdieu 2015; Loveman 2014; Norton 2014). Scholars have shown, however,
how cultural or cognitive constraints (Mohr and Duquenne 1997; Steensland
2006) and professional or bureaucratic conflicts over jurisdictions and flows of resources
(Barman 2013; Goldberg 2005; Strand 2011) can create barriers to the adoption
and implementation of new classification schemes. In this article, I expand on this
literature in two ways. First, I elaborate another potential impediment, which emerges
when an attempt to recategorize populations administered by different wings of the
state creates a “bureaucratically split personality.” A bureaucratically split personality is
a classification that implies that the same people belong to multiple categories that
entail inconsistent behavioral expectations, principles of expert evaluation, and institutional
trajectories. Second, I show the mechanisms by which bureaucrats resolve these
unworkable and unstable combinations by favoring one side of the “split” over the
other.
Until 2005, the French state directed people with intellectual or physical disabilities
and those with mental illnesses into separate institutions, run by different sets of
professionals, under the aegis of different bureaucracies. In 2005, however, legislators
anointed a new classification, “psychic handicap.”2
It established that a “substantial,
durable, or definitive alteration of … psychic functions”3
caused by mental illness
could create a “restriction of activity or of life in society” and thus constitute a
“disability.” According to the President of the country’s organization for the families
of people with mental disorders (UNAFAM), this transformation was “truly revolutionary.”
People with severe mental disorders would continue to receive care as patients
1
I use “severe mental disorders” to refer to the disturbances of thoughts, emotions, and behaviors usually
associated with serious forms of conditions like schizophrenia, bipolar disorder, or major depression; these
disturbances are then typically classified as “symptoms” and labeled “mental illnesses.”
2
“Handicap” in French can be translated as either “disability” or “handicap.” For clarity, I use “psychic
handicap” (“handicap psychique”) to refer to the specific category advanced by advocates for people with
mental disorders and “disabled” to refer to the broader administrative classification that could contain people
with “physical,” “intellectual,” “sensorial,” or “psychic” handicaps. All translations are by the author.
3
République Française. 2005. Loi n°2005-102 pour l’égalité des droits et des chances, la participation et la
citoyenneté des personnes handicapées. Retrieved October 4, 2017 (https://www.legifrance.gouv.fr/).
Theory and Society (2019) 48:753–784754
in the public psychiatric system, but also be incorporated into the state’s broader
attempt to make people with disabilities into empowered actors capable of managing
an autonomous “life project” (see Power et al. 2014). Psychic handicap thus seemed
like a case of the dramatic shift in “biographical trajectory, memberships, or location”
created when a new classification enters the “process of making people” in state
institutions (Bowker and Star 1999, p. 223).
As the opening anecdote suggests, psychic handicap instead represents a case where
a new classification scheme does not achieve its intended effect. Ten years after its
codification in public policy, key actors were moving away from psychic handicap as a
lever for radically changing the state’s response to people with severe mental disorders.
They were instead proposing ways to advance individual autonomy and recovery
without a designation of psychic handicap.4
Compared to the United States, where
the idea that “mental illness” causes “mental disability” has long been recognized,5
people with mental disorders in France remain primarily in services and institutions that
serve people with illnesses (not disabilities) and under the jurisdiction of psychiatrists
(see Fig. 1).
This outcome is puzzling. The introduction of psychic handicap had the support of
the upper administration and civil society groups, overcoming the previous cultural and
administrative separation of “mental illness” and “disability.” Moreover, rather than
struggling over the new jurisdictions and flows of resources the classification would
create, professionals and bureaucrats recognized that—in a context where France’s
health system was facing budgetary austerity—a recognition of disability was crucial
for providing housing, work, and social supports for the severely mentally disordered.
In explaining the trajectory of psychic handicap as an instance of a bureaucratically
split personality, I offer an alternative explanation that focuses on how a new classification
scheme can be undermined by the unexpected but irreconcilable contradictions
created when multiple classifications are supposed to be applied to the same people.
Empirically, this article uses ethnographic observations in institutions tasked with
applying and acting on classifications of mental illness, psychic handicap, and disability.
I supplement this with 186 interviews with policymakers, advocates, medical
professionals, and bureaucrats as well as unpublished statistics, classification tools,
and archives. I start by outlining the parallel development of France’s systems for the
“mentally ill” and “disabled” and how the introduction of the category “psychic
handicap” in 2005 appeared well-poised to allow someone to be both. The second
empirical section analyzes, in turn, how psychic handicap emerged as a bureaucratically
split personality as professionals applied for its recognition, bureaucrats sought to
evaluate it, and disability service providers worked to integrate those so classified into
existing institutions. At each stage, incongruities between the expected passivity and
activity of people with mental disorders, the subjective or objective character of their
troubles, and the variability or stability of these limitations made the notion of psychic
handicap practically incoherent. Through three mechanisms, refractory looping,
outsourcing expertise, and classification by default, these contradictions resulted in
4
This shift is visible in the national ministerial strategy, which talks more about “avoiding” psychic handicap
than giving benefits based on it. See, Secrétariat d’État chargé des personnes handicapées, 2016, Stratégie
quinquennale de l’évolution de l’offre médico-social: Volet handicap psychique (Paris, France).
5
See, Comptroller General of the United States, 1977, Returning the Mentally Disabled to the Community
(Washington, DC).
Theory and Society (2019) 48:753–784 755
the return of responsibility for people with mental disorders to the health system, where
they had a single bureaucratic identity as mentally ill.
In the conclusion, I reflect on how bureaucratically split personalities can be used to
expand our understanding of the conditions that shape how official state classification
schemes enter bureaucratic practice and reshape the definition of populations. Distinct
resolutions of classificatory contradictions can result in profoundly different and
consequential distributions of people across state institutions.
Theoretical framework
Cultural constraints and classification struggles
Modern states govern their populations through applying a dizzying and expanding
range of classifications (Bourdieu 2015; Foucault 1991; Starr 1992). But while states
can “forcefully remake social relations” (Morgan and Orloff 2017, p. 11) by obligating
their subjects to engage with official categories on census forms, for welfare benefit
applications, or in courtrooms, the exercise of this classificatory power is neither
automatic nor straightforward. The British state struggled for decades to stymie piracy,
despite having the military might to do so, because it lacked the “cultural infrastructure”
necessary to constitute pirates as a distinctive “social object” (Norton 2014, p. 1551).
Doing so required finding a way of classifying pirates that resolved the contradiction
between pirates’ rights as English subjects and the need of state agents to identify,
judge, and execute them in a coordinated fashion quickly.
Existing scholarship, then, points to potential impediments to the introduction and
institutionalization of new state classification schemes. In their study of the poverty
relief organizations that constituted the American welfare state at the start of the
twentieth century, Mohr and Duquenne (1997) find a strict and consistent separation
between the organizations and services offered to different categories of recipients, like
Fig. 1 Medical and disability services for people with mental disorders in France and the United States.
Sources: OECD Health Statistics; Davis et al. (2012); Direction de la recherche, des études, de l’évaluation et
des statistiques, 2013, Les Établissements et Services Pour Adultes Handicapés (Paris, France)
Theory and Society (2019) 48:753–784756
the “worthy” and the “fallen.” Certain classificatory combinations may simply be
cognitively unthinkable: for much of US history, being recognized as a “woman” and
belonging to the category “soldier” were, in the eyes of policymakers and
administrators, mutually exclusive. Steensland (2006) examines what happens when
enterprising reformers trespass these cultural separations, as with a proposal in the
1970s in the United States to provide General Income Assistance to all poor people.
Opponents argued doing so would “symbolically contaminate” (2006, p. 1283) the
deserving working poor by placing them in the same category as undeserving welfare
recipients. Policymakers ultimately opted to create separate programs for people with
disabilities and low-wage workers, which fit with prior divisions in the population. In
short, existing state classifications create cognitive schema and cultural resources that
constitute “symbolic constraints” (Mayrl and Quinn 2016, p. 2) to new classifications.
As this research shows, new categories are not just about making sense of the world,
but also “sites of power relationships [and] political action” (Friese 2010, p. 148). A
related literature on “classification struggles” explores contestation over not just the
symbolic and moral implications of new classifications, but also their material ones
(Barman 2013; Goldberg 2005; Mora 2014). For example, the “mentally ill” only came
into being as a group once psychiatrists overcame the resistance of priests and judges,
who saw them rather as morally deviant or criminally dangerous (Abbott 1988;
Goldstein 1987). Today, classification struggles in the United States continue over
whether jurisdiction over the mentally ill belongs to psychiatrists, social workers, or
psychologists (Bosk 2013; Craciun 2016; Strand 2011). At stake in such struggles is
which professional group will benefit from the authority, resources, and prestige for the
management of the mad. Classification struggles are not limited to the upper echelons
of the state, where new classifications are elaborated and official criteria for their
application defined. A separate literature has looked at classifications as deployed by
“street-level bureaucrats,” the state agents who are in direct contact with citizens. While
nominally bound by sanctioned categories, these bureaucrats have significant discretion
in how those categories are used and in allocating resources attached to them (Lipsky
2010; Prottas 1979). Scholars have analyzed how bureaucratic resistance slowed
welfare reforms’ efforts to convert benefit “recipients” into “workers” in France and
the United States (Dubois 2003; Sandfort 2000; Watkins-Hayes 2009). Loveman
(2007) similarly shows how census workers in Puerto Rico employed their own
existing racial categories (based on appearance) in carrying out their surveys,
undermining the official racial classification (based on ancestry). Struggles might be
between bureaucrats and their superiors—for example, over a new classification that
assigns them responsibility for an undesirable clientele—or between bureaucratic
groups—as resource constrained agencies push clients onto one another (Lara-Millán
2014; Seim 2017).
These literatures provide useful guidance for identifying where the process of using
a new classification like psychic handicap to transform state interventions into the lives
of the mentally disordered might break down. A deeply engrained cultural divide might
make intermixing the “mentally ill” and “disabled” either unimaginable or morally
suspect. Existing cultural divisions could be mobilized in classification struggles by
professional groups threatened by the new repartition of resources and responsibilities.
On the ground, street-level bureaucrats might more subtly resist implementing the new
classification in an effort to shrug off a burdensome new obligation. Whether at the
Theory and Society (2019) 48:753–784 757
macro-level of policy, the meso-level of professions, or the micro-level of bureaucratic
practice, each calls attention to various kinds of resistance a new classification might
provoke.
From boundary object to bureaucratically split personality
This article examines the trajectory of a new state category from a slightly different
angle, emphasizing not struggles over classifications but rather contradictions between
them. At times, new categories are fabricated out of whole cloth and layered on top of
existing classifications, as was the case for “Hispanic” in the United States (Mora
2014). This article focuses more specifically on cases where a new classification is
intended to serve as a bridge through which previously mutually exclusive classifications
come to apply to the same people. The recognition of the autism spectrum by
health and human services agencies, for example, has taken children once called
“mentally retarded” and placed in specialized institutions and enabled them to receive
benefits on the basis of “disability” and access mainstream educational services as
“students” (Eyal 2013).
The literature on classification in science and medicine has analyzed such
recombinations as depending on what Star and Griesemer (1989, p. 409) call a
“boundary object.” Boundary objects are categories that “live in multiple social worlds”
and facilitate cooperation among them. A classification like the autism spectrum serves
as a boundary object because it is flexible enough to be used by multiple communities
(educators, psychologists, or autistic people themselves) without unilaterally imposing
one group’s understanding of that object on the others (Bowker and Star 1999, p. 297).
I conceptualize psychic handicap as an attempt to create a similar boundary object that
would bridge the previously partitioned worlds of mental illness and disability.
In practice, however, psychic handicap was a bureaucratically split personality: an
attempt to combine classifications that were, in practice, irreconcilable without a
fundamental alteration to one or the other. Each term in “bureaucratically split personalities”
captures a distinct feature of my approach. I use “bureaucratically” to differentiate
from the literature on cognitive or cultural constraints. Bureaucratically split
personalities are not so much “unthinkable” (Mohr and Duquenne 1997, p. 355) at
the level of policymaking but rather “unworkable” in practice. Actors may be in
agreement over what classifications should be used and how, but they may struggle
to adapt existing tools and procedures in ways that allow them to integrate them into
organizational processes. After all, “even the most dramatic change at the macro level
comes to nothing if it is not collectively embraced in practice by those who must do
their work in a new way” (Kellogg 2011, p. 7).
Second, I refer to “splits” to emphasize that bureaucratically split personalities stem
from institutional divisions between the “many hands” of the state (Joyce and Mukerji
2017; Morgan and Orloff 2017). Different agencies develop distinctive practices for
applying categories, expectations for those they classify, and processes for allocating
resources based on those classifications. Mol (2002, pp. 119, 138) points to how such
radically different visions (in her case, of patients and of pathologies) can co-exist in a
single organization as long as boundary objects remain “fuzzy” enough to allow
cooperation without standardization. This article, however, examines situations where
the meaning of psychic handicap needed to be precisely defined and applied. In these
Theory and Society (2019) 48:753–784758
moments, the distinctive visions of and expectations for the mentally ill and the
disabled developed by different wings of the French state snapped into sharp relief.
Such moments may create a process of “splitting” by which previously latent “perceived
gaps” between classifications “widen,” “thereby reinforcing [the] mental separateness”
(Zerubavel 1996, p. 424) of populations that a new classification was
supposed to bring together.6
Finally, I use the term “personalities” to highlight that bureaucratic splits are most
likely to be problematic when classifications imply that one is a certain kind of person.
Hacking (1995, pp. 352, 360) describes “human kinds” as population groups about
which experts claim “systematic, general, and accurate knowledge” that provides
“principles through which [to] interfere, intervene, and improve.” A child can carry
multiple psychiatric diagnoses—like “oppositional defiant disorder” and “ADHD”—
without any contradiction. Both place them within a single human kind, “mentally ill,”
and imply a common set of interventions. But, as Eyal (2013) shows, diagnoses of
“autism” and “mental retardation” are profoundly incompatible, because they imply
different life trajectories, principles of expert intervention, and institutional homes.
Discrepancies such as these, which are characteristic of classification schemes that I
am characterizing as bureaucratically split personalities, are not easily reconciled with
simple changes to diagnostic criteria or classification procedures.
Bureaucratically split personalities in practice
In this section, I consider in greater detail what differentiates a new state classification
that serves as a boundary object from one that becomes a bureaucratically split
personality, drawing on literatures that analyze applying to be classified, evaluating
eligibility for a classification, and institutionalizing that classification into organizational
processes. I then build on the literature on street-level bureaucrats, which already
identifies a range of “coping behaviors and adaptive attitudes” (Lipsky 2010, p. 181)
that state agents deploy when faced with contradictory mandates or inconsistent policy.
As Lipsky (2010, p. 181) observes, these studies often leave unspecified the “orientation
of adaptive attitudes”—that is, how these behaviors aggregate into specific outcomes.
I theorize three mechanisms—refractory looping, outsourcing expertise, and
classification by default—that link individual responses to a bureaucratically split
personality to the broader structure of the state, resulting in particular resolutions of
the “split.”
Applying—State classifications vary in the extent to which they can be unilaterally
imposed: states can conscript “soldiers” but generally not “patients.”7
Classifications
that require some cooperation, then, must contend with what Hacking (1998, p. 21)
calls “looping.” People “tend to conform to or grow into the ways they are described”
in administrative or professional classification schemes, but this is in part because they
themselves act in ways that change those categories. Paradoxically, this dynamic
engagement actually makes those categories seem more natural and entrenched, as
6
Maryl and Quinn (2016, p. 5) refer to this as “boundary activation” (see Lamont and Molnár 2002).
7
Forced treatment for mental illness is a notable exception.
Theory and Society (2019) 48:753–784 759
when racial classifications added to censuses became the basis for political mobilization
in Latin America (Loveman 2014).
This article considers how the process of applying one state classification might
affect the other classifications previously applied to the same people. That people might
“loop” with respect to multiple classifications simultaneously is latent but not explicitly
analyzed in existing studies. Hacking’s (1998) study of trauma victims shows how the
way women responded to talk therapy but not to medication seemed to confirm that
they had Multiple Personality Disorder—which meant that they did not have schizophrenia,
as many were previously diagnosed. I call refractory looping the process by
which a population “growing into” one category implies them “growing out” of
another.8
As I show, psychic handicap implied contradictory expectations for the way
people would apply for benefits. Over time, the criteria for psychic handicap changed in
a way that made it more synonymous with mental illness and clarified the meaning of
disability in a way that increasingly excluded the mentally disordered.
Evaluating—A new classification scheme has little meaning unless some bureaucratic
agency or professional group has jurisdiction for assigning it (Barman 2013;
Kellogg 2014; Mora 2014). As Porter (1996) shows, the jurisdiction granted by states
to bureaucrats increasingly comes with the expectation of “objectivity;” that is, that
decisions stem from rules that appear impersonal, reliable, and publicly defensible (see,
also, Bracci and Llewellyn 2012; Sandfort 2000). Psychiatrists, for instance, have
responded to challenges to their jurisdiction by adopting new tools, criteria, and
symptom scales intended to show the scientific basis of diagnosis (Horwitz 2001;
Strand 2011; Whooley 2016). This article analyzes how the professionals evaluating
psychic handicap were supposed to “link together” (Eyal 2013, p. 864) a set of
information about mental illness in a person’s application from the health system and
evaluation tools from the disability system in order to perform an expert, objective
evaluation (see, also, Berg and Bowker 1997; Mol 2002).
Psychic handicap was a bureaucratically split personality insofar as these two were a
poor match for one another. Evaluators did not, however, respond by “burden shifting”
(Seim 2017) responsibility for psychic handicap onto other agencies or “rubber
stamping” (Lipsky 2010, p. 130; Prottas 1979, p. 68) psychiatric evaluations as their
own, as the literature on classification struggles in bureaucracy suggests. Instead they
outsourced expertise, achieving a semblance of objectivity by incorporating measures
that were not supposed to be part of their evaluation. This collapsed one classification
into another rather than linking them through a boundary object. Showalter (2019)
shows a similar process in how, in the early twentieth century, psychiatrists in New
York prisons defined “psychopaths” using behavioral criteria produced by the justice
system. Outsourcing expertise in this way meant that re-classifying people as psychopaths
did little to transform them from being a “criminalized” to a “medicalized”
population.
Institutionalizing—The application of a classification has concrete meaning only
once it ties individuals into a broader infrastructure that acts upon that label (Bowker
and Star 1999, p. 319). Public health surveys that identify many US prisoners as
8
Navon and Eyal (2016, p. 1426) describe how looping with respect to autism “disabled” other interpretations
of children’s condition. This terminology would be confusing in this article, and those authors’ use of the term
does not consider competing classifications per se.
Theory and Society (2019) 48:753–784760
mentally ill, for example, have little import so long as inmates remain in prisons
without treatment. When the goal of a new classification is to help the classified
population enter into extant institutions, the key bureaucratic task is one of “commensuration”:
showing that new individuals are comparable to existing ones along some
relevant common metric (Espeland and Stevens 1998; Lakoff 2005; Lamont and
Molnár 2002). One successful example was the use of IQ, which allowed parents in
the United States to claim that their “learning disabled” children were close enough to
“normal” to enter schools; the lack of a common measure in Germany left similar
individuals in segregated institutions (Powell 2010).
A boundary object implies that a person has characteristics commensurate with the
populations served by multiple sets of state institutions. When contradictions in these
expectations emerge and a bureaucracy rejects someone as “incommensurate,” that
person does not necessarily disappear into a classificatory vacuum. They instead tend to
fall back into programs with more flexible criteria, which I call classification by default.
In the United States, people who do not fit the increasingly narrow eligibility criteria for
welfare benefits on the basis of being “poor” wind up receiving services from emergency
health, disability, or penal institutions that serve broader categories of persons
(Hansen et al. 2014; Lara-Millán 2014; Seim 2017). This mechanism thus calls
attention to how the resolution of bureaucratic “splits” depends on the mix between
universal and targeted programs in different wings of the state.
Bureaucratically split personalities do not end in the wholesale rejection of a new
classification as symbolically contaminated or through the triumph of one group or
another in a classification struggle. Rather, this framework focuses on how, as a new
classification scheme is applied, evaluated, and institutionalized, contradictions within
that scheme can lead to refractory looping, outsourcing expertise, and classification by
default. These processes gradually undermine its continued use.
Data and methods
Data for this article were collected as part of a broader project on public institutions
tasked with managing people with severe mental disorders in Paris, France. Through
preliminary interviews and documents, I identified key sites in the health and disability
sectors where consequential decisions in the lives of people with mental disorders are
made (see Fig. 2). This process frequently started in a Medical Psychological Center
(“Centre médico psychologique,” hereafter “Psychiatric Clinic”). Over the course of
7 months in 2016, I observed weekly triage meetings in which nurses would present
new cases to psychiatrists, who would determine what level of psychiatric care each
patient would be given. I also attended discussions among nurses, psychiatrists, and
social workers of difficult cases who might apply to be recognized as disabled based on
having a psychic handicap.
All applications for disability benefits were then processed by a multi-disciplinary
team of generalist doctors, psychologists, and social workers in a Departmental Home
for Disabled Persons (“Maison départementale des personnes handicapées,” hereafter
“Disability Office”). I spent 6 months (also in 2016) observing evaluations of new
demands, meetings to discuss employment of disabled persons, and sessions of the
departmental commission responsible for making final determinations on complex
Theory and Society (2019) 48:753–784 761
cases and appeals.9
The Disability Office would then orient applicants towards providers
of professional insertion, protected work, aid in daily life, or long-term housing,
with whom I conducted interviews and site visits.10
In addition to my asking for approvals from the directors of the Clinic and Office, as
well as from the teams I was directly observing, this research was approved by the
University of California, Berkeley Committee for the Protection of Human Subjects. In
these settings, collective decisions are made outside of the presence of the applicants;
the primary research subjects of this project were professionals, not patients/clients. I
avoided recording any potentially identifiable information on the latter group. I show
how people with mental disorders engaged with the category of psychic handicap
through looking at bureaucratic traces and reports of their behavior, rather than direct
interviews.11
I contextualized my observational data from the Clinic and Disability Office through
186 qualitative interviews with professionals and policymakers. Among interviewees,
38 were involved in direct service provision in the health sector and 47 in the disability
sector; questions focused on classificatory decision-making and their appropriation of
Fig. 2 The French health and disability systems and intended impact of the 2005 reform
9
I also obtained unpublished statistics from the Disability Office, which show that the cases I directly
observed were representative of the broader range of demands treated by the Disability Office and of their
outcomes.
10
Direct observation of decision-making over access to downstream services for housing and employment
was not practical, because many structures (such as a fifty-bed nursing home) would only admit one or two
new people per year.
11
This focus on observable responses to a classification rather than phenomenological experiences of being
classified is consistent with other studies on “looping” (Eyal 2013; Navon and Eyal 2016).
Theory and Society (2019) 48:753–784762
official regulations and tools. I conducted interviews on the production of these regulations
and tools with 25 people involved in public policies for health and 24 in
disability. This included nearly all the key figures in the relevant national ministries
who worked specifically on mental illness or psychic handicap, as well as those with the
same responsibilities in regional and departmental agencies around Paris. The remainder
were representatives of key advocacy organizations, professional groups (such as the
psychiatrists’ unions), research institutes, and the legal sector. Among the total, 50 were
trained as psychiatrists, 39 as nurses or psychologists, 21 as social workers or special
educators, 17 as general or public health doctors, 17 as lawyers, and 31 as administra-
tors.12
Interviews lasted between 30 min and 2 h. I also examined official government
reports, directives, classification scales and other tools, unpublished policy evaluations,
and the archives of the Ministry of Health (which, to emphasize their separation from
published academic sources, I reference in the footnotes).
I developed a coding scheme iteratively, starting by differentiating the various kinds
of demands faced by the Psychiatric Clinic and Disability Office, then considering the
range of possible responses to those requests. I subsequently elaborated codes that
captured the reasoning professionals used to justify those decisions. I focused on 296
decisions at the Disability Office where decisions were made over the allocation of
benefits or services, in particular the 206 cases in which someone did not receive at
least one of the benefits for which they applied. In addition to the documentary
evidence, I coded 9000 separate units of texts from field notes and interviews in the
software Dedoose.
Paris is undoubtedly a distinctive case within France. It is also the place where the
introduction of psychic handicap would be most likely to succeed. The key actors that
pushed psychic handicap (government ministries and advocacy organizations) are
based in Paris. As many informants insisted, actors making classificatory decisions in
Paris are thus subject to much greater oversight and scrutiny in their implementation of
national policy. Moreover, services in both the disability and health systems are wellfunded
and widely available, suggesting that the limited impact of psychic handicap
was not just a question of resources. Interviews and archives, combined with several
weeks of comparative observations in a separate Disability Office outside of Paris and a
different clinic and hospital, increased my confidence that the patterns I describe are
reflective of the general bureaucratic challenge of using psychic handicap in the way
policymakers intended.
Broadly, this project follows the extended case method (Burawoy 1998) in
which the researcher attempts to understand concrete situations by looking outward
to structural forces and backwards to the history weighing on those situations. My
goal was thus to put theories of classification struggles, street-level bureaucracy,
and medical decision-making and diagnosis into dialogue with my observations.
The single case of psychic handicap cannot establish the necessary conditions for
the success of a new classification or determine what would happen if a “split
personality” appeared at the level of applying categories but not institutionalizing
them. My goal, instead, is to elaborate theoretical mechanisms that can be applied
and tested in other empirical situations (see Hirschman et al. 2016).
12
The remainder, mostly in advocacy organizations, had educational backgrounds not directly related to
health, disability, or public policy.
Theory and Society (2019) 48:753–784 763
The politics of psychic handicap
Separating and recombining mental illness and disability
This section considers the separate development of France’s health and disability
systems and the divergent meanings they attached to “mental illness” and “disability.”
Driven by the scandal of tens of thousands of asylum residents dying of hunger during
World War II in institutions that were more holding-areas than hospitals, the post-War
period saw a significant medicalization of social responses to mental disorders. In 1960,
the state cut France into “sectors” of 70,000 people and assigned a public psychiatric
team to each, responsible for providing universally accessible and free “prevention,”
“cure,” and “re-insertion” for the mentally ill.13
In 1968, psychiatry was recognized as a
medical specialty (Pinell 2004) and in 1971 psychiatric hospitals were designated “acute
care” institutions subsumed under the regulations governing general hospitals. Nonetheless,
at the end of the 1960s, the boundaries of psychiatry’s jurisdiction remained
fluid: in addition to people with iconic mental disorders like schizophrenia or depression,
psychiatrists were also variously tasked with addressing mentally retardarded and
“senile” persons, adult “marginals,” and “inadapted children” (Chauvière 1980).
Psychiatry’s domain narrowed significantly in the 1970s. The 1975 Law in Favor of
Handicapped Persons brought together two disparate sets of institutions: centers to
retrain “invalid workers” and “wounded soldiers” as a matter of national solidarity, and
residential schools created by the parents of “deficient children” as a matter of charity
(Barral 2007, pp. 217–220; Winance et al. 2007, pp. 165–166). As the Health and
Social Affairs Minister14
declared, the law “created a true social statute” (quoted in
Chapireau 2016, p. 5) for a new, state-recognized kind of person. These people with
disabilities would receive non-medical interventions from protected workshops, supported
housing, and services for aid and accompaniment in daily life. The law drove the
removal of people with intellectual disabilities and the dependent aged from psychiatric
hospitals.
The reform sparked a classification struggle over whether the “mentally ill” could
also be “disabled,” which highlighted the different meanings of the two categories. The
dominant image of a handicapped person was of someone with clearly-identifiable
incapacities that would be stable over time. For the psychoanalytic currents dominant in
French psychiatry, lumping the mentally ill with the disabled was a form of unacceptable
“symbolic contamination” (Steensland 2006, p. 1283). They saw mental illness as
a troubled subjectivity that had a dynamic trajectory that could be treated and transformed,
in contrast to a permanently broken body or brain (Lanteri-Laura 1972). As
one psychiatrist active at the time told me, “The handicapped are people who have no
more need for medicine, but just assistance and a bed. For psychiatrists to say, ‘We’re
not going to treat these people anymore,’ that’s intolerable.” As the literature on
professions and jurisdictional conflicts would predict, the struggle also had a material
dimension. Psychiatrists’ unions feared that relabeling people as “handicapped” was a
13
See, Ministère de la santé publique, Circulaire du 15 mars 1960 relative au programme d’organisation et
d’équipement des départements en matière de lutte contre les maladies mentales (Paris, France).
14
Both “disability” and “health” are covered by the (frequently renamed) Ministry for Health and Social
Affairs, but in separate directions. For simplicity, I speak of the “health” and “disability” ministries.
Theory and Society (2019) 48:753–784764
“modality of escaping the expensive designation of being a ‘sick person’” (Ayme 2002,
p. 185).
The state sided with psychiatry, concluding that insofar as the law on handicap was a
“correction destined … to cover up the holes in the social protection system,”15
the
mentally ill did not need to be included, precisely because they were already protected
by the French welfare state on the basis of being “mentally ill.” In the ensuing decades,
mental illness and disability were “mutually exclusive” categories both in administrative
practice and national discourse (Henckes 2011, p. 8). The disability and health
systems had distinct financing streams, regulatory agencies, and logics of social
protection (see Table 1). Meanwhile, psychiatrists had near-undisputed authority to
classify illness and intervene in the lives of the mentally disordered: as one researcher
explained, psychiatric secteurs developed “therapeutic work, therapeutic lodging, therapeutic
activities … but all around treating ‘our patients.’”
Growing budgetary pressures in the 1990s made a revision to this classification
scheme seem increasingly necessary. Reports from the period identified many psychiatric
hospitalizations as “inadequate” because stabilized patients who no longer required
intensive medical care stayed for lack of a supported living situation outside.16
Although like most Western countries France was closing hospital beds by the 1970s,
in the 1990s the transfer of funds was particularly stark: the state closed 33,000 beds in
psychiatry and opened 30,000 in publicly-funded, privately-run supported housing
establishments for the disabled (see Fig. 3). With no official recognition that mentally
ill people could be disabled, however, their access to these new structures was limited.
This exclusion was increasingly problematic because these same fiscal pressures meant
that psychiatric teams were abandoning activities like managing apartments or helping
patients find work that did not fit into a progressively more biomedical definition of
treatment.
In 2001, the association of parents of the mentally ill announced a desire to “render
existent a population” and “make real” that population’s rights within the disability
system through the recognition of psychic handicap.17
The organization recognized not
only the shift in financing from the health to disability system, but also strengthening
European and international norms that promoted the rights of people with disabilities
(Waldschmidt 2009). Thus, when the organization’s Vice-President described the initiative
as a “demand to make a category of persons that didn’t exist before,” she meant
that the new classification entailed not just new services, but also a new “personality” for
the mentally disordered as active players in increasingly autonomous lives.
Thanks in part to the organization’s lobbying, President Chirac called for public
action to “remediate the difficult situation … [of persons] with psychic handicaps, who
are currently without any solution.”18
This official recognition of this population was
rolled into a 2005 law for the “equality, citizenship, and participation” of the disabled.19
15
See, Ministère de la solidarité nationale, 1983, Bilan de la politique en direction des personnes handicapées
(Paris, France).
16
See, Direction de l’hospitalisation et de l’organisation des soins, 2008, “Éléments d’analyse des
inadéquations de prise en charge en hospitalisation complète,” DGOS 2014 / 011 / 3.
17
See, UNAFAM. 2001, Le livre blanc des partenaires de Santé Mentale (Paris, France), p. 12.
18
See, Chirac, Jacques, 2002, “Discours devant le Conseil national consultatif des personnes handicapées”
(Paris, France), retrieved August 19, 2016 (http://discours.vie-publique.fr/).
19
See, République Française, Loi n°2005-102, article 3.
Theory and Society (2019) 48:753–784 765
It recognized that mental illness could create a “disability” insofar as it produced a
“restriction of activity or a restriction of life in society” as a result of an alteration of
“psychic functions.”
Psychic handicap as boundary object
After the 2005 law, UNAFAM changed its name to the “families of sick and/or
psychically handicapped people.” This suggested their vision of psychic handicap as
a boundary object: it allowed the mentally disordered to exist in the bureaucratic worlds
of mental illness and disability simultaneously and created “durable cooperation”
between them (Bowker and Star 1999, p. 297). One advocate explained their position
at the time: “we don’t refuse psychiatry. We know that our family members need
Table 1 Overview of France’s health and disability systems
Health Disability
Dominant profession Psychiatrists, Nurses Administrators, Educators, Psychologists
Organization type Public Private Non-Profit
Funding Central State, National Insurance Sub-National Departments, Central State
Access Universal; all people in a given Psychiatric
Clinic’s territory (with or without
self-identification as “mentally ill”)
Targeted; accessible only to applicants
who self-identify with and are evaluated
as “disabled” by the Disability Of-
fice
Key criterion Subjective symptoms; disturbances of
thoughts, emotions, or behaviors
Objectively-determined limitations in
daily life or participation in society; the
life project of a person with a disability
Regulating body Directorate of Health Directorate of Social Cohesion
Cost per day €900 (Psychiatric Hospital) €240 (Nursing home)
Fig. 3 The decline in public psychiatric beds and rise in dedicated disability services in France (1960-2013).
Source: Institut de recherche et de documentation en économie de la santé, Statistical Yearbooks
Theory and Society (2019) 48:753–784766
psychiatric care…. We’re not hostile to them.” At the same time, though, UNAFAM’s
Vice-President insisted that “for these citizens” with a psychic handicap, “everything
like lodging, activities, help with employment—they’re going to be accompanied by
people not from the health sector. Not psychiatrists, not nurses, but social workers and
educators.”
Numerous factors appeared to favor the success of this new classification system in
transforming the institutional trajectories of people with mental disorders. The centralized
French state used its power to “produce and canonize … social classifications”
(Bourdieu 2015, p. 9; see, also, Carson 2007; Fourcade 2009; Porter 1996) to support
this new scheme. The official responsible for implementing the 2005 reform stated that
“psychic handicap will be the marker of whether our disability policies have
succeeded.”20
The President in 2014 declared it a “national priority.”21
A ministerial
delegate for handicap insisted, “Clearly, at the level of people with disabilities, today,
for the government, a big preoccupation—for the minister and her cabinet—is people
with psychic handicaps.”
In stark contrast to 1975, medical professionals largely did not “struggle” to prevent
the mentally ill from being classified as disabled. Two major psychiatric professional
organizations actually co-signed UNAFAM’s original call. One psychiatrist who conducted
trainings around the country stated, “We used to hear, ‘We’re not going to dump
our patients in the disability system.’ Not anymore.” Of fifty psychiatrists I interviewed,
only two expressed reluctance to call their patients “psychically handicapped.” While in
the past psychiatrists feared contaminating the mentally ill with the hopeless-sounding
label of disability, they now recognized that disability services offered their patients the
best hope for “professional and social integration.”22
They further accepted that their
jurisdiction over the new classification was limited: their role was to provide a medical
certificate confirming someone’s illness, which would accompany a written “life
project” and application prepared with the help of a Clinic social worker. The expert
determination of eligibility on the basis of a psychic handicap would fall to the
Disability Office.
Finally, architects of the law had a vision for how mental illness and disability could
be bridged at a bureaucratic level (see Fig. 2). Multi-disciplinary teams at the Disability
Office would use standardized scales and evaluation tools to translate the mental
symptoms characterized by psychiatrists into objective, functional impairments. The
Disability Office would then convert these impairments into “limitations in everyday
life,” which would provide a stable rendering of psychic handicaps. This would make
this population commensurable with those with physical, intellectual, or sensorial
handicaps and thus admissible to institutions serving people with disabilities. Advocates
further believed that the appeal of these benefits and services, increasingly only
available through the disability system, would encourage applications from people with
psychic handicaps. In so doing, they would join other people with disabilities in
20
See, Fédération d’aide à la santé mentale, 2016, “Égalité, citoyenneté et handicap psychique,” Pratiques en
santé mentale 1(February), p. 84.
21
Présidence de la République, 2014, Conférence nationale de handicap: relève des conclusions (Paris,
France), Service de presse, retrieved June 26, 2016 (http://www.elysee.fr/).
22
Massé, Gérard, 1992, La psychiatrie ouverte: une dynamique nouvelle en santé mentale: rapport (Paris,
France), Ministère de la santé et de l’action humanitaire, p. 234.
Theory and Society (2019) 48:753–784 767
becoming active participants and self-advocates in their own institutional trajectory
(Winance et al. 2007).
The law of 2005 did not just introduce a new category, then, but sought to create a
new human kind: a “person with a psychic handicap.” The prior symbolic distancing of
disability and mental illness would be broken down by a boundary object, facilitating a
shared understanding of the limitations and challenges caused by mental disorders (see
Star and Griesemer 1989, p. 393). Rather than provoking a classification struggle, this
new scheme was backed by civil society organizations, professional groups, and the
state; and the scheme appeared poised to enter bureaucratic practice.
Psychic handicap as a bureaucratically split personality
Application: passive patients versus self-advocates
The Law of 2005 envisioned that disability benefits would be granted based on the
“aspirations of the handicapped person as expressed in their life project,”23
a document
that would be included in their application to the Disability Office. As I show in this
section, however, the expected active self-advocacy of disabled persons was in contradiction
with the passivity and recalcitrance of people with mental illness.
A majority of the most severe patients served by the Psychiatric Clinic arrived
through a hospitalization. Half of these hospitalizations were legally imposed without
the consent of the person, who was deemed unable to recognize his or her own need for
care (see Gong 2017). Social workers and psychiatrists who wanted their patients to
receive disability benefits subsequently spent significant time cajoling obstinate patients
to provide the minimum information necessary to make an application. One
psychiatrist described:
To do a dossier for the [Disability Office], the person needs to be in agreement
and for them to be in agreement, they need to be conscious of their handicap. And
in fact it’s variable with patients, because there are some who are conscious right
away that they’re not going to get there and that they can’t get there, and then
others not at all. I have a patient who is twenty-five, it’s his second hospitalization
in two years, he’s had electroshocks, treatments that are very heavy—and he
refuses the demand for handicap. And he tells me, several times, ‘If I accept, that
means I’m sick, that means I’m crazy, and there’s nothing left but to die.’ So what
do you want me to do? For us, there are the criteria, for him, there aren’t.
I once posited that perhaps manipulative people might come to the Psychiatric Clinic in
the hopes of getting a medical certificate that would allow them to get the Handicap
Allocation, which at €800 a month is €300 more than France’s general minimum
income. A social worker brushed me off: “I basically do handicap demands for people
who don’t think they are handicapped.”
23
République Française, Loi n°2005-102.
Theory and Society (2019) 48:753–784768
My observations of the social workers’ weekly meetings at the Clinic confirmed that
people often evaded the label of psychic handicap to their obvious detriment. Some
patients remained involuntarily hospitalized because they would not apply to a lessrestrictive
nursing home that only accepted voluntary residents. While scholars have
pointed to how fear of stigma can prevent people from applying for benefits to which
they are entitled (Hasenfeld et al. 1987; Soss 2000), advocates in UNAFAM had
presumed—probably correctly—that “handicapped” would be less stigmatizing than
“mentally ill” (see Liu et al. 2010, p. 1393). What they missed was that “mentally ill”
could be imposed on a person regardless of the stigma attached to it, while “psychic
handicap” required a degree of cooperation and acceptance that was frequently lacking.
Even when patients were willing to support an application, they did not engage with
the category of psychic handicap in a way congruent with the model personality of a
person with disabilities. As one social worker laughed, “People don’t give a shit!—for
them, it’s completely paralyzing to ask them what their ‘life project’ is. Often it’s,
‘survive with 800 euros a month’ and so we just leave it blank. What’s the point of
writing, ‘Be in care and eat’ as a ‘project’?” Another averred, “Their social demands are
often not very linked with the state where they are. Some tell me, ‘I’m ready to live
alone, I can manage without psychiatry, I’m doing very well, thank you very much.’”
Rather than increasing their patients’ agency, for these professionals the goal was to
render the mentally ill passive enough to accept the supports they needed but which the
health system could no longer provide.
Psychiatric Clinic professionals’ interactions with the Disability Office highlighted
how the professionals expected very different behaviors from the same people, even
when a shared recognition that they were dealing with the psychically handicapped
should have served as a boundary object. One Disability Office official described,
“Once, we actually did a visit to the home of someone whose application said she was
‘autonomous in everything, handles eating fine, everything is fine, she just needs a little
income’ and we discovered that all she ate—breakfast, lunch, and dinner—were cans of
sauerkraut.” This behavior, she noted, was congruent with “mental illness” but not
“disability”: “For the nurse from the Psychiatric Clinic, that was fine—‘at least she’s
eating!’—but for us, it was like, ‘maybe this person could use some help making better
choices.’”
The 2005 law envisioned that, after participating in the preparation of an application,
an applicant’s involvement would continue at the Disability Office. As one official in
the Ministry explained, “What has really changed [with the law of 2005] is that our
evaluations are participative. The person is at the center, and we do everything as a
function of their life-project.” Yet even when convinced to apply, people recently
released from the hospital would still come to information-gathering interviews at the
Office and declare “all is well.” The head of the National Federation of Patients in
Psychiatry told me: “When we have a meeting at the Disability Office, we’re going to
wash up, put on a tie, and say, ‘I can do everything’ and the evaluator is never going to
ask, ‘Is he saying this just to show he’s still human, or because it’s really true?’”
When the time came to create a “personalized plan of compensation” on the basis of
the life project and medical certificate, the Disability Office was only allowed to make
decisions on the specific benefits and services that the person had asked for, not what
they thought might be most appropriate. In the case of a 32-year-old man diagnosed
with severe bipolar disorder, the head of the Professional Insertion Team asked,
Theory and Society (2019) 48:753–784 769
“Concretely, is this someone who can work in a regular company?” The person who
had reviewed the dossier replied quickly, “No way. But he crossed out the box for a
protected workshop.” “And,” added a psychologist who had interviewed him, “It’s a
waste of time to even mention it to him.” Indeed, in 19 of the 77 cases of people with
severe mental disorders that I saw adjudicated in the Disability Office, reference was
made to either the person being in “denial” or having an “inadapted” life project. The
latter might refer to someone with behavioral issues or a history of hospitalizations
asking to live in a facility with the elderly or a person on numerous anti-psychotic
medications requesting to operate heavy machinery.
The result of the growing awareness that psychic handicap was a bureaucratically
split personality was a kind of refractory looping, by which psychic handicap was
redefined as more and more a synonym of mental illness and further from disability.
Even when psychiatrists couldn’t see it, social workers in the Clinic recognized how the
Disability Office had come to expect not just people who self-advocated, but people
who accurately understood their limitations and asked for the “right” benefits. This
excluded their clients. As one social worker laughed, “Oh, here’s a good one—they [the
psychiatrist] wants us to send a new application for [patient name]. Do they remember
that she was banned from the Disability Office for threatening an evaluator who told
her she was disabled?”
As theories of “looping” suggest (Hacking 1995; Loveman 2014; Navon and Eyal
2016), the ways that people with severe mental disorders did (or did not!) engage with
the category ultimately altered its official meaning. Two guides on working with and
evaluating people with psychic handicap published in 2016 and 2017 noted an “absence
of a demand”24
and “lack of consciousness of needs”25
as key components.
Hypothetically, the disability system as a whole could be changed to account for this,
allowing for legally mandating people with mental disorders to be evaluated by the
Disability Office or to live in disability institutions like nursing homes (as is possible in
the United States). But doing so would directly contradict the work of associations of
handicapped persons to differentiate the disability system from the health one: as one
proudly stated, in France “care without consent exists, but not [disability] compensation
without consent.”26
Instead, some associations developing innovative programs to encourage
employment and provide housing for the mentally disordered were “looking
for ways to promote access without an application to the Disability Office,” as
one director explained. Although many of these programs still used the term
psychic handicap,27
they sought to return financing for these services to the
health system. In so doing, they would evade the classification scheme created
24
L’agence nationale de l’évaluation et de la qualité des établissements et services médico-sociaux, 2016,
Spécificités de l’accompagnement des adultes handicapés psychiques (Paris, France), p. 27.
25
Caisse nationale de solidarité pour l’autonomie, 2017, Troubles psychiques: Guide d’appui pour
l’élaboration de réponses aux besoins des personnes vivant avec des troubles psychiques (Paris, France), p.
16.
26
Caisse nationale de solidarité pour l’autonomie, 2009, Handicaps d’origine psychique: Une évaluation
partagée pour mieux accompagner les parcours des personnes (Paris, France), p. 64.
27
Hardy-Baylé, Marie-Christine, 2015, Données de preuves en vue d’améliorer le parcours de soins et de vie
des personnes présentant un handicap psychique, Centre de Preuves en Psychiatrie.
Theory and Society (2019) 48:753–784770
by the law of 2005, relying instead on a psychiatrist’s evaluation of mental
illness—made with or without the cooperation of the patient.
Evaluation: subjective symptoms versus objective limitations
Even when applications reached the Disability Office, the evaluation of psychic
handicap further detached it from the aspiration that it would serve as a boundary
object between two worlds in equal partnership. Previous research on France’s Disability
Offices has shown how they struggle with the classic challenges of street-level
bureaucracies: a seemingly-impossible mandate to offer individualized evaluations,
despite limited personnel and time, in the face of an avalanche of demands (Baudot
and Revillard 2015; Bertrand et al. 2014; Bureau et al. 2013). In my observations, the
mix of generalist doctors, psychologists, and special educators there did not abdicate
their role in evaluating psychic handicap through directly “burden shifting” responsibility
back onto the health system. At one training at the Disability Office, new
evaluators eagerly asked questions about the meaning of psychic handicap and nearly
half the time was spent explaining how to qualify people with mental disorders as
disabled. In a 2014 survey, 87% of departmental Disability Offices indicated that
psychic handicap was the disability for which they most sought to improve evaluation
practices.28
Moreover, in interviews, evaluators clearly evinced that determining whether someone
with psychic handicap qualified for disability benefits required that they carry out
an expert assessment separate from that made by psychiatrists. As the director of the
Disability Office insisted, “You can have very different diagnoses [of mental illness]
that result in the same consequences…. The pathology is an element that is going to
clarify things … but it’s not the foundation of what we do.” His discourse paralleled
that of the referent for psychic handicap in the Ministry, who confirmed, “For disability,
the diagnosis is not super important, what is important is the consequence. We don’t
need to know if someone is bipolar or schizophrenic. It was a conscious choice not to
make that central to evaluation, because we’re in handicap here, not in health.” On the
ground, the doctor on the evaluation team who reviewed the medical certificates would
frequently not mention diagnoses in discussions of individual cases, focusing only on
the observable consequences of those diagnoses in functioning. From the administration
to the ground level, disability professionals thus embraced the frequent critique of
psychiatric diagnoses as unreliable, dependent on the subjective and inscrutable judgment
of the psychiatrist (Craciun 2016; Strand 2011; Whooley 2016).
Indeed, looking at diagnosis alone would not help evaluators respond to pressures to
objectify and justify their determinations. One member of the team needed to fill out the
40-page “guide of evaluation of needs for disabled persons” that precisely explained
someone’s limitations in daily life, independent living, or employment. These determinations
could potentially be scrutinized by a departmental commission, composed of
representatives of the government and associations for people with disabilities, which
heard appeals and reviewed a random sample of dossiers in detail. That commission, in
turn, faced oversight from the central state, which was seeking to standardize
28
Haut Conseil de la santé publique, 2014. Enquête quantitative sur les modes d’évaluation et de traitement
des demandes de compensation du handicap par les MDPH (Paris, France).
Theory and Society (2019) 48:753–784 771
evaluations and eliminate “deviations in the interpretation of [regulatory] texts,” as one
ministry document put it. This goal was reaffirmed in 2017 in a new project to ensure
“the objectification … [of[ the needs of the person … and responses to them.”29
French
disability evaluators were thus far from the old “politics of the administrative window”
(Alexis 2008; Dubois 2003) in which low-level functionaries could exercise professional
judgment largely free from outside surveillance.
Despite the stated commitment to using psychic handicap as a boundary object to
establish disability as separate from mental illness, in practice it was a bureaucratically
split personality because the tools that were supposed to be used to produce “objective”
evaluations were a poor match for mental disorders. For example, one benefit offered
personalized aid for people with “severe” or “absolute” difficulties in one of sixteen
different activities. The criteria reflected the vision of disabilities as something that
could be precisely measured and documented: a simple set of procedures could show
that a person who was paraplegic had an “absolute” barrier to feeding themselves or
getting out of bed. The Law of 2005 included activities like “orient oneself in time and
space,” “manage one’s security,” or “master one’s behavior in relationship with others”
explicitly in order to cover people with psychic handicaps. But it was very difficult to
show that limitations in these domains were “severe” or “absolute.”30
The lead
evaluator for this benefit explained, “We are trying to see what is possible or impossible
for someone … for psychic handicap, that’s complicated.” In my coding of reasons for
rejection from various benefits, I found that people with mental disorders were far less
likely to be found to be concretely “insufficiently handicapped” or “too handicapped”
to receive benefits (the latter most often referring to help finding employment), but
more likely to have an application deferred because of a lack of clear evidence either
way (see Table 2).
Like most tools under the 2005 law, the medical certificate attached to each
application was supposed to cover all handicaps, regardless of their origin. It included
a series of checkboxes to be filled out by doctors about whether applicants were
autonomous in various daily activities. To Disability Office evaluators’ frustration,
however, medical certificates describing people with grave, treatment-resistant schizophrenia
nonetheless frequently indicated that the person was “autonomous” in activities
like “clothe oneself” or “travel outside the home,” thus disqualifying them from many
benefits. The problem, one representative from UNAFAM pointed out, was that
“physically they can get up, get dressed, cook dinner—but they don’t.” Disability has
long served as an exemption from certain social expectations—like employment—
based on what someone cannot do (Stone 1984). A psychic handicap, defined by
limitations to the activities someone will do, was inapplicable within this framework.
A boundary object must “satisfy the information requirements” (Star and Griesemer
1989, p. 393) of each world. But the evidence psychiatrists supplied to show psychic
handicap often spoke to a troubled subjectivity but not objective incapacities. In one
debate over whether to give a Handicap Allocation to someone whose medical
29
Caisse nationale de solidarité pour l’autonomie, 2017, Une réforme tarifaire pour faciliter les parcours des
personnes handicapées (Paris, France), p. 2.
30
15.8% of these allowances were given to people with psychic handicaps, even though they constituted
closer to 25% of total applicants for disability benefits. Amara, Fadéla, Danièle Jourdain-Menninger, Myriam
Mesclon-Ravaud, and Gilles Lecoq, 2011, La prise en charge du handicap psychique (Paris, France),
Inspection générale des affaires sociales, pp. 30, 70.
Theory and Society (2019) 48:753–784772
certificate described him as having “behavioral problems,” one evaluator commented,
“Maybe this really isn’t a psychic handicap—it’s just someone who’s frustrated and
compulsive!” The example displays how, in these evaluation situations, the “personality”
of people who ostensibly had a psychic handicap was “bureaucratically split”: the
fact that the issue was “psychic” seemed to conflict with the notion that it was a
“handicap.”
These difficulties in conducting evaluations that satisfied bureaucratic requirements
were particularly visible in cases of “anxiety-depression syndrome.” Mood disorders
have been a central component of the rise of disability rolls in the United States
(Berkowitz and DeWitt 2013, p. 223) and now account for 39% of new awards for
disability income on the basis of mental disorders.31
But in Paris (no national data are
available for France as a whole), mood disorders were only 21% of an already smaller
portion of awards given for mental disorders.32
My observations suggest that this was
not based on a stigmatizing belief that applicants were malingerers or that giving
disability benefits would demotivate a search for employment (see Table 2).
Rather, rejections stemmed from the way that, in an attempt to comply with the
requirement that they document a clear justification for their decisions, evaluators chose
indicators that widened the gulf between “people with depression” and “people with
disabilities.” In particular, in the absence of concrete evidence of how anxiety or
depression created functional limitations in everyday life, evaluators used two shortcuts:
is the person treated by a psychiatrist in a public Psychiatric Clinic? Has the
person been hospitalized? Explained one evaluator:
We’re obligated to grab onto something: is there a treatment, consultations,
hospitalizations? Often, the people that show they don’t have treatments, that
are not hospitalized … by definition are not handicapped. If it’s hard to find
something in terms of a restriction [of activities], and we don’t have lots of
elements of treatment, of [medical] following, then it’s not serious.
Table 2 Reasons for rejection from at least one disability benefit
Mental disorder (n = 59) Other (n = 147)
Administrative ineligibility (working, migration status) 6.7% (4) 22.4% (33)
Benefits de-motivating 3.4% (2) 4.8% (7)
Incomplete evidence of handicap 23.7% (14) 19.0% (28)
Not in treatment 13.6% (8) 4.8% (7)
Not sufficiently handicapped 5.1% (3) 35.4% (52)
Still in treatment 39.0% (23) 4.1% (6)
Too handicapped 1.7% (1) 6.8% (10)
31
Calculation based on Social Security Administration, 2015, Annual Statistical Report on the Disability
Insurance Program (Washington, DC).
32
Unpublished statistics (see Note 9).
Theory and Society (2019) 48:753–784 773
In discussing one case, a person wrote in her own application for a Handicap Allocation
that, in addition to physical limitations, she also had “psychic fragility” and depression
as a result of an automobile accident. One evaluator exclaimed, “We don’t know what
that means!” Another cut in, “Well, is she being followed by a psychiatric clinic?”
When the doctor looking at the dossier replied, “No,” the head of the team declared,
“Okay, well, at a physical level she doesn’t meet our criterion, and she’s not being
followed for her ‘fragility.’ We can reconsider this if she starts being followed at a
[public sector] clinic.” The evaluator was not saying that she directly needed an
evaluation by a psychiatrist that she could “rubber stamp” as a determination that the
person’s depression constituted a psychic handicap. Rather, she was using a following
by public psychiatry as an indicator of disability in and of itself in her own expert
evaluation.
Ultimately, this heuristic outsourced expertise because it made evaluations contingent
on the prior decisions of public sector psychiatrists over whether to treat someone
in a psychiatric clinic, prescribe them medication, or hospitalize them. Psychic handicap
did not serve as a boundary object which bridged separate classifications “without
imposing … categories from one community” onto another (Bowker and Star 1999, p.
297). Instead, psychiatric diagnoses—which French psychiatrists did not have to
objectify using formal tools or measurements, and which many in the disability system
critiqued in the abstract as unreliable and subjective—in effect became disability
evaluations. They thus excluded some people whose functional limitations were
difficult to evaluate but which might formally qualify them for disability benefits.
In a sign of the state and civil society’s commitment to the spirit behind the
recognition of psychic handicap, various state agencies discussed bureaucratic fixes
for these evaluation procedures.33
One option would be for evaluations to rely officially
on psychiatric expertise, which would defeat advocates’ goal of moving beyond a
purely-medical understanding of the challenges of people with severe mental disorders.
Another would be to create separate tools and procedures for this population, which
would make the challenge of commensurating them with other people with disabilities,
discussed in the next section, more severe. In either case, the attempt to use psychic
handicap in evaluations showed it to be a bureaucratically split personality. Illnesses
that created difficult-to-document troubles of volition could not be easily translated into
incapacities that the Disability Office could objectively establish.
Institutionalization: stable compensation versus unstable troubles
In some cases, the team at the Disability Office could evaluate the medical certificate
and life project of a mentally ill applicant, recognize the existence of a psychic
handicap, and directly grant a disability benefit (like the Handicap Allocation). The
full aspiration of advocates, however, was that psychic handicap would enable people
with severe mental disorders to enter an existing infrastructure of professional
retraining centers, protected workshops, and supported housing for people with disabilities
(see Fig. 2). Institutionalizing psychic handicap, then, required establishing
that people with psychic handicaps were commensurate with the disabled people these
33
A key example was “Potentiel Emploi,” a national experimentation in a detailed evaluation of working
capacity centered on people with psychic handicaps. The experiment was not renewed.
Theory and Society (2019) 48:753–784774
institutions already served. Here, a tension long embedded in the categories of disability
and mental illness—that the former was fixed and stable, the latter unpredictable and
variable—constituted another way that psychic handicap was an unworkable bureaucratically
split personality.
This contradiction was evident to the Team for Professional Insertion at the Disability
Office. They determined whether a person who had already been designated
“disabled” would be offered help in finding employment through France’s Centers
for Professional Retraining, which were created for people disabled by war or workplace
accidents (Romien 2005). These centers operated based on a model of “compensating”
for the limitations of a disability through a long trajectory of evaluation, preorientation,
re-training, development of a “project for professional insertion,” internships,
and only then job searching. These institutions’ requirement that people be
“stabilized”—“a strict precondition” according to one government report34
—was at
odds with the very definition of psychic handicap given by UNAFAM and other
advocates, which emphasized its “variability” and “the need for constant adjust-
ments.”35
As one psychologist who worked on professional insertion observed at a
training, “We should remember that we’re dealing with a population [mentally disordered]
that at any moment could fall apart and we have to start again from zero.”
The delay with which the Disability Office treated applications was undoubtedly
frustrating for all applicants but particularly problematic for those with mental disorders.
One day, the team debated a 53-year-old woman whose two-year-old medical
certificate described a “chronic psychosis” and “heavy treatment.” The team’s doctor
commented, “If she hasn’t been in the hospital since then, maybe she’s okay….” But
the psychologist from the retraining center replied, “No. She could be totally delirious.
We’ve done this before and it hasn’t worked out. Can we just say she’s ‘still in care’
and give her a Handicap Allocation?” While evoked in only 5% of other rejected
demands, the team used “still in treatment” as a justification 40% of the time in refusing
a request from someone whose limitations stemmed primarily from a mental disorder
(Table 2). Confronted with this observation, one psychiatrist working in a service for
professional insertion admitted that “the notion of ‘handicap’ in psychiatry is not that
interesting if you say that it’s something stable and fixed. ‘Stable’ means nothing, in
terms of psychiatric pathologies … just call it a chronic illness.”
This same contradiction appeared in France’s extensive system of protected workshops.
These are organized around the model of a disabled person (classically, someone
with intellectual disabilities) who could produce predictably and reliably at some level
lower than that of a “normal” worker, with the difference made up for by state
subsidies. These facilities had an interest in taking people with psychic handicaps
because prenatal screening and abortion meant the population of people with conditions
like Down’s Syndrome was decreasing. Yet when the directors of these institutions
actually brought people designated as having a psychic handicap into their establishments,
they struggled to commensurate them with their institutionalized schema of
disability:
34
Le Houérou, Annie, 2014, Dynamiser l’emploi des personnes handicapées en milieu ordinaire (Paris,
France), Assemblée Nationale, p. 24.
35
UNAFAM, Le livre blanc, p. 12.
Theory and Society (2019) 48:753–784 775
These are people [with a psychic handicap] who are one day capable of taking
lots of initiative, lots of responsibilities, very technical things, and another day,
they can’t tie their shoes…. That’s the difficulty, not because they don’t know
how, not because they can’t do it, but because some days, they just don’t do it!
My interviews suggested these difficulties were more about practicality than stigma.
Because protected workshops largely cannot fire workers once they have been hired,
managers had to continue to pay mentally disordered persons even when they were on
long sick leave due not to their handicap, but their illness. They thus declared that
people with intellectual disabilities were preferable because the key metric against
which they measured disabled people was their “capacity to maintain their work over
time.” The result was that only 6% of places in protected workshops in the Parisian
region were dedicated to individuals with primarily mental disorders.36
In France, being psychically handicapped presents a disordered, contradictory split
personality with respect to work. On one hand, France has long been seen as a leader in
promoting the employment of disabled persons (Power et al. 2014). I observed that at
least some of the gatekeepers of employment for persons with disabilities—the Disability
Office, “ordinary” firms, and protected workshops—made genuine attempts to
accommodate people with mental disorders. But, once admitted, they determined that
they were a poor fit, and thus began turning around new candidates. By default, then,
they made the primary bureaucratic identity of this population mentally ill. And, in
contrast to its employment-focused disability policy, France’s generous paid sick leave
means that it has among the highest proportions of the population among rich countries
who are not working because of illness.37
Thus, when government statistics show that
very few people labeled as having a psychic handicap are employed, the reason is that,
as one workshop director explained: “I’ve realized that these people aren’t really
handicapped! They’re just sick!”
Difficulties of commensuration, rather than overt classification struggles, created
problems for accessing long-term disability housing as well. Across the
health and disability systems, nearly everyone I interviewed endorsed the goal
of reducing the number of chronically ill patients languishing in hospitals. Ideally,
these individuals would go to disability housing establishments that were charged
with promoting the active social participation and rehabilitation of their tenants.
But, as the person responsible for handling housing placements at the Disability
Office explained:
When you’re at the hospital, everything is organized for you. They tell you
when to get up, when you sleep, and the rest of the time, you do nothing. In
supported housing, you have a collective life, activities all day. If you’re not
used to it, it’s exhausting. And this constant solicitation to do things that
36
Agence régionale de santé, 2011, Projet Régional de Santé: Schéma d’organisation médico-social, Île-deFrance,
p. 45.
37
OECD Health Statistics, 2012, retrieved September 13, 2017 (http://www.oecd.org/els/health-
systems/health-data.htm).
Theory and Society (2019) 48:753–784776
you don't have the capacity to do, it creates frustration and behavioral
problems … and you go back to the hospital.
A social worker from the Psychiatric Clinic recounted the case of a patient who had
been involuntarily hospitalized for 3 years, and who would collect cigarette butts and
eat them. Working with the Disability Office, they found a place for him in a nursing
home but they “sent him back to us because they said that he wasn’t willing to
participate in group activities, he just wanted to eat cigarette butts. And I was like,
‘yeah, that’s the point.’” This back-flow was significant, equivalent to about 22% of the
people leaving psychiatric hospitals for disability housing.38
These challenges were not a matter of simple gatekeeping by housing establishments.
The directors of the private associations that run them actually wanted to tap into
a growing population. But, in a manner consistent with the idea of psychic handicap as
a boundary object, they usually demanded that the public psychiatric teams sign a
convention agreeing to re-hospitalize those individuals when necessary. The problem
was that, when they did hospitalize patients whose symptoms flared, the homes were
required to hold that person’s bed, but were only paid one-third their usual rate by the
state. The solution, the head of placements in disability establishments at the Disability
Office explained, was that “now the establishments want a guarantee that the person is
‘stabilized’ … which means they [people with psychic handicap] are never going to
make it.” This refractory looping reaffirmed a meaning of disability that increasingly
excluded psychic handicaps. In 2014 statistics for the Parisian region, 45% of the
people who had been approved for residential services by the Disability Office but had
not found a structure willing to take them had a psychic handicap.39
At the time of research, efforts were being made to facilitate the entrance of these
populations into disability housing. Doing so, however, required a profound break with
a model that provided the same daily payment for all categories of disabled persons. It
thus entailed the slow process of building specialized institutions only for the psychically
handicapped—which, increasingly, were simply converted pavilions of psychiatric
hospitals. There, one ministry official admitted, “the color of the blouses has
changed, but the practices are the same.” As was the case when the Disability Office
outsourced expertise, psychic handicap was being treated more as a synonym of severe
mental illness than a true boundary object. Other innovative programs—like worksite
“Job Coaching” or rapid-entry “Housing First”— relied on re-directing funding and
expertise back into the health system, which was not bound by the expectation of
“stability” around which the disability system was organized.
These findings speak to the role of the particular configuration of the French welfare
state in shaping what happened when the contradictions of a bureaucratically split
personality emerged. Unlike the targeted disability system, in which people needed an
affirmative evaluation from the Disability Office and to be accepted by an institution,
France’s universal health system had no prerogative to refuse people (Bauduret and
Jaeger 2005). Through classification by default, psychiatric hospitals became the home
for mentally disordered people with irregular immigration statuses, criminal justice
38
Direction de la recherche, des études, de l’évaluation et des statistiques, 2013, Les établissements et services
pour adultes handicapés: Résultats de l’enquête ES 2010 (Paris, France), pp. 269, 273.
39
MDPH 75, 2014, Rapport d’activité (Paris, France), p. 61.
Theory and Society (2019) 48:753–784 777
involvement, or, in some cases, a history of violence towards the staff at establishments
for people with disabilities. Outside the hospital, the lower level Psychiatric Clinic staff
remained the de-facto sources of advice in looking for employment, facilitators of
social activity, and aids for daily life (“we’re nurses specialized in assembling IKEA
furniture,” one joked). They encouraged people with severe mental disorders to eke out
a living using public housing or the minimum welfare payment available to all lowincome
people. These were clearly not the dynamic services envisioned for people in
the disability system, but they required neither stability, nor self-identification, nor a
clear documentation of functional limitations. They thus were the fallback source of
solidarity for those deemed incommensurable with other people with disabilities.
Conclusion
Why did the introduction of “psychic handicap” fail to effect the intended transformation
of people with severe mental disorders from passive objects of medicine to
individuals capable of managing an autonomous life project, aided simultaneously by
both the medical and disability systems? For some advocates, the law of 2005 had not
carried sufficient funding or political will for implementing the gamut of services that
official recognition of psychic handicap had promised. They blamed professional
resistance and a mutual fear of symbolic contamination from psychiatrists and disability
institutions, with each group seeing the people served by the other as undesirable.
These explanations correspond to theoretical ones in the literature on classification
struggles (Barman 2013; Goldberg 2005) and cultural constraints to the introduction of
new classifications (Norton 2014; Steensland 2006) and are undoubtedly partly true.
I also, however, presented counterevidence to these explanations. Policymakers and
administrators endorsed the change and psychiatric professionals acquiesced to it. My
ethnographic observations show how, at the micro-level, social workers in the Psychiatric
Clinic and evaluators in the Disability Office labored collectively to try to adapt
the tools, procedures, and regulations given to them to a new population (see Kellogg
2011). The recognition of psychic handicap as a basis of disability, on top of an existing
bureaucratic identity of mentally ill, entailed clear material benefits that ought to have
driven requests for classification (Bailey 2008; Liu et al. 2010). This counter-factual of
a transformative new classification is even more plausible given the success of French
parents in the 1990s in establishing autism (which was ultimately called “cognitive
handicap”) as the basis for a bureaucratic recognition of their children as disabled
(Chamak 2010).
My analysis starts with the recognition that, in analyzing a state classification like
psychic handicap, “although it is possible to pull out a single classification … in reality
none of them stands alone” (Bowker and Star 1999, p. 38). In this case, the introduction
of psychic handicap was supposed to serve as a boundary object that would facilitate
the simultaneous classification of people with severe mental disorders as mentally ill
and disabled. But both categories continued to bear the imprint of the different arms of
the state in which they originated. Each limb had contradictory behavioral expectations,
principles of expert evaluation, and typologies of persons they served, which, suddenly,
were supposed to apply to the same people. This prompted processes of outsourcing
expertise, classification by default, and refractory looping. Although these mechanisms
Theory and Society (2019) 48:753–784778
harken to classic street-level bureaucratic practices like “rubber-stamping,” “gatekeeping,”
or “burden shifting” (Lipsky 2010; Prottas 1979; Seim 2017), they specify how a
bureaucratic split personality led to a distinctive outcome based on the particular
structure of the French welfare state. As in the case of the clinical treatment of “split”
or “multiple personality disorder” (now referred to as “dissociative identity disorder”),
the ultimate solution was the “reintegration” of these contradictory bureaucratic personalities
into a single, unified human kind: mentally ill.
These findings are surprising in light of research on “boundary objects” that
claims that very different professionals can maintain radically different practices
and visions of the objects of their work so long as they are linked by some
common categories or tools (Bowker and Star 1999; Mol 2002; Star and
Griesemer 1989). Why was the mighty French state unable simply to reconfigure
psychic handicap to satisfy the needs of both the health and disability systems?
My results speak to the issue of scale: psychic handicap really was a logically
consistent boundary object in national policy and even, to an extent, for the heads
of the Psychiatric Clinic or Disability Office. But for the nurses and social workers
who bore the brunt of preparing applications or the psychologists and special
educators responsible for filling in the tables and checklists of the Disability
Office, that psychic handicap was a bureaucratically split personality was evident.
Even once these problems filtered upwards, they were difficult to resolve because
what was at stake was not just an administrative label but a personality. Interventions
by the health and disability systems were organized by distinctive visions of
the kinds of people they served. Trying to change those kinds—by, for example,
suggesting that the preferences of people with disabilities could be over-ridden if,
like people with psychic handicaps, they lacked “insight” into their needs and
limitations (Gong 2017)—would most likely have undermined the consensus on
which the 2005 reform rested and sparked a true classification struggle.
This negative case points to reasons why the classification of disability has been
much easier to apply to the mentally disordered in other contexts. In the United States,
the dominant system of psychiatric diagnoses—the Diagnostic and Statistics Manual—
incorporates an evaluation of functional limitations (Whooley 2016). Disability evaluations
are initially more purely medical (Powell 2010; Stone 1984) and diagnosis is
thus deliberately “objectified” as disability. People with psychiatric disabilities are
commensurate with other people with disabilities insofar as the category of “mental
disability” was already in use by government agencies by the 1970s; mentally disabled
persons formed an early part of the model around which the disability system is based
(Bagnall and Eyal 2016). Moreover, as the United States has cut back even residual
welfare programs, mentally disordered persons have no option other than asking for the
designation of disability, which gives access to a minimum income and Medicaid
insurance (Hansen et al. 2014). Ironically, while the rising numbers of people receiving
disability benefits on the basis of mental illness has provoked continuous political
consternation in the United States (Berkowitz and DeWitt 2013), the classifications
mentally ill and disabled have proven practically compatible, even without a bridge like
psychic handicap. As this example suggests, we need greater attention to cross-national
differences not just in the prevalence of classifications like disability but the meaning
and nature of state categories that make them easier or more difficult to apply (see
Carson 2007; Fourcade 2009).
Theory and Society (2019) 48:753–784 779
This analysis thus focuses on the meta-characteristics of categories themselves and
the relations among them as presenting a potential barrier to the reclassification of
populations. Classifications may require validation from a single bureaucrat or demand
information and collaboration of a network of experts. Institutions can vary in the
extent to which they broadly or narrowly construct the populations they serve and thus
in the degrees of commensuration they expect among them. Crucially, categories can be
to a greater or lesser degree ascribed. States can unilaterally incarcerate those they label
criminals and intern those they label mentally ill, but new welfare programs that seek to
“empower” or “make autonomous” the poor require a greater degree of engagement,
even if that engagement loops back to modify the meaning of the category. Finally, reclassification
may take the form of a wholesale replacement of one classification by
another, like relabeling people with autism as disabled and not mentally ill (Chamak
2010; Eyal 2013). Alternatively, it may consist of layering classifications on top of one
another, as was the case with psychic handicap. The latter may be more likely to create
bureaucratically split personalities; the former overt classification struggles.
This perspective has important implications for a growing literature on how states
establish key lines of division and social identities in their subject populations
(Bourdieu 2015; Loveman 2014; Starr 1992). Psychic handicap could have been a
case of the seeming buildup of classifications over time through projects that apply
previously mutually-exclusive categories to the same people. Research points, for
example, to the simultaneous management of the poor through both medicalization
and criminalization, processes once thought to be opposing principles for state intervention
(Bosk 2013; Lara-Millán 2014; Seim 2017). These analyses run parallel to
work in the sociology of mental health that identifies an endlessly expanding accumulation
of psychiatric diagnoses (Conrad 2007; Horwitz 2001; Rose 2006). Yet even if
the state and medicine appear to construct an ever-more-elaborate “iron cage” of
classification (Bowker and Star 1999, p. 320; see, also, Foucault 1991), this research
helps identify the circumstances that limit attempts to layer and recombine classifications.
Work that develops the now frequent recognition that the state’s “many hands”
can act in ways that are “incoherent” (Morgan and Orloff 2017, pp. 3, 18), “contradictory”
(Goldberg 2005, p. 342), or “ambiguous” (Mora 2014, p. 188) should consider
how bureaucrats do not just struggle over categories with one another, but struggle with
the categories themselves.
What this case reveals is that it is precisely the state’s strength—its ability to marshal
the “material and symbolic resources to impose the … classificatory schemes … with
which bureaucrats, judges, teachers, and doctors must work” (Brubaker and Cooper
2000, p. 16; see, also, Loyal and Quilley 2017)—that made psychic handicap a
contradictory bureaucratically split personality. There are no inherent characteristics
of disability and mental illness that make them incompatible: rather, it was precisely the
long process of forging true, distinct human kinds out of these classifications in state
institutions that made merging one into the other so difficult. Orderly systems of
population classification, then, must emerge from resolving the disordered systems
that states themselves can create.
Acknowledgements The author gratefully acknowledges the assistance of Martin Eiermann, Gil Eyal, Neil
Fligstein, Marion Fourcade, Matty Lichtenstein, Michael Long, Mara Loveman, Marie Mourad, Isabel Perera,
Gisele Sapiro, Tonya Tartour, two anonymous reviewers, and the Theory and Society Editors, as well as the
Theory and Society (2019) 48:753–784780
Culture Organizations and Politics Workshop, Berkeley Medical Sociology Working Group, the BerkeleySciencesPo
Collaboration Conference, the Columbia Science Knowledge and Technology Working Group,
and the Center for European Sociology. A version of this article was presented at the 2017 ASA Annual
Meeting in the Regular Session on Health Policy. Research was made possible with the generous support of
the Chateaubriand and Georges Lurcy Fellowships and the Institute for International Studies and Center for
European Studies at Berkeley.
References
Abbott, A. (1988). The system of professions: An essay on the division of expert labor. Chicago: University of
Chicago Press.
Alexis, S. (2008). Accueillir ou reconduire. Enquête sur les guichets de l’immigration. 7Paris: Éditions
Raisons d’Agir.
Ayme, J. (2002). La loi de 1975 et les réactions syndicales. In J. P. Arveiller (Ed.), Pour une psychiatrie
sociale. 50 ans d’action de la Croix Marine (pp. 181–186). Erès: Ramonville-Sainte-Agne, Paris.
Bagnall, A., & Eyal, G. (2016). Foreverchildren? and autonomous citizens: Comparing the deinstitutionalizations
of psychiatric patients and developmentally disabled individuals in the United States. In B. Perry (Ed.), 50
years after deinstitutionalization: Mental illness in contemporary communities (pp. 27–61). Bingley: Emerald.
Bailey, S. R. (2008). Unmixing for race making in Brazil. American Journal of Sociology, 114(3), 577–614.
https://doi.org/10.1086/592859.
Barman, E. (2013). Classificatory struggles in the nonprofit sector: the formation of the national taxonomy of
exempt entities, 1969–1987. Social Science History, 37(1), 103–141. https://doi.org/10.1017
/S0145553200010580.
Barral, C. (2007). Disabled persons’ associations in France. Scandinavian Journal of Disability Research,
9(3–4), 214–236. https://doi.org/10.1080/15017410701680506.
Baudot, P.-Y., & Revillard, A. (2015). L’autonomie de l’équilibriste. Gouvernement et action publique, 4(4), 83–113.
Bauduret, J.-F., & Jaeger, M. (2005). Rénover l’action sociale et médico-sociale: histores d’une refondation
(2nd ed.). Paris: Dunod.
Berg, M., & Bowker, G. (1997). The multiple bodies of the medical record: toward a sociology of the artifact.
The Sociological Quarterly, 38(3), 513–537. https://doi.org/10.1111/j.1533-8525.1997.tb00490.x.
Berkowitz, E. D., & DeWitt, L. (2013). The other welfare: Supplemental security income and U.S. social
policy. Ithaca: Cornell University Press.
Bertrand, L., Caradec, V., & Eideliman, J.-S. (2014). Situating disability: the recognition of “disabled workers”
in France. ALTER-European Journal of Disability Research, 8(4), 269–281. https://doi.org/10.1016/j.
alter.2014.09.006.
Bosk, E. A. (2013). Between badness and sickness: reconsidering medicalization for high risk children and
youth. Children and Youth Services Review, 35(8), 1212–1218. https://doi.org/10.1016/j.
childyouth.2013.04.007.
Bourdieu, P. (2015). On the state. Cambridge, UK: Polity.
Bowker, G. C., & Star, S. L. (1999). Sorting things out: Classification and its consequences. Cambridge, MA:
MIT Press.
Bracci, E., & Llewellyn, S. (2012). Accounting and accountability in an Italian social care provider:
contrasting people-changing with people-processing approaches. Accounting, Auditing & Accountability
Journal, 25(5), 806–834.
Brubaker, R., & Cooper, F. (2000). Beyond “identity.”. Theory and Society, 29(1), 1–47.
Burawoy, M. (1998). The extended case method. Sociological Theory, 16(1), 4–33.
Bureau, M.-C., Rist, B., Lima, L., & Trombert, C. (2013). La traduction de la demande d’aide sociale: les cas
du handicap et de l’insertion des jeunes. Revue française d’administration publique, 145(1), 175–188.
Carson, J. (2007). The measure of merit: Talents, intelligence, and inequality in the French and American
republics, 1750-1940. Princeton: Princeton University Press.
Chamak, B. (2010). Autisme, handicap et mouvements sociaux. ALTER-European Journal of Disability
Research, 4(2), 103–115. https://doi.org/10.1016/j.alter.2010.02.001.
Chapireau, F. (2016). Le handicap psychique. In L’Encyclopédie médico-chirurgicale. http://www.emconsulte.com/article/1068052/handicap-psychique.
Accessed 19 June 2016.
Chauvière, M. (1980). Enfance inadaptée: l’héritage de Vichy. Paris: Les Editions Ouvrières.
Theory and Society (2019) 48:753–784 781
Conrad, P. (2007). The medicalization of society: On the transformation of human conditions into treatable
disorders. Baltimore: Johns Hopkins University Press.
Craciun, M. (2016). The cultural work of office charisma: maintaining professional power in psychotherapy.
Theory and Society, 45(4), 361–383. https://doi.org/10.1007/s11186-016-9273-z.
Davis, L., Fulginiti, A., Kriegel, L., & Brekke, J. S. (2012). Deinstitutionalization? Where have all the people
gone? Current Psychiatry Reports, 14(3), 259–269. https://doi.org/10.1007/s11920-012-0271-1.
Dubois, V. (2003). La vie au guichet: relation administrative et traitement de la misère (2nd ed.). Paris:
Economica.
Espeland, W. N., & Stevens, M. L. (1998). Commensuration as a social process. Annual Review of Sociology,
24(1), 313–343. https://doi.org/10.1146/annurev.soc.24.1.313.
Eyal, G. (2013). For a sociology of expertise: the social origins of the autism epidemic. American Journal of
Sociology, 118(4), 863–907. https://doi.org/10.1086/668448.
Foucault, M. (1991). Governmentality. In G. Burchell, C. Gordon, P. Miller (Eds.), The Foucault effect (pp.
87–104). Chicago: University of Chicago Press.
Fourcade, M. (2009). Economists and societies: Discipline and profession in the United States, Britain, and
France, 1890s to 1990s. Princeton: Princeton University Press.
Friese, C. (2010). Classification conundrums: categorizing chimeras and enacting species preservation. Theory
and Society, 39(2), 145–172. https://doi.org/10.1007/s11186-009-9103-7.
Goldberg, C. A. (2005). Contesting the status of relief workers during the new Deal. Social Science History,
29(3), 337–371.
Goldstein, J. (1987). Console and classify: The French psychiatric profession in the nineteenth century.
Cambridge: Cambridge University Press.
Gong, N. (2017). “That proves you mad, because you know it not”: Impaired insight and the dilemma of
governing psychiatric patients as legal subjects. Theory and Society, 46(3), 201–228. https://doi.
org/10.1007/s11186-017-9288-0.
Hacking, I. (1995). The looping effects of human kinds. In D. Sperber, D. Premack, & A. J. Premack (Eds.),
Causal cognition: A multidisciplinary approach (pp. 351–383). Oxford, UK: Oxford University Press.
Hacking, I. (1998). Rewriting the soul: Multiple personality and the sciences of memory. Princeton: Princeton
University Press.
Hansen, H., Bourgois, P., & Drucker, E. (2014). Pathologizing poverty: new forms of diagnosis, disability, and
structural stigma under welfare reform. Social Science & Medicine, 103, 76–83. https://doi.org/10.1016/j.
socscimed.2013.06.033.
Hasenfeld, Y., Rafferty, J. A., & Zald, M. N. (1987). The welfare state, citizenship, and bureaucratic
encounters. Annual Review of Sociology, 13(1), 387–415.
Henckes, N. (2011). La politique du handicap psychique. DREES/MiRe–Convention de Recherche n° 08-
2547. https://hal.archives-ouvertes.fr/file/index/docid/769756/filename/Henckes_2011_Rapport_MiRe.
pdf. Accessed 1 Oct 2015.
Hirschman, D., Berrey, E., & Rose-Greenland, F. (2016). Dequantifying diversity: affirmative action and
admissions at the University of Michigan. Theory and Society, 45(3), 265–301. https://doi.org/10.1007
/s11186-016-9270-2.
Horwitz, A. V. (2001). Creating mental illness. Chicago: University of Chicago Press.
Joyce, P., & Mukerji, C. (2017). The state of things: state history and theory reconfigured. Theory and Society,
46(1), 1–19. https://doi.org/10.1007/s11186-017-9282-6.
Kellogg, K. C. (2011). Challenging operations: Medical reform and resistance in surgery. Chicago:
University of Chicago Press.
Kellogg, K. C. (2014). Brokerage professions and implementing reform in an age of experts. American
Sociological Review, 79(5), 912–941. https://doi.org/10.1177/0003122414544734.
Lakoff, A. (2005). Diagnostic liquidity: mental illness and the global trade in DNA. Theory and Society, 34(1),
63–92. https://doi.org/10.1007/s11186-005-6233-4.
Lamont, M., & Molnár, V. (2002). The study of boundaries in the social sciences. Annual Review of Sociology,
28, 167–195.
Lanteri-Laura, G. (1972). La chronicité dans la psychiatrie moderne française. Note d’histoire théorique et
sociale. Annales. Histoire, Sciences Sociales, 27(3), 548–568. https://doi.org/10.3406/ahess.1972.422522.
Lara-Millán, A. (2014). Public emergency room overcrowding in the era of mass imprisonment. American
Sociological Review, 79(5), 866–887. https://doi.org/10.1177/0003122414549552.
Lipsky, M. (2010). Street-level bureaucracy: dilemmas of the individual in public service (30th anniversary.).
New York: Russell Sage Foundation.
Liu, K., King, M., & Bearman, P. S. (2010). Social influence and the autism epidemic. American Journal of
Sociology, 115(5), 1387–1434. https://doi.org/10.1086/651462.
Theory and Society (2019) 48:753–784782
Loveman, M. (2007). The U.S. Census and the contested rules of racial classification in early twentiethcentury
Puerto Rico. Caribbean Studies, 35(2), 79–113.
Loveman, M. (2014). National colors: Racial classification and the state in Latin America. Oxford, UK:
Oxford University Press.
Loyal, S., & Quilley, S. (2017). The particularity of the universal: critical reflections on Bourdieu’s theory of
symbolic power and the state. Theory and Society, 46(5), 429–462. https://doi.org/10.1007/s11186-017-
9298-y.
Mayrl, D., & Quinn, S. (2016). Defining the state from within: boundaries, schemas, and associational
policymaking. Sociological Theory, 34(1), 1–26. https://doi.org/10.1177/0735275116632557.
Mohr, J. W., & Duquenne, V. (1997). The duality of culture and practice: poverty relief in New York City,
1888-1917. Theory and Society, 26(2–3), 305–356. https://doi.org/10.1023/A:1006896022092.
Mol, A. (2002). The body multiple: Ontology in medical practice. Durham: Duke University Press.
Mora, G. C. (2014). Cross-field effects and ethnic classification. American Sociological Review, 79(2), 183–
210. https://doi.org/10.1177/0003122413509813.
Morgan, K. J., & Orloff, A. S. (Eds.). (2017). Introduction. In The many hands of the state: Theorizing
political authority and social control (pp. 1–32). Cambridge, UK: Cambridge University Press.
Navon, D., & Eyal, G. (2016). Looping genomes: diagnostic change and the genetic makeup of the autism
population. American Journal of Sociology, 121(5), 1416–1471. https://doi.org/10.1086/684201.
Norton, M. (2014). Classification and coercion: the destruction of piracy in the English maritime system.
American Journal of Sociology, 119(6), 1537–1575. https://doi.org/10.1086/676041.
Pinell, P. (2004). La normalisation de la psychiatrie française. Regards sociologiques, 29, 3–21.
Porter, T. M. (1996). Trust in numbers: The pursuit of objectivity in science and public life. Princeton, NJ:
Princeton University Press.
Powell, J. J. W. (2010). Change in disability classification: redrawing categorical boundaries in special
education in the United States and Germany, 1920–2005. Comparative Sociology, 9(2), 241–267.
Power, A., Lord, J. E., & DeFranco, A. S. (2014). Active citizenship and disability: Implementing the
personalisation of support. Cambridge, UK: Cambridge University Press.
Prottas, J. (1979). People processing: The street-level bureaucrat in public-service bureaucracies. Lexington:
Lexington Books.
Romien, P. (2005). À l’origine de la réinsertion professionnelle des personnes handicapées : la prise en charge
des invalides de guerre. Revue française des affaires sociales, 2, 229–247.
Rose, N. (2006). Disorders without borders? The expanding scope of psychiatric practice. BioSocieties, 1(04),
465–484. https://doi.org/10.1017/S1745855206004078.
Sandfort, J. R. (2000). Moving beyond discretion and outcomes: examining public management from the front
lines of the welfare system. Journal of Public Administration Research and Theory, 10(4), 729–756.
https://doi.org/10.1093/oxfordjournals.jpart.a024289.
Seim, J. (2017). The ambulance: toward a labor theory of poverty governance. American Sociological Review,
82(3), 451–475. https://doi.org/10.1177/0003122417702367.
Showalter, D. (2019). Misdiagnosing medicalization: penal psychopathy and psychiatric practice. Theory and
Society, 48(1), 67–94. https://doi.org/10.1007/s11186-018-09336-y.
Soss, J. (2000). Unwanted claims: the politics of participation in the U.S. welfare system. Ann Arbor:
University of Michigan Press.
Star, S. L., & Griesemer, J. R. (1989). Institutional ccology, translations and boundary objects: amateurs and
professionals in Berkeley’s Museum of Vertebrate Zoology, 1907-39. Social Studies of Science, 19(3),
387–420. https://doi.org/10.1177/030631289019003001.
Starr, P. (1992). Social categories and claims in the liberal state. Social Research, 59(2), 263–295.
Steensland, B. (2006). Cultural categories and the American welfare state: the case of guaranteed income
policy. American Journal of Sociology, 111(5), 1273–1326. https://doi.org/10.1086/ajs.2006.111.issue-5.
Stone, D. A. (1984). The disabled state. Philadelphia: Temple University Press.
Strand, M. (2011). Where do classifications come from? The DSM-III, the transformation of American
psychiatry, and the problem of origins in the sociology of knowledge. Theory and Society, 40(3), 273–
313. https://doi.org/10.1007/s11186-011-9142-8.
Waldschmidt, A. (2009). Disability policy of the European Union: the supranational level. ALTER-European
Journal of Disability Research, 3(1), 8–23. https://doi.org/10.1016/j.alter.2008.12.002.
Watkins-Hayes, C. (2009). The new welfare bureaucrats: Entanglements of race, class, and policy reform.
Chicago: University of Chicago Press.
Whooley, O. (2016). Measuring mental disorders: the failed commensuration project of DSM-5. Social
Science & Medicine, 166, 33–40. https://doi.org/10.1016/j.socscimed.2016.08.006.
Theory and Society (2019) 48:753–784 783
Winance, M., Ville, I., & Ravaud, J.-F. (2007). Disability policies in France: changes and tensions between the
category-based, universalist and personalized approaches. Scandinavian Journal of Disability Research,
9(3–4), 160–181. https://doi.org/10.1080/15017410701680795.
Zerubavel, E. (1996). Lumping and splitting: notes on social classification. Sociological Forum, 11(3), 421–
433.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
Alex V. Barnard is Assistant Professor of Sociology at New York University, with a PhD from the University
of California, Berkeley. His work examines how people with severe mental disorders are classified by and
distributed between institutions of care and control in France and the United States. He uses archives and
interviews to understand historical shifts in mental health policy alongside ethnographic observations to
explore how medical and bureaucratic decision-making is based on society-specific typologies of mentally
ill persons and boundaries around health, disability, and deviance. His previous work, on radical environmentalism
and the political economy of waste, has been published as “Making the City ‘Second Nature’” in the
American Journal of Sociology (2016) and as Freegans: Diving Into the Wealth of Food Waste in America
(University of Minnesota Press 2016).
Theory and Society (2019) 48:753–784784